For men, there are only three out of the eighty-six
categories
for which the number of "positive" and "negative" instances is equal, and only one (Category 2 3c) which shows a slight trend in the direction op- posite to the one expected.
Adorno-T-Authoritarian-Personality-Harper-Bros-1950
But, on the other hand, he keeps pulling back and is overcome with doubt and indecision about these things.
He emphasizes the basic importance of happiness and the emptiness of "success" without any personal satisfaction.
M44: His speech abounds in qualifying phrases and overintellectualization. He seems repeatedly unable to verbalize a generalization before he is overwhelmed by a rush of qualifications. Further, his thinking is rich in philosophizing, psychologiz- ing, and poetic statement. Moreover, these characteristics are not shallow but have much substance. ". . . Well, I don't think you should obey anyone or anything with- out question. I think it's man's unique function to question and when he ceases to question, he ceases to be man. (Have you ever had serious doubts about your re- ligious beliefs? ) Oh, not especially serious, I'd say, because I believe there should be changes. "
M48: "I'm what they always call an agnostic. Sounds sort of prosaic. . . . I'm skep- tical-though I believe Christ was a great man . . . persecuted. . . . "
M53: (How do you account for your growing away from the conventionalism of your background? ) "I don't know. It wasn't simply a change of locale. I think, probably, through reading. From 15 to 16 I did a lot of reading and became rather dissatisfied with it (i. e. , with conventional ideas with which brought up). (Were there any people who especially influenced you? ) No, must have been a hell of a lot of people. (Q) I don't know. I think through reading. I enjoy reading for read- ing's sake as well as a means of securing information. "
There is in the records of the low scorers a tendency to use a great deal of qualifying phrases and other devices characteristic of an approach that is judicious rather than prejudicial through dogma, convention or a fixed set. Impressionistic ratings based on synopsis as employed here are perhaps not
? THE AUTHORITARIAN PERSONALITY
the best means to nail down this difference. More concrete experimental studies on intolerance toward ambiguity now in progress (for an advance report see Frenkel-Brunswik, 37, and forthcoming papers dealing with the relationship between emotional ambivalence and perceptual ambiguity), and on rigidity (Rokeach, 98) point toward the relative prominence in eth- nically prejudiced as compared with unprejudiced children of a tendency to impose, in a rigid manner, certain preconceived sets upon ambiguous per- ceptual data or upon the solving of reasoning problems.
There seems to be a general tendency on the part of low scorers to expose themselves to broad experience-emotional, cognitive, perceptual-even at the risk of having to modify one's preconceived notions and of having to sustain conflicts. Thus all the evidence seems to point toward a greater over- all rigidity in the high scorers as compared with the low scorers. These results however, as many others, may well be valid only for the extreme groups. In the medium range, on the other hand, rigidity may be neither pervasive nor relatively absent throughout the personality as a whole, but may be differentially distributed over some areas of personality whereas other areas retain their flexibility.
Although low scorers seem in general to accept more realities even if com- plex and unstructured, there is a distinct sub-variety of the ethnically ex- tremely unprejudiced who cling to the liberal ideology in a rigid and stereo- typed manner. This often goes with a personality structure not dissimilar to that of the typical high scorer. (See also the characterization of the "rigid low" in Chapter XIX. )
Intolerance of emotional and cognitive ambiguity seems not only to be a characteristic of the personality of the ethnically prejudiced; it also ap- pears as part of the explicitly stated ego-ideal of exponents of the Nazi ideology in professional psychology. The most notable case is that of E. R. Jaensch with his programatic glorification of a personality type character- ized by fixed relationships between stimuli and perceptual responses, and with his rejection of the school of Gestalt psychology mainly on the ground of its stress on the concept of ambiguity. (For a summary on Jaensch see the paper by Boder, in Harriman, 47; see also 37. )
3. NEGA TIVE A TTITUDE TOW ARD SCIENCE. SUPERSTITION
The inability to "question" matters and the need for definite and dog- matic answers, as frequently found in high scorers, leads either to an easy acceptance of stereotyped, pseudoscientific answers, of which escape into ready-made hereditarian explanations is but one manifestation, or else to an explicitly antiscientific attitude. Explanations by accidental factors are like- wise included under this general heading. Its opposite is a scientific-natural- istic attitude, found predominantly in the low scorers. The entire Category 59 encompassing these alternatives differentiates significantly (at the 1 per
? PERSONALITY ORGANIZATION SEEN THROUGH INTERVIEWS 465
cent level) for both men and women. The difference in attitudes involved has been described previously, primarily in discussing the subjects' concepts of their "selves. " It will be remembered that it was the low scorers who showed a tendency toward explanations in terms of sociopsychological dynamics.
The antiscientific thinking of the typical high scorer is closely connected with his tendency toward superstition, as discussed in Chapter VII. The tendency toward superstition is illustrated by the following records of high- scoring women:
F31: "I am not superstitious. Mother is a little bit superstitious. She believes in old-world customs and palmistry. I'm not a fatalist, either; what I do will control my life, what I make of it. But I don't know about that-there are the boys on the battle- field, for instance. They say a bullet has your name on it. "
F36: Subject does not believe in formal religion; this is why she likes the Uni- tarians. She does not think the churches should have a narrow, strict creed and tell you exactly what you should believe. She has read a great deal of theosophy, Madame Blavatsky, etc. She believes in reincarnation and divine will: reincarnation in the sense that the soul goes from one body to another and that you will be sub- jected to those experiences that are necessary to learning, to enable the soul to reach a higher state.
F6o: (Why did you come to ? -) "Why, I don't know! It just happened. Don't you think some things just happen to us? " (Superstition? ) She was just "called" to do it. Once she was out walking in the early morning-the birds were singing-she raised her hands and her face to the sky, and they were wet. (What was it? ) She considered it a supernatural phenomenon.
Along the same line Lentz (67) reports that conservatives are more an- tagonistic to science, especially with regard to its future activities, and, conversely, are more superstitious. They feel much more favorable toward the conventional, the traditional, and the routine. They are less tolerant and sympathetic towards the underdog, less aesthetic and less imaginative.
The fact that high scorers on ethnocentrism are more often given to stereotyping, pre-judgments and ready generalizations, or else to overcon- creteness, should not blind us to the fact that there also are tendencies of this kind in the low scorers. The increasing complexity of the social realities and their partial unintelligibility to the individual sometimes may necessitate a falling back on stereotypes so that opaque events appear more compre- hensible (see Chapter XVII). Ethnic prejudice is but one of many pos- sible media for this tendency. It must be held in favor of the ethnically un- prejudiced, however, that they make a serious effort to counteract such stereotyping in one of the areas of paramount social significance.
4. ANTI-INTRACEPTIVENESS AND AUTISM
Likewise previously discussed but rated here directly and in its own right is the tendency, found primarily in the high scorer, toward what may be
? THE AUTHORIT ARIAN PERSONALITY
called anti-intraceptiveness as contrasted with a greater readiness toward intraception in the low scorer (Category 6o). The difference is significant at the I per cent level. The concept of intraception covers the tendency toward introspection, as well as a readiness toward gaining insights into psychological and social mechanisms. It is contrasted primarily with ex- ternalizations of various kinds as referred to above. The greater creativity, imagination, and ability for empathy of the low scorers just discussed is likewise related to their greater intraceptiveness.
It may suffice to cite here in addition to previous quotations only one record of a low-scoring woman and one of a low-scoring man in which the tendency toward reflection becomes apparent.
F7o: Always wanted to work with people. . . . (What does religion give you? ) "I suppose going to church takes a load off of me of thinking about things I should think about. I think it covers my social ideas, and it causes me to think about things I must think about for some reason. (What about? ) Moral values, the relative value between peace and liberty in wartime, for instance. Pacifism and its ins and outs, interpersonal relationship as moral values. I don't think of those values as right or wrong, except as it has social implications. "
M so: (Future? ) "l don't know. I just recently came of the opinion that it is not necessary for me to be a commercial artist. Important to do whatever I do well. Not important what I do any more (Q) Well, I was the prima donna type . . . built up myself into an idea that not suited for the work . . . now it doesn't matter any more . . . it began when I was here about a month . . . the child guidance center psychol- ogist gave a speech . . . he gave the idea that (deviation in behavior has a cause) . . . . It may go back to infancy. . . . I play around with it (leading to nothing) . . . until a friend also interested made a startling remark. . . . Then I thought a lot about it . . . put down reasons on both sides, on two sheets of paper and decided he was right. I don't mean I am in love with my mother, but I have a dependency com- plex . . . married a woman older than myself . . . and always depend on others . . . leave responsibility to others . . . it seems on looking back that I have always done that. . . . "
The -absence of proper orientation toward social and psychological dy- namics in the typical high scorer may be linked to a general tendency toward autistic thinking in goal behavior as contrasted with the somewhat more realistic attitude that might be expected to be present in the typical low scorer (Category 62). Though not significantly differentiating between the two extreme groups of interviewees, this category shows a distinct trend in the direction that we expected (2I positive as compared with I2 negative instances in men, and I9 positive as compared with I2 negative instances in women). The trend is particularly evident in the vocational choices and economic expectations of our subjects.
The somewhat adolescent and fantastic, glamour-seeking character of the vocational ambitions of high-scoring subjects is illustrated by the following records:
F66: "If you're good, you can get up to ambassador. I think there have been
? PERSONALITY ORGANIZATION SEEN THROUGH INTERVIEWS 467
some women ambassadors. Or maybe there were only women ministers. I made the choice about three years ago. I just heard some friends of mine talking about it, and it sounded interesting. (Why vocational choice? ) Well, salary had a little to do with it. I think if I could really put myself to it, I could do it. If I really wanted to and had nothing else to stand in the way. The times have much to do with my choice. I would like to know why they do what they do. "
F34: Subject wants to be a journalist. This is purely utilitarian. She likes journal- ism, but her real desire is to do creative writing. She has imagined herself as a great actress. "But my acting is purely amateur stuff. I was always active in school dramatics as well as high school journalism. The family used to laugh because I was always play-acting. I always said I'd earn my living at either acting or writing. . . . "
5. SUGGESTIBILITY
Submission to authority and lack of independence and of critical judg- ment tend to lead the high scorer toward being suggestible and gullible, as contrasted with the greater autonomy of the low scorer (Category 6I). Again, the difference is significant at the I per cent level for both men and women. The social implications of a tendency on the part of the gullible person to fall easy prey to unsound and destructive political propaganda are obvious. The high scorer's dependence, in his personal life, upon author- ity, conventionalized values, church dogma, public opinion, and prestige figures, and the low scorer's relative independence of, and occasional rebel- lion against, these authorities have been pointed out so often throughout this volume that no further comment seems necessary.
The present chapter will be summarized together with a general synopsis of the interview results at the conclusion of the chapter which follows.
? CHAPTER XIII
COMPREHENSIVE SCORES AND SUMMARY OF INTERVIEW RESULTS Else Frenkel-Brunswik
A. THE DISCRIMINATORY POWERS OF THE MAJOR AREAS STUDIED
1. VERIFICA TION OF ANTICIP A TED TREND BY CA TEGQRIES
In the preceding four chapters (IX to XII), the technique of interviewing was described, and specific results of interviews with eighty subjects scor- ing extremely high or extremely low on the overt anti-Semitism or Ethno- centrism scales were presented and discussed in terms of the approximately ninety rating categories of the Scoring Manual used to classify the records of the subjects.
As shown in the tables of these chapters, a sizable proportion of the cate- gories differentiated high scorers from low scorers to a statistically sig- nificant degree in the direction anticipated for either or both sexes. The vast majority of the remaining categories showed at least a trend in the expected direction.
For men, there are only three out of the eighty-six categories for which the number of "positive" and "negative" instances is equal, and only one (Category 2 3c) which shows a slight trend in the direction op- posite to the one expected. As can be seen from Tables 3(X), 1(XI), and 2 (XI), to which the four exceptions are limited, the absolute number of rat- ings involved is very small in each case; the items involved deal with certain aspects of attitude toward siblings, sex, and people. For women there are two out of the ninety categories that show equality and four that show reversal, all six exceptions being confined to Tables 1(X), 2 (X), and 3(X), and deal- ing with aspects of the relationships to parents and siblings. In no case does the reversal approach statistical significance; nor does it appear in the case of any category that yielded less than 50 per cent "Neutral" ratings and was thus included in the list of those for which intensive evaluation was under- taken (72 for men and 65 for women, see end of Chapter IX).
468
? Based on categories presented in tables
20 men and 25 women
20 men and 15 women "low scorers"
Sums of Means '
en
1 ( X ) , 2 ( X ) , 3(X)
P a r e n t s ? a n d Childhood
Men 8. . . . . . 613 women 7. 47
L 37 3. 26
3. 11 1. 68
8. 42 5. 11
17. 10 12. 58
4. 47 4. 94
1(Xl)
Attitude toward Sex
Men 10. 57 Women 10. 43
0 "':: ,. . . .
2(Xl)
Attitude toward
M 3. 89 i1:l
3(Xl)
Attitude toward Men Present Self Women
13. 00 11. 33
2. 44 3. 56
2. 44 1. 11
13. 11 9. 67
26. 11 21. 00
4. 88 4. 67
~
4(XI)
Attitude toward Men Childhood Self Women
8. 50 8. 00
3. 75 5. 75
2. 25 12. 75 1. 00 8. 75
21. 25 16. 75
6. 00 6. 75
q r --l en
1(XII)
Dynamic Character Men Structure Women
11. 65 11. 75
2. 53 3. 59
3. 10 11. 35 1. 95 9. 12
23. 00 20. 87
5. 63 5. 54
2(XII)
Cognitive Personality Men Organization Women
14. 67 14. 67
3. 00 3. 50
3. 00 14. 00 1. 83 11. 00
28. 67 25. 67
6. 00 5. 33
All tables
All categories Men Women
10. 91 10. 38
2. 24 3. 34
2. 81 10. 59 1. 69 8. 10
21. 50 18. 48
5. 05
Dealing with
Sex "high scorers"
"! :! ositive?
? negative?
~
Men People Women
10. 67 10. 67
1. 67 2. 78
2. 22 1. 44
11. 67 8. 00
22. 33 18. 67
4. 22 <. . . . . . M
TABLE 1 (XIII}
COMPOSITE RATINGS (MEANS} FOR MAJOR AREAS OF STUDY FOR "HIGH" AND "LOW" SCORING GROUPS OF INTERVIEWEES
Means o f High(H) and Low(L) ratings received by
! ! .
L
H
. k
a;: > ~
2. 86 1. 86
2. 57 2. 29
5. 86 9. 00
16. 43 19. 43
5. 43
4. 14 z
+ 5. 03 0\
--l
i1:l
M
en
\0
? THE AUTHORITARIAN PERSONALITY
Thus there are no "negative validities" in the evaluation of the interviews. This result is not surprising in view of the fact that the categories of the Scoring Manual were derived from previous empirical evidence, including a preliminary inspection of the interview material; even so, the fact that the subsequent "blind" rating procedure supported most of the hypotheses with which the evaluation began is strong evidence of their validity.
2. COMPOSITE RATINGS FOR SEVEN MAJOR AREAS
Considering the almost universally confirmatory trend of the results just discussed, the computation of average ratings for larger groups of categories seems justified. The results of this procedure are presented in Table 1 (XIII) in the same manner as in the preceding tables except for the fact that indica- tion of statistical significance is here omitted.
As can be seen by an inspection of the last two columns of the table, dif- ferentiation between high scorers and low scorers is consistently somewhat less clear-cut for women than for men. This may be due either to the un- equal size of the two samples of women interviewees, or to the fact, re- peatedly referred to above, that women were in the main interviewed at an earlier stage of the investigation than were men. It may also be that, in our culture at least, patterns of behavior and ideology are more coherent and consistent in men than they are in women.
Among all the areas studied, that of attitudes toward present self yields the most clear-cut differentiation between high scorers and low scorers. The ratio between "positive" (confirmatory) and "negative" instances is among the most favorable. At the same time, the absolute values of the positive in- stances are among the highest. This indicates, furthermore, that the number of "Neutral" ratings is relatively low for this area.
The categories pertaining to dynamic and, especially, cognitive personality organization likewise show excellent differentiation, a very high proportion of the differences being statistically significant.
Contrariwise, over-all results are least sharply defined in the area of attitude toward parents and siblings, for reasons outlined in Chapter X. This is in line with what was stated above concerning absence of reversal of the expected trend in the case of a few scattered categories pertaining to these topics. Even for this area, however, the ratio of positive to negative instances is, for the men, better than three to one, and thus the differentiation is quite satisfactory.
An explanation of the differences in discriininatory power among the various major areas covered by the ratings of interviews may be in the fol- lowing. An adequate survey of the early family situation requires a great deal more specific information than could be obtained in our necessarily circumscribed interviews, and this made for the large proportion of "Neutral"
? SUMMARY OF INTERVIEW RESULTS
ratings. Such aspects as cognitive organization and attitude toward self, on the other hand, may be inferred from expressions in a greater variety of fields, and thus the likelihood of insufficient information is reduced. Over and above this, the very nature of such formal variables permits the subject himself to choose from among a variety of "alternate" manifestations, many of them encompassed in a single rating category. This principle of shifting expres- sion of identical motivational tendencies is apt to give an advantage to mo- tivational categories regardless of how complete the specific information on any particular aspect of life may be. It is for the same reason that the value of generalized, synoptic ratings, as contrasted with specific or situationally limited quantification, was so much stressed at the beginning of the discus- sion of the problems connected with the evaluation of interviews (Chapter IX; see also 3I, 32, 34, 36).
B. V ALIDITY OF OVER-ALL SCORES AND RA TINGS OF THE INTERVIEWS
1. INDIVIDUAL COMPOSITE SCORE BASED ON ALL AREAS OF RA TING
Composite ratings, covering the entire range of the areas of rating, were also obtained for each individual interviewee. In a sense, these composite rat- ings define what may be called the "score" of the subject on the interview. (It was with a view to this final score that the manual was designated as the Interview Scoring Manual in spite of the fact that for each of the constituent categories ratings rather than scores were obtained. ) Since, as was pointed out in the preceding section, there were no reversals of the expected dis- criminatory trend on any of the categories yielding less than 50 per cent "Neutral" ratings, all of these categories were included in the final score. These scores are shown, for all the So subjects interviewed, in Tables I (IX) and 2 (IX). The score itself is given in two parts, defined by the sums of the "High" and of the "Low" ratings received by the individual. The difference between the sum of these two parts and the fixed total (72 for men and 65 for women), not shown in the tables, indicates the number of "Neutral" ratings received by the individual on the categories in question.
The column referring to composite standing shows the letter H or L, de- pending on the preponderance of the High or the Low score in the preced- ing pair of columns. 1
Means of the quantitative data are shown in the bottom rows of the tables. The ranges of the number of "high" and "low" interview scores for each of the four groups can be read directly from the two tables.
1 Concerning the use of capital and lower-case symbols and other technical data perti- nent to this section see Chapter IX, Section F, 3.
? 472 THE AUTHORITARIAN PERSONALITY
2. OVER-ALL INTUITIVE RA TING
AND ITS AGREEMENT WITH THE COMPOSITE SCORE
The last column in Tables r (IX) and 2 (IX) adds what was briefly men- tioned in Chapter IX as the intuitive rating of the interviewee. This is based on an informal synopsis and estimate after the blind interview-rating pro- cedure had been completed, and it was made without the rater's making explicit to himself the quantitative results in terms of specific ratings on the single categories.
Composite scores and intuitive ratings agree very highly with one another. There is only one subject, F72, among the So interviewees for whom there is a discrepancy between the two values. In her case the composite score
(H) is correct; the over-all intuitive rating (L) is incorrect.
3. AGREEMENT WITH THE QUESTIONNAIRE RESULTS
It will be noted that while there is consistency as to "low" vs. "high" standing in the A-S and E columns of Tables I (IX) and 2 (IX)-this standing having been the very criterion of selection of the interviewees-the subse- quent columns show a good deal of inconsistency. This inconsistency reaches major proportions in such cases as F2z or F39, both overt low scorers who nonetheless score among the highest on the interview, with respect to both the composite score and the synoptic over-all ratings. Similar instances of opposition, though not all as striking ones, can be found among the high scorers. Various other kinds of inconsistency of trend may be noted upon inspection of the tables.
The general agreement between various columns in Tables I (IX) and 2 (IX) containing letter indices may be expressed conveniently in terms of percentage. "Percentage agreement" is then defined as the percentage of equal-letter combinations (HH or LL) in any pair of columns considered. The figures on percentage agreement could easily be transformed into tetra- choric correlation coefficients.
The defining criterion of selection, extremely high vs. extremely low standing on the overt anti-Semitism or Ethnocentrism scale, shows a per- centage agreement of about 85 with both the over-all intuitive ratings and the composite standing on the interview. (This figure is an average of an agreement of about 95 per cent achieved by the rater whose material hap- pened to include the most complete interviews, and of an agreement of 75 per cent achieved by the other of the two raters whose data were more frag- mentary. )
Since composite ratings agree with intuitive ratings almost completely (see above), the figures for the agreement of overt ethnocentrism with the intuitive ratings are practically identical with those mentioned above for
composite ratings.
? SUMMARY OF INTERVIEW RESULTS
473
Intuitive as well as composite interview ratings show slightly less agree- ment with standing on the F and the PEC scales, but even here the figures are between 75 and So per cent.
C. SUMMARY OF THE PERSONALITY P A TTERNS DERIVED FROM THE INTERVIEWS
1. INTRODUCTION
The results of the "blind" ratings of the interviews, as discussed in the present and in the preceding four chapters, have shown that individuals ex- tremely high on overt ethnic prejudice tend to differ with respect to a great variety of personality traits from those extremely low on prejudice. Some of the personality traits discussed were also measured by other techniques, especially the questionnaire. The results obtained by the various methods independently are very similar. However, rating by categories describes and substantiates in a more systematic, organized, and controlled way the im- pressions formed about the personality differences between high scorers and low scorers in the course of intensive study of individual cases.
It has to be emphasized, of course, that these differences are based on an analysis of group trends within statistical samples and do not imply that every individual will exhibit most or even a large proportion of the features belonging to either the "high" or the "low" syndrome, as the case may be. As can be seen from Tables r (IX) and 2 (IX), certain individuals seem to possess a relatively large number of either "high" or "low" features while others seem to have features of both patterns, with a relatively slight prev- alence of one or the other. It should thus be kept in mind that the summary which follows deals with composite pictures of these patterns, abstracted from the study of groups, rather than with individual cases. Were we to lay greater stress on concrete personalities, the most frequent syndromes or combinations of trends within single individuals would have to be determined as an intermediate step, leading to the definition of subtypes within the prejudiced and the unprejudiced patterns. Some remarks pointing in this direction will be made in the following summary. The typology, as such, of the ethnically prejudiced, will be taken up more systematically in Chap- ter XIX.
The results are furthermore limited to trends found in individuals with extreme standing on the prejudice scale. How far the relations would hold for those with middle scores on prejudice has to be left open, since such in- dividuals were not included in the present intensive investigation by means of interviews.
Finally, it remains for future investigation to ascertain how far the inter- relationships found are interculturally valid or whether they are restricted
? 474
THE AUTHORITARIAN PERSONALITY
to certain specific cultures or subcultures, such as the one from which our sample of subjects has been drawn; namely, an urban and suburban popula- tion on the West Coast of the United States.
In an attempt to summarize the social and psychological factors which, within our limited framework, have been found to be related to prejudice, we will proceed in the following manner: First, the aspects and "themes" of the personality structure of the high scorer as differentiated from the low scorer will be described in a synoptic fashion.
M44: His speech abounds in qualifying phrases and overintellectualization. He seems repeatedly unable to verbalize a generalization before he is overwhelmed by a rush of qualifications. Further, his thinking is rich in philosophizing, psychologiz- ing, and poetic statement. Moreover, these characteristics are not shallow but have much substance. ". . . Well, I don't think you should obey anyone or anything with- out question. I think it's man's unique function to question and when he ceases to question, he ceases to be man. (Have you ever had serious doubts about your re- ligious beliefs? ) Oh, not especially serious, I'd say, because I believe there should be changes. "
M48: "I'm what they always call an agnostic. Sounds sort of prosaic. . . . I'm skep- tical-though I believe Christ was a great man . . . persecuted. . . . "
M53: (How do you account for your growing away from the conventionalism of your background? ) "I don't know. It wasn't simply a change of locale. I think, probably, through reading. From 15 to 16 I did a lot of reading and became rather dissatisfied with it (i. e. , with conventional ideas with which brought up). (Were there any people who especially influenced you? ) No, must have been a hell of a lot of people. (Q) I don't know. I think through reading. I enjoy reading for read- ing's sake as well as a means of securing information. "
There is in the records of the low scorers a tendency to use a great deal of qualifying phrases and other devices characteristic of an approach that is judicious rather than prejudicial through dogma, convention or a fixed set. Impressionistic ratings based on synopsis as employed here are perhaps not
? THE AUTHORITARIAN PERSONALITY
the best means to nail down this difference. More concrete experimental studies on intolerance toward ambiguity now in progress (for an advance report see Frenkel-Brunswik, 37, and forthcoming papers dealing with the relationship between emotional ambivalence and perceptual ambiguity), and on rigidity (Rokeach, 98) point toward the relative prominence in eth- nically prejudiced as compared with unprejudiced children of a tendency to impose, in a rigid manner, certain preconceived sets upon ambiguous per- ceptual data or upon the solving of reasoning problems.
There seems to be a general tendency on the part of low scorers to expose themselves to broad experience-emotional, cognitive, perceptual-even at the risk of having to modify one's preconceived notions and of having to sustain conflicts. Thus all the evidence seems to point toward a greater over- all rigidity in the high scorers as compared with the low scorers. These results however, as many others, may well be valid only for the extreme groups. In the medium range, on the other hand, rigidity may be neither pervasive nor relatively absent throughout the personality as a whole, but may be differentially distributed over some areas of personality whereas other areas retain their flexibility.
Although low scorers seem in general to accept more realities even if com- plex and unstructured, there is a distinct sub-variety of the ethnically ex- tremely unprejudiced who cling to the liberal ideology in a rigid and stereo- typed manner. This often goes with a personality structure not dissimilar to that of the typical high scorer. (See also the characterization of the "rigid low" in Chapter XIX. )
Intolerance of emotional and cognitive ambiguity seems not only to be a characteristic of the personality of the ethnically prejudiced; it also ap- pears as part of the explicitly stated ego-ideal of exponents of the Nazi ideology in professional psychology. The most notable case is that of E. R. Jaensch with his programatic glorification of a personality type character- ized by fixed relationships between stimuli and perceptual responses, and with his rejection of the school of Gestalt psychology mainly on the ground of its stress on the concept of ambiguity. (For a summary on Jaensch see the paper by Boder, in Harriman, 47; see also 37. )
3. NEGA TIVE A TTITUDE TOW ARD SCIENCE. SUPERSTITION
The inability to "question" matters and the need for definite and dog- matic answers, as frequently found in high scorers, leads either to an easy acceptance of stereotyped, pseudoscientific answers, of which escape into ready-made hereditarian explanations is but one manifestation, or else to an explicitly antiscientific attitude. Explanations by accidental factors are like- wise included under this general heading. Its opposite is a scientific-natural- istic attitude, found predominantly in the low scorers. The entire Category 59 encompassing these alternatives differentiates significantly (at the 1 per
? PERSONALITY ORGANIZATION SEEN THROUGH INTERVIEWS 465
cent level) for both men and women. The difference in attitudes involved has been described previously, primarily in discussing the subjects' concepts of their "selves. " It will be remembered that it was the low scorers who showed a tendency toward explanations in terms of sociopsychological dynamics.
The antiscientific thinking of the typical high scorer is closely connected with his tendency toward superstition, as discussed in Chapter VII. The tendency toward superstition is illustrated by the following records of high- scoring women:
F31: "I am not superstitious. Mother is a little bit superstitious. She believes in old-world customs and palmistry. I'm not a fatalist, either; what I do will control my life, what I make of it. But I don't know about that-there are the boys on the battle- field, for instance. They say a bullet has your name on it. "
F36: Subject does not believe in formal religion; this is why she likes the Uni- tarians. She does not think the churches should have a narrow, strict creed and tell you exactly what you should believe. She has read a great deal of theosophy, Madame Blavatsky, etc. She believes in reincarnation and divine will: reincarnation in the sense that the soul goes from one body to another and that you will be sub- jected to those experiences that are necessary to learning, to enable the soul to reach a higher state.
F6o: (Why did you come to ? -) "Why, I don't know! It just happened. Don't you think some things just happen to us? " (Superstition? ) She was just "called" to do it. Once she was out walking in the early morning-the birds were singing-she raised her hands and her face to the sky, and they were wet. (What was it? ) She considered it a supernatural phenomenon.
Along the same line Lentz (67) reports that conservatives are more an- tagonistic to science, especially with regard to its future activities, and, conversely, are more superstitious. They feel much more favorable toward the conventional, the traditional, and the routine. They are less tolerant and sympathetic towards the underdog, less aesthetic and less imaginative.
The fact that high scorers on ethnocentrism are more often given to stereotyping, pre-judgments and ready generalizations, or else to overcon- creteness, should not blind us to the fact that there also are tendencies of this kind in the low scorers. The increasing complexity of the social realities and their partial unintelligibility to the individual sometimes may necessitate a falling back on stereotypes so that opaque events appear more compre- hensible (see Chapter XVII). Ethnic prejudice is but one of many pos- sible media for this tendency. It must be held in favor of the ethnically un- prejudiced, however, that they make a serious effort to counteract such stereotyping in one of the areas of paramount social significance.
4. ANTI-INTRACEPTIVENESS AND AUTISM
Likewise previously discussed but rated here directly and in its own right is the tendency, found primarily in the high scorer, toward what may be
? THE AUTHORIT ARIAN PERSONALITY
called anti-intraceptiveness as contrasted with a greater readiness toward intraception in the low scorer (Category 6o). The difference is significant at the I per cent level. The concept of intraception covers the tendency toward introspection, as well as a readiness toward gaining insights into psychological and social mechanisms. It is contrasted primarily with ex- ternalizations of various kinds as referred to above. The greater creativity, imagination, and ability for empathy of the low scorers just discussed is likewise related to their greater intraceptiveness.
It may suffice to cite here in addition to previous quotations only one record of a low-scoring woman and one of a low-scoring man in which the tendency toward reflection becomes apparent.
F7o: Always wanted to work with people. . . . (What does religion give you? ) "I suppose going to church takes a load off of me of thinking about things I should think about. I think it covers my social ideas, and it causes me to think about things I must think about for some reason. (What about? ) Moral values, the relative value between peace and liberty in wartime, for instance. Pacifism and its ins and outs, interpersonal relationship as moral values. I don't think of those values as right or wrong, except as it has social implications. "
M so: (Future? ) "l don't know. I just recently came of the opinion that it is not necessary for me to be a commercial artist. Important to do whatever I do well. Not important what I do any more (Q) Well, I was the prima donna type . . . built up myself into an idea that not suited for the work . . . now it doesn't matter any more . . . it began when I was here about a month . . . the child guidance center psychol- ogist gave a speech . . . he gave the idea that (deviation in behavior has a cause) . . . . It may go back to infancy. . . . I play around with it (leading to nothing) . . . until a friend also interested made a startling remark. . . . Then I thought a lot about it . . . put down reasons on both sides, on two sheets of paper and decided he was right. I don't mean I am in love with my mother, but I have a dependency com- plex . . . married a woman older than myself . . . and always depend on others . . . leave responsibility to others . . . it seems on looking back that I have always done that. . . . "
The -absence of proper orientation toward social and psychological dy- namics in the typical high scorer may be linked to a general tendency toward autistic thinking in goal behavior as contrasted with the somewhat more realistic attitude that might be expected to be present in the typical low scorer (Category 62). Though not significantly differentiating between the two extreme groups of interviewees, this category shows a distinct trend in the direction that we expected (2I positive as compared with I2 negative instances in men, and I9 positive as compared with I2 negative instances in women). The trend is particularly evident in the vocational choices and economic expectations of our subjects.
The somewhat adolescent and fantastic, glamour-seeking character of the vocational ambitions of high-scoring subjects is illustrated by the following records:
F66: "If you're good, you can get up to ambassador. I think there have been
? PERSONALITY ORGANIZATION SEEN THROUGH INTERVIEWS 467
some women ambassadors. Or maybe there were only women ministers. I made the choice about three years ago. I just heard some friends of mine talking about it, and it sounded interesting. (Why vocational choice? ) Well, salary had a little to do with it. I think if I could really put myself to it, I could do it. If I really wanted to and had nothing else to stand in the way. The times have much to do with my choice. I would like to know why they do what they do. "
F34: Subject wants to be a journalist. This is purely utilitarian. She likes journal- ism, but her real desire is to do creative writing. She has imagined herself as a great actress. "But my acting is purely amateur stuff. I was always active in school dramatics as well as high school journalism. The family used to laugh because I was always play-acting. I always said I'd earn my living at either acting or writing. . . . "
5. SUGGESTIBILITY
Submission to authority and lack of independence and of critical judg- ment tend to lead the high scorer toward being suggestible and gullible, as contrasted with the greater autonomy of the low scorer (Category 6I). Again, the difference is significant at the I per cent level for both men and women. The social implications of a tendency on the part of the gullible person to fall easy prey to unsound and destructive political propaganda are obvious. The high scorer's dependence, in his personal life, upon author- ity, conventionalized values, church dogma, public opinion, and prestige figures, and the low scorer's relative independence of, and occasional rebel- lion against, these authorities have been pointed out so often throughout this volume that no further comment seems necessary.
The present chapter will be summarized together with a general synopsis of the interview results at the conclusion of the chapter which follows.
? CHAPTER XIII
COMPREHENSIVE SCORES AND SUMMARY OF INTERVIEW RESULTS Else Frenkel-Brunswik
A. THE DISCRIMINATORY POWERS OF THE MAJOR AREAS STUDIED
1. VERIFICA TION OF ANTICIP A TED TREND BY CA TEGQRIES
In the preceding four chapters (IX to XII), the technique of interviewing was described, and specific results of interviews with eighty subjects scor- ing extremely high or extremely low on the overt anti-Semitism or Ethno- centrism scales were presented and discussed in terms of the approximately ninety rating categories of the Scoring Manual used to classify the records of the subjects.
As shown in the tables of these chapters, a sizable proportion of the cate- gories differentiated high scorers from low scorers to a statistically sig- nificant degree in the direction anticipated for either or both sexes. The vast majority of the remaining categories showed at least a trend in the expected direction.
For men, there are only three out of the eighty-six categories for which the number of "positive" and "negative" instances is equal, and only one (Category 2 3c) which shows a slight trend in the direction op- posite to the one expected. As can be seen from Tables 3(X), 1(XI), and 2 (XI), to which the four exceptions are limited, the absolute number of rat- ings involved is very small in each case; the items involved deal with certain aspects of attitude toward siblings, sex, and people. For women there are two out of the ninety categories that show equality and four that show reversal, all six exceptions being confined to Tables 1(X), 2 (X), and 3(X), and deal- ing with aspects of the relationships to parents and siblings. In no case does the reversal approach statistical significance; nor does it appear in the case of any category that yielded less than 50 per cent "Neutral" ratings and was thus included in the list of those for which intensive evaluation was under- taken (72 for men and 65 for women, see end of Chapter IX).
468
? Based on categories presented in tables
20 men and 25 women
20 men and 15 women "low scorers"
Sums of Means '
en
1 ( X ) , 2 ( X ) , 3(X)
P a r e n t s ? a n d Childhood
Men 8. . . . . . 613 women 7. 47
L 37 3. 26
3. 11 1. 68
8. 42 5. 11
17. 10 12. 58
4. 47 4. 94
1(Xl)
Attitude toward Sex
Men 10. 57 Women 10. 43
0 "':: ,. . . .
2(Xl)
Attitude toward
M 3. 89 i1:l
3(Xl)
Attitude toward Men Present Self Women
13. 00 11. 33
2. 44 3. 56
2. 44 1. 11
13. 11 9. 67
26. 11 21. 00
4. 88 4. 67
~
4(XI)
Attitude toward Men Childhood Self Women
8. 50 8. 00
3. 75 5. 75
2. 25 12. 75 1. 00 8. 75
21. 25 16. 75
6. 00 6. 75
q r --l en
1(XII)
Dynamic Character Men Structure Women
11. 65 11. 75
2. 53 3. 59
3. 10 11. 35 1. 95 9. 12
23. 00 20. 87
5. 63 5. 54
2(XII)
Cognitive Personality Men Organization Women
14. 67 14. 67
3. 00 3. 50
3. 00 14. 00 1. 83 11. 00
28. 67 25. 67
6. 00 5. 33
All tables
All categories Men Women
10. 91 10. 38
2. 24 3. 34
2. 81 10. 59 1. 69 8. 10
21. 50 18. 48
5. 05
Dealing with
Sex "high scorers"
"! :! ositive?
? negative?
~
Men People Women
10. 67 10. 67
1. 67 2. 78
2. 22 1. 44
11. 67 8. 00
22. 33 18. 67
4. 22 <. . . . . . M
TABLE 1 (XIII}
COMPOSITE RATINGS (MEANS} FOR MAJOR AREAS OF STUDY FOR "HIGH" AND "LOW" SCORING GROUPS OF INTERVIEWEES
Means o f High(H) and Low(L) ratings received by
! ! .
L
H
. k
a;: > ~
2. 86 1. 86
2. 57 2. 29
5. 86 9. 00
16. 43 19. 43
5. 43
4. 14 z
+ 5. 03 0\
--l
i1:l
M
en
\0
? THE AUTHORITARIAN PERSONALITY
Thus there are no "negative validities" in the evaluation of the interviews. This result is not surprising in view of the fact that the categories of the Scoring Manual were derived from previous empirical evidence, including a preliminary inspection of the interview material; even so, the fact that the subsequent "blind" rating procedure supported most of the hypotheses with which the evaluation began is strong evidence of their validity.
2. COMPOSITE RATINGS FOR SEVEN MAJOR AREAS
Considering the almost universally confirmatory trend of the results just discussed, the computation of average ratings for larger groups of categories seems justified. The results of this procedure are presented in Table 1 (XIII) in the same manner as in the preceding tables except for the fact that indica- tion of statistical significance is here omitted.
As can be seen by an inspection of the last two columns of the table, dif- ferentiation between high scorers and low scorers is consistently somewhat less clear-cut for women than for men. This may be due either to the un- equal size of the two samples of women interviewees, or to the fact, re- peatedly referred to above, that women were in the main interviewed at an earlier stage of the investigation than were men. It may also be that, in our culture at least, patterns of behavior and ideology are more coherent and consistent in men than they are in women.
Among all the areas studied, that of attitudes toward present self yields the most clear-cut differentiation between high scorers and low scorers. The ratio between "positive" (confirmatory) and "negative" instances is among the most favorable. At the same time, the absolute values of the positive in- stances are among the highest. This indicates, furthermore, that the number of "Neutral" ratings is relatively low for this area.
The categories pertaining to dynamic and, especially, cognitive personality organization likewise show excellent differentiation, a very high proportion of the differences being statistically significant.
Contrariwise, over-all results are least sharply defined in the area of attitude toward parents and siblings, for reasons outlined in Chapter X. This is in line with what was stated above concerning absence of reversal of the expected trend in the case of a few scattered categories pertaining to these topics. Even for this area, however, the ratio of positive to negative instances is, for the men, better than three to one, and thus the differentiation is quite satisfactory.
An explanation of the differences in discriininatory power among the various major areas covered by the ratings of interviews may be in the fol- lowing. An adequate survey of the early family situation requires a great deal more specific information than could be obtained in our necessarily circumscribed interviews, and this made for the large proportion of "Neutral"
? SUMMARY OF INTERVIEW RESULTS
ratings. Such aspects as cognitive organization and attitude toward self, on the other hand, may be inferred from expressions in a greater variety of fields, and thus the likelihood of insufficient information is reduced. Over and above this, the very nature of such formal variables permits the subject himself to choose from among a variety of "alternate" manifestations, many of them encompassed in a single rating category. This principle of shifting expres- sion of identical motivational tendencies is apt to give an advantage to mo- tivational categories regardless of how complete the specific information on any particular aspect of life may be. It is for the same reason that the value of generalized, synoptic ratings, as contrasted with specific or situationally limited quantification, was so much stressed at the beginning of the discus- sion of the problems connected with the evaluation of interviews (Chapter IX; see also 3I, 32, 34, 36).
B. V ALIDITY OF OVER-ALL SCORES AND RA TINGS OF THE INTERVIEWS
1. INDIVIDUAL COMPOSITE SCORE BASED ON ALL AREAS OF RA TING
Composite ratings, covering the entire range of the areas of rating, were also obtained for each individual interviewee. In a sense, these composite rat- ings define what may be called the "score" of the subject on the interview. (It was with a view to this final score that the manual was designated as the Interview Scoring Manual in spite of the fact that for each of the constituent categories ratings rather than scores were obtained. ) Since, as was pointed out in the preceding section, there were no reversals of the expected dis- criminatory trend on any of the categories yielding less than 50 per cent "Neutral" ratings, all of these categories were included in the final score. These scores are shown, for all the So subjects interviewed, in Tables I (IX) and 2 (IX). The score itself is given in two parts, defined by the sums of the "High" and of the "Low" ratings received by the individual. The difference between the sum of these two parts and the fixed total (72 for men and 65 for women), not shown in the tables, indicates the number of "Neutral" ratings received by the individual on the categories in question.
The column referring to composite standing shows the letter H or L, de- pending on the preponderance of the High or the Low score in the preced- ing pair of columns. 1
Means of the quantitative data are shown in the bottom rows of the tables. The ranges of the number of "high" and "low" interview scores for each of the four groups can be read directly from the two tables.
1 Concerning the use of capital and lower-case symbols and other technical data perti- nent to this section see Chapter IX, Section F, 3.
? 472 THE AUTHORITARIAN PERSONALITY
2. OVER-ALL INTUITIVE RA TING
AND ITS AGREEMENT WITH THE COMPOSITE SCORE
The last column in Tables r (IX) and 2 (IX) adds what was briefly men- tioned in Chapter IX as the intuitive rating of the interviewee. This is based on an informal synopsis and estimate after the blind interview-rating pro- cedure had been completed, and it was made without the rater's making explicit to himself the quantitative results in terms of specific ratings on the single categories.
Composite scores and intuitive ratings agree very highly with one another. There is only one subject, F72, among the So interviewees for whom there is a discrepancy between the two values. In her case the composite score
(H) is correct; the over-all intuitive rating (L) is incorrect.
3. AGREEMENT WITH THE QUESTIONNAIRE RESULTS
It will be noted that while there is consistency as to "low" vs. "high" standing in the A-S and E columns of Tables I (IX) and 2 (IX)-this standing having been the very criterion of selection of the interviewees-the subse- quent columns show a good deal of inconsistency. This inconsistency reaches major proportions in such cases as F2z or F39, both overt low scorers who nonetheless score among the highest on the interview, with respect to both the composite score and the synoptic over-all ratings. Similar instances of opposition, though not all as striking ones, can be found among the high scorers. Various other kinds of inconsistency of trend may be noted upon inspection of the tables.
The general agreement between various columns in Tables I (IX) and 2 (IX) containing letter indices may be expressed conveniently in terms of percentage. "Percentage agreement" is then defined as the percentage of equal-letter combinations (HH or LL) in any pair of columns considered. The figures on percentage agreement could easily be transformed into tetra- choric correlation coefficients.
The defining criterion of selection, extremely high vs. extremely low standing on the overt anti-Semitism or Ethnocentrism scale, shows a per- centage agreement of about 85 with both the over-all intuitive ratings and the composite standing on the interview. (This figure is an average of an agreement of about 95 per cent achieved by the rater whose material hap- pened to include the most complete interviews, and of an agreement of 75 per cent achieved by the other of the two raters whose data were more frag- mentary. )
Since composite ratings agree with intuitive ratings almost completely (see above), the figures for the agreement of overt ethnocentrism with the intuitive ratings are practically identical with those mentioned above for
composite ratings.
? SUMMARY OF INTERVIEW RESULTS
473
Intuitive as well as composite interview ratings show slightly less agree- ment with standing on the F and the PEC scales, but even here the figures are between 75 and So per cent.
C. SUMMARY OF THE PERSONALITY P A TTERNS DERIVED FROM THE INTERVIEWS
1. INTRODUCTION
The results of the "blind" ratings of the interviews, as discussed in the present and in the preceding four chapters, have shown that individuals ex- tremely high on overt ethnic prejudice tend to differ with respect to a great variety of personality traits from those extremely low on prejudice. Some of the personality traits discussed were also measured by other techniques, especially the questionnaire. The results obtained by the various methods independently are very similar. However, rating by categories describes and substantiates in a more systematic, organized, and controlled way the im- pressions formed about the personality differences between high scorers and low scorers in the course of intensive study of individual cases.
It has to be emphasized, of course, that these differences are based on an analysis of group trends within statistical samples and do not imply that every individual will exhibit most or even a large proportion of the features belonging to either the "high" or the "low" syndrome, as the case may be. As can be seen from Tables r (IX) and 2 (IX), certain individuals seem to possess a relatively large number of either "high" or "low" features while others seem to have features of both patterns, with a relatively slight prev- alence of one or the other. It should thus be kept in mind that the summary which follows deals with composite pictures of these patterns, abstracted from the study of groups, rather than with individual cases. Were we to lay greater stress on concrete personalities, the most frequent syndromes or combinations of trends within single individuals would have to be determined as an intermediate step, leading to the definition of subtypes within the prejudiced and the unprejudiced patterns. Some remarks pointing in this direction will be made in the following summary. The typology, as such, of the ethnically prejudiced, will be taken up more systematically in Chap- ter XIX.
The results are furthermore limited to trends found in individuals with extreme standing on the prejudice scale. How far the relations would hold for those with middle scores on prejudice has to be left open, since such in- dividuals were not included in the present intensive investigation by means of interviews.
Finally, it remains for future investigation to ascertain how far the inter- relationships found are interculturally valid or whether they are restricted
? 474
THE AUTHORITARIAN PERSONALITY
to certain specific cultures or subcultures, such as the one from which our sample of subjects has been drawn; namely, an urban and suburban popula- tion on the West Coast of the United States.
In an attempt to summarize the social and psychological factors which, within our limited framework, have been found to be related to prejudice, we will proceed in the following manner: First, the aspects and "themes" of the personality structure of the high scorer as differentiated from the low scorer will be described in a synoptic fashion.
