In January of 2008 I had an email exchange with John Jost, political psychologist at NYU and lead author of the 2003 article, “Political Conservativism as Motivated Social Cognition.” (Psychological Bulletin 129: 339-375). He had seen an advance copy of the book, and while favorably disposed toward most of it he took exception to the following passages in the conclusion:
“The data presented in this book for the most part agree with the findings from my earlier studies. Political conservatives in America tend to sleep well with a diminished range of dreaming, while American liberals are more likely to sleep poorly with an expanded range of dreaming. The differences are not absolute, but the trends seem consistent with their respective political ideals.
These findings correspond fairly well with other research on political psychology. For example, John T. Jost and his colleagues argued in an influential 2003 article that “the core ideology of conservatism stresses resistance to change and justification of inequality and is motivated by needs that vary situationally and dispositionally to manage uncertainty and threat.”[i] Reviewing the results of eighty-eight studies involving more than twenty thousand people, Jost et al. found that political conservatism was psychologically correlated with high degrees of death anxiety and dogmatism and low degrees of openness to experience, tolerance of uncertainty, and integrative complexity. Without corresponding data on political liberals it’s hard to know exactly what to make of these findings, but they seem consistent in many ways with the patterns identified in my dream research, and thus supportive of a classic social scientific view of conservatives going back to Adorno and his 1950 study of “the authoritarian personality.” Conservatives seem to have thicker psychological boundaries than do liberals, with less interest in anything that deviates from their traditional ways of living and more concern about possible threats to those traditions.[ii]
My hesitation to fully endorse this line of research stems from 1) its pathologizing approach to conservative beliefs and ideals and 2) its premise that there’s a clear, stable distinction between a conservative and a liberal personality. I believe it’s better to start political psychology research with the recognition that no one is purely conservative or liberal. Everybody’s personality includes aspects of both tendencies… Samuels, like Lakoff, is an avowed political liberal, and Jost et al., leave little doubt as to their greater sympathy for liberal qualities. The leftward-leaning tendencies of most social scientists gives us good reason to question the motivations of researchers who argue that conservatives are somehow less mentally healthy or psychologically mature than liberals. Personal bias plays a role in political psychology just like it does in every academic field, and the best we researchers can do is try to be honest with ourselves and continually test our ideas against new sources of evidence.”
Jost replied that “you seem to be perpetuating some common misunderstandings (or misrepresentations) of our work. For example, you state that our meta-analysis did not include “corresponding data on political liberals” and that our work assumes a “clear, stable distinction between a conservative and a liberal personality.” Both of these statements are false. In fact, participants’ political orientation scores were treated as continuous variables (ranging from extremely liberal to extremely conservative) in nearly all of the 88 studies in the meta-analysis. Thus, our research contains extensive data on liberals and does not assume a rigid categorical distinction at all; rather, all conclusions are comparative. You also claim that our research takes a “pathologizing approach to conservative beliefs and ideals” (p. 154) and imply that our (assumed) “liberal bias” led us to conclude that “conservatives are somehow less mentally healthy or psychologically mature than liberals” (p. 157). These claims are also false, and we clearly said so in an August 2003 Washington Post Op-Ed piece. All of the 9 cognitive and motivational style variables that we investigated are part of normal psychological functioning, and there was nothing in our articles that “pathologized” conservatives. What you are doing here is perpetuating false claims that circulated on right-wing websites by people who never actually read our research.”
Jost also offered links to several articles in which he makes his case against his conservative attackers:
My response to Jost starts with an acknowledgment of my imprecision in summarizing his research. In the conclusion I briefly discuss my findings in light of four different perspectives on political psychology—cognitive scientist George Lakoff, Jungian analyst Andrew Samuels, Neo-orthodox theologian Reinhold Niebuhr, and Jost and his colleagues. In the paragraph on Jost I discuss his work in connection to the long lineage of social scientific research on political “authoritarianism” starting with Adorno, and clearly from Jost’s perspective I failed to adequately credit his methodological advances over the early Adorno work. The primary target of my critical comments is indeed Adorno, and I could have done more to distinguish Jost’s research from his. That said, I stand by the general methodological concerns raised in that passage. Research like Adorno’s and Jost’s is predicated on psychological categories that may correspond to actual features of cognitive structure but also have the tendency to obscure the fluid dynamism of actual lived experience (of the liberal and conservative varieties). My approach in the book tries to integrate both dimensions, the cognitive-structural and the dynamic-experiential, and even though my efforts are very limited and preliminary, the initial results have been sufficiently promising to encourage the pursuit of more and better research. Indeed, my primary point in bringing up Jost’s work was to highlight the convergence in our findings. The sleep and dream data I present in the main portion of the book strike me as remarkably consistent with the basic results of the meta-analysis of Jost and his colleagues in their 2003 article. I’m curious to know how much further these correspondences extend.