SEARCH

SEARCH BY CITATION

Keywords:

  • authoritarianism;
  • political beliefs;
  • ideology;
  • psychological conflict

Abstract

  1. Top of page
  2. Abstract
  3. The RWA Concept: Construct Validity
  4. The RWA Scale: Predictive and Discriminant Validity
  5. Our Speculative Interpretations
  6. Acknowledgments
  7. References

We try to clarify Kenneth Thomas's (“Wild Analysis in Politics”) mistakes about our analysis of the question: Do right-wing authoritarian (RWA) beliefs originate from psychological conflict? We suggest, contrary to Thomas's analysis, an approach that builds on the kinds of psychodynamic causes we use to explain RWA and the importance of confirming them empirically.

To begin, we wish to thank the editors for an opportunity to respond to Kenneth Thomas's (2013) “Wild Analysis in Politics,” an article critical of our thinking about right-wing authoritarianism (RWA; Crouse & Stalker, 2007). We hope that our comments will raise some issues that future work in this area of political thought and psychology will investigate, especially by finding testable hypotheses and bringing social science evidence to bear on them.

The stance of our article should be made clear. The main point is that RWA followers, as defined by high scores on Altemeyer's RWA scale (Altemeyer, 1996), hold beliefs that may be influenced by powerful psychodynamic causes which are often outside of conscious awareness. We try to spell out in some detail what these causes might be. We describe, in effect, a psychodynamic model of how these beliefs might come about and the personalities they might form. Our article thus postulates a psychogenesis of certain political beliefs and characteristics of those who believe them. These two components of our article are, obviously, connected: i.e., these causes make for these personalities.

We make it clear over and over again that our article is speculative. Our intent is to introduce a psychodynamic account of the genesis of certain beliefs. We ask what sort of person would believe the statements in Altemeyer's RWA questionnaire (Altemeyer, 1996) and how would they come to believe them. We attempt to answer these questions along the lines of a model of belief formation that involves negative affects and conflicting impulses in a compromise: viz., the beliefs in question. We make it clear that we do not try to evaluate the empirical adequacy of each posit we use and that our claims should be prefaced by “pending further verification” rather than “definitive explanations of.” We wrote our article, in short, to speculate and provoke further analysis, not to present a laundry list of well-supported empirical conclusions.

Thomas does not seem to approve of our stance and seems to view our article as an attack on American conservatives. Our article is no such thing. It merely attempts to follow out the implications of a model of belief formation wherever they may go. Our article is hardly a veiled attempt to denigrate; it is an exploratory analysis. It is premised on the simple observation that some beliefs may be caused, and we seek to build a model of these causes from an analytic standpoint. While some of its interpretations may not be welcome to some people, that is no objection to them. If one can't follow a chain of reasoning to its conclusion, one is not doing one's rational duty in exploring hypotheses and their consequences—and that is what scientific study is all about, happy consequences or otherwise.

Thomas ignores all the caveats and qualifications that we preface the article with and include in the opening and concluding sections. He also misreads our article in various important ways, and he seems to misunderstand how one ought to go about criticizing social science work. Moreover, Thomas also displays confusion about basic social science matters in psychometrics, largely basic points about what it would take to disconfirm our claims or implications of them. His main tactic seems to cherry pick our article for sentences that he particularly dislikes and then to argue that they are false. We counsel readers to first look at our article before concluding that Thomas has made a cogent objection. More likely than not, he has simply lifted something out of context and ignored important parts of our argument. In this response, we will ignore Thomas's rather aggressive and ad hominem attacks on us and our article, save for mentioning that while one of us is liberal in political matters, the other is conservative. Bias, we guess, is in the eye of the beholder.

We organize our responses to Thomas's article under three topics: the RWA concept, the RWA scale, and our explanations of RWA followers. Our article is mostly about the last of these three topics, but Thomas's criticisms of our article provoke us to mention research that we believe psychologists and political scientists should pursue under the first two topics.

The RWA Concept: Construct Validity

  1. Top of page
  2. Abstract
  3. The RWA Concept: Construct Validity
  4. The RWA Scale: Predictive and Discriminant Validity
  5. Our Speculative Interpretations
  6. Acknowledgments
  7. References

Thomas does not seem to like Altemeyer's (1996) concept of RWA. He especially dislikes Altemeyer's use of the label “right” in RWA. We use RWA essentially the way Altemeyer does. RWA followers tend more than others to submit to established authorities, show more aggression in support of authorities than others, and be more socially conventional than others. Obviously, the authorities can be of many kinds, differ with historical period, and from country to country. What (or who) is considered an established authority also depends on a multitude of factors (e.g., forms of government and religion). In addition, what is considered (and by whom) submission and aggression in support of authorities differs in probably just as many ways, as does what is considered socially conventional. Thomas seems to think that Altemeyer's RWA concept can be undercut and discredited by armchair observations about some of these differences.

Altemeyer to his credit seems very aware of these complexities and does not seem to engage in quibbles about “your definition doesn't account for this or that.” He resolves them in the standard way of defining things precisely in the social sciences—he operationalizes RWA with his RWA scale. He recognizes, as we do, that someone else may operationalize RWA differently, or may operationalize different kinds of authoritarianism. Altemeyer seems to recognize, correctly we think, that no single scale will cover everything that someone else might want to include or exclude under RWA. To illustrate, take years of schooling as a measure of education. Asking a person their number of years of schooling, or their highest degree, fails fully to capture the conceptual meaning of their education or academic achievement (e.g., one's field, one's awards, one's grades, where one went to school) and also indirectly measures irrelevancies like their socioeconomic background. Even so, measures of years of schooling or highest degree pick out something worthwhile and can be used to measure educational achievement well enough in various circumstances.

Altemeyer, of course, recognizes the established ideas in psychology that no measure is a perfect measure of the concept it is intended to measure, that all measures are fallible, and that all measures are influenced by something or other they are not intended to measure. This view recognizes that RWA is hardly unidimensional and that much more research is needed to assess the potentially common factor structure underlying Altemeyer's measure of RWA and, say, measures of other kinds of authoritarianism, social dominance, ethnocentrism, systems justifications, and other related concepts. It may turn out that RWA is not a unique concept in that its measure or measures have unique variance not shared by other important and potentially more fundamental social or personality concepts like the ones just mentioned. But finding this out requires developing and assessing new and competing measures of authoritarianism and other concepts, not just armchair criticism of Altemeyer's RWA concept and its measure as Thomas has done. We encourage research that investigates the common, correlated, and unique factor structure of RWA and other related measures. Thomas seems unaware of the value of this avenue of research. We would especially like to see techniques like LISREL and TETRAD brought to bear on relevant data because they can ferret out latent causal structures far better than the usual methods of correlation and regression analysis.

The RWA Scale: Predictive and Discriminant Validity

  1. Top of page
  2. Abstract
  3. The RWA Concept: Construct Validity
  4. The RWA Scale: Predictive and Discriminant Validity
  5. Our Speculative Interpretations
  6. Acknowledgments
  7. References

Thomas discounts and ignores Altemeyer's work assessing the predictive and discriminant validity of his RWA scale. Altemeyer covers this work in three books available at the time of Thomas's article (Altemeyer, 1981, 1988, and 1996). The items on Altemeyer's RWA scale may not capture every nuance about RWA that everyone can think of, as Thomas seems to demand. It was never meant to. It is a sorting device, and it is to be judged by whether it sorts groups well or poorly as Altemeyer defines RWA. It seems unfair, and beside the point, to pick apart the wording of Altemeyer's RWA items because you think they don't measure your concept of RWA while at the same time ignoring Altemeyer's extensive research supporting both the predictive and discriminant validity of his scale. Altemeyer has shown that his RWA scale, despite Thomas's dislike of it, has good reliability, predicts many things it should predict, and does not predict things it should not predict. The scale thus appears to have both predictive and discriminant validity. It does its sorting job well enough to be a standard psychometric instrument on this topic, if not the standard instrument.

Moreover, predictive and discriminant validity is a far more complicated matter than Thomas makes it. Thomas complains that RWA correlations explain only 25% of the criterion variance of a variety of predicted outcomes such as ethnocentrism and beliefs in a dangerous world. (At least that is what we take Thomas to mean by this 25% figure; we presume it is an “eyeball” computation from correlations in our Table 1.) For one thing, 25% of the variance is substantial considering the amount of measurement error that may be present. Anyone who thinks that 25% of the variance is too low should reflect on whether he or she would enjoy living in a society where RWA explained more than 25% of the variance of these outcomes.

But far more important, considerations of predictive and discriminant validity must assess how predictive and discriminant correlations arise, in tandem with construct validity as we discussed above. For example, techniques exist to model the latent structure of the constructs assessed by the scale (construct validity), together with the causal relationships of these constructs to outcomes of interest (predictive and discriminant validity). This is not the place to discuss this avenue of study in detail, save to mention that any good introductory text on structural equation modeling is a good place to begin. It is this kind of work that we think eventually will demonstrate how useful the RWA scale will prove to be in social science, and it is also the kind of work that Thomas does not engage in or even allude to as appropriate here.

Our Speculative Interpretations

  1. Top of page
  2. Abstract
  3. The RWA Concept: Construct Validity
  4. The RWA Scale: Predictive and Discriminant Validity
  5. Our Speculative Interpretations
  6. Acknowledgments
  7. References

However, we can set aside all the issues raised above for the moment. They mainly suggest directions for debate and discussion, and thus the direction research should take. Thomas saves his most biting criticisms for our efforts to broaden psychological interpretations of what goes on in the minds of persons who agree with the items on Altemeyer's RWA questionnaire. He says that we may be unprofessional, apparently only for positing these explanations. He does not appear to understand that the scale's beliefs stand on their own—people hold them—indeed, a proportion of people agree to them, for example, in group after group to whom Altemeyer has given the scale; by now, the scale has been given to more than 50,000 people. This is true even if you don't like the beliefs—or whatever label you put on them. We say there are a spectrum of cases and that many things can produce political beliefs, not just the things to which we point. We also say there are a spectrum of believers—from the most adamant hate group, say, to those who have more traditional conservative social, economic, and political views. This merely indicates that some score high on the scale and others less so; it does not equate them but it does indicate commonalities of one sort or another. As we note, as does Altemeyer, there are also commonalities between right- and left-wing authoritarianism, and thus some things true of one group are true of the other. Altemeyer has also developed a LWA scale, and he discusses this concept and how to operationalize it in his 1996 book.

We emphasize that our accounts of these beliefs are of value because they use familiar ideas from psychoanalytic conflict theory to shed light on new things. They apply ideas from one area (conflict theory) to a different area (beliefs from Altemeyer's RWA questionnaire) to shed new light on the latter area. We posit a small number of general causes from contemporary conflict theory (not classical Freudian conflict theory, as Thomas misunderstands), from which we deduce many specific characteristics of RWAs, and we see if this squares with social science evidence when we can. Our deductions are tendency claims, not universal or categorical: Xs tend to be Ys, as in RWAs tend to believe that clearer rules are necessary. Sometimes it amounts to most Xs are Ys, but in all cases it is more than some few Xs are Ys, which is what Thomas thinks would show our claims wrong. A few cases to the contrary don't matter; many, many cases would, and an armchair isn't where you find this kind of proportion. It takes, of course, assembling a good sample and then generalizing from it. Thomas does not have the sample-based studies to offer as disconfirming evidence.

Thomas does not seem to approve of our explanatory strategy: We take beliefs from the high-scoring end of the RWA scale and then speculate about what sort of person would hold such beliefs. We make use of standard psychodynamic explanations from contemporary conflict theory and their associated belief outcomes on the RWA scale, and when we know of relevant data, we allude to it. We do not attempt to defend every aspect of contemporary conflict theory any more than one argues for Newton's mechanics when one uses it to compute where NASA satellites will go. That is for another time. We use it to explain, speculatively, RWA.

Thomas also does not even like our taking beliefs from the RWA scale to study. He complains that the RWA items are confusing to respondents but apparently doesn't see that the RWA scale's good test-retest reliability (e.g., 0.85 over 28 weeks) is not consistent with confused respondents. He says that some of the items fit left-wing authoritarians, or that they aren't true of today's Republicans, or they are true of today's Democrats, or that still other things are true of today's Republicans. He claims repeatedly that beliefs we say are true of high-scoring RWAs are also true of many others besides high-scoring RWAs. We believe all this is obviously true. Thomas seems to think that when we say, for example, that RWAs have certain beliefs that we are saying all and only RWAs have these traits. He thus points to some who may not, or others who do, and believes he has disconfirmed our claim.

We never make universal generalizations, and we never claim that things are unique to RWAs. We never say or imply that any belief from the RWA scale is true only for high-scoring persons, only that it shows up to a certain notable degree in that group as shown by the results of Altemeyer's many surveys—enough so that Altemeyer's scale reliably identifies those who hold certain political and social beliefs. If Thomas doesn't like the beliefs we posit psychodynamic conflict-oriented explanations for, he can always select different ones. However, we believe that the beliefs on the RWA scale may be representative of beliefs held by (some, perhaps more than is widely known) governmental and nongovernmental leaders, as work has shown in the fields of political-personality profiling and political leadership and leader-follower relationships (e.g., Dean, 2006; Post, 2003). We believe they deserve recognition and study for this reason alone.

Whether we—or anyone—are correct in calling the beliefs we posit explanations for “RWA beliefs” depends, of course, on resolution of the conceptual and validity concerns raised above. As we noted, the RWA concept eventually may be better replaced by more fundamental concepts. But the beliefs stand on their own, no matter what label they are given. The posits and interpretations we make do not depend on resolution of every conceptual and validity concern. What is at issue is whether you want to call these RWA beliefs or use some other label for them. They stand by themselves as interesting in their own right. At this point, subject to the concerns raised above, we remain comfortable with the RWA label.

Our general strategy—certainly not new to us—is applicable to many social concepts: namely to take any operationally defined definition of the social concept and assess how well psychodynamic concepts (in our case from contemporary conflict theory) explain it. Of course, this is speculative explanation of the social concept and requires empirical evidence to back up the explanation at some point. We make this very clear in our article. We could have used any collection of beliefs with this strategy. We chose RWA beliefs because we are interested in them, not because we have a political agenda to denigrate alleged opponents. We chose concepts from contemporary conflict theory to explain RWA because we believe empirically minded social and personality psychologists are more ready to take these contemporary conflict concepts seriously as explanations of social and political phenomena than they have in the past. It seems to us, and others too, that social and cognitive psychology has in the past 10 years or so recognized the usefulness of quick, automatic, emotional, largely unconscious impulses like those from contemporary conflict theory to explain social and political phenomena. An excellent example of this is the recent study in this journal by Thórisdóttir and Jost (2011) showing that motivated closed-mindedness—of the kind covered by Altemeyer's RWA scale—mediates the effect of threat on political conservatism. To us, this is a clear example of empirical support for the kind of psychodynamic posits we make in our article; namely, that defenses against threats from uncertainty and other sources lie behind RWA beliefs, and they are projected onto society as aspects of political conservatism.

We encourage others, like we have tried to do, to examine the usefulness of contemporary conflict theory for speculative explanations of what is going on in the social and political lives of leaders and followers. We also encourage studies like Thórisdóttir and Jost (2011) that use quantitative methodology such as experimental, quasi-experimental, and latent variable modeling to garner support or rejection of these speculations.

Acknowledgments

  1. Top of page
  2. Abstract
  3. The RWA Concept: Construct Validity
  4. The RWA Scale: Predictive and Discriminant Validity
  5. Our Speculative Interpretations
  6. Acknowledgments
  7. References

Correspondence concerning this article should be sent to James Crouse, 3200 Gulf Shore Blvd. North, Unit 109, Naples, FL 34103. E-mail: Jcrouse@udel.edu

References

  1. Top of page
  2. Abstract
  3. The RWA Concept: Construct Validity
  4. The RWA Scale: Predictive and Discriminant Validity
  5. Our Speculative Interpretations
  6. Acknowledgments
  7. References
  • Altemeyer, B. (1981). Right-wing authoritarianism. Winnipeg: University of Manitoba Press.
  • Altemeyer, R. A. (1988). Enemies of freedom: Understanding right-wing authoritarianism. San Francisco: Jossey-Bass.
  • Altemeyer, R. A. (1996). The authoritarian specter. Cambridge, MA: Harvard University Press.
  • Crouse, J., & Stalker, D. (2007). Do right-wing authoritarian beliefs originate from psychological conflict? Psychoanalytic Psychology, 24, 2544.
  • Dean, J. (2006). Conservatives without a conscience. New York: Viking Press.
  • Post, J. M. (Ed.). (2003). The psychological assessment of political leaders: With profiles of Saddam Hussein and Bill Clinton. Ann Arbor: University of Michigan Press.
  • Thórisdóttir, H., & Jost, J. (2011). Motivated closed-mindedness mediates the effect of threat on political conservatism. Political Psychology, 32, 785811.
  • Thomas, K. R. (2013). Wild analysis in politics. Political Psychology, 34, 927934.