Verified by 
This is a guest post by Dr. Ernest Davis, a computer scientist at NYU. Although this essay was inspired by my article on how a lack of political diversity in social psychology undermines its scientific research, his essay goes well beyond our article, and includes religion and history.
Enjoy.
Lee
Contempt for conservatism, and its dangerous consequences
Dr. Ernest Davis, NYU
The analysis in (Duarte et al. 2014) of the bias in the social sciences against conservatives and conservatism is important and timely. (I am not at all competent to evaluate either the effectiveness or the reasonableness of their proposed remedies, and therefore will not discuss them.)
If anything, the article understates the blatant, explicit contempt shown to conservative views in the published scientific literature. For instance, Wilson, Ausman, and Mathews (1973) wrote:
The “ideal” conservative is characterized as conventional, conforming, antihedonistic, authoritarian, punitive, ethnocentric, militaristic, dogmatic, superstitious, and antiscientific.
After that long stream of insults, the reader is quite prepared to be told that conservatives also smell funny. Obviously, most or all of those adjectives could be have been replaced with equally accurate adjectives of a neutral or positive valence e.g. respectful of tradition, uneccentric, abstemious, forceful, stern, patriotic, and so on. They proceed to characterize conservatism as a “syndrome” that has to be “explained” whereas, by implication, being liberal is just the way normal people are, and as such demands no particular explanation. (Stankov (2008) also refers to the “Conservative syndrome”.) This kind of language would be appropriate for an article in Mother Jones, but seems entirely out of place in a scientific paper in Journal of Personality and Social Psychology.
Duarte et al. are primarily concerned with the damage that this kind of bias does to social science research. Equally or more important is the poisonous impact of these kinds of publications on the state of political discourse, particularly when echoed gleefully in articles in liberal publications such as (Mooney 2014). First, such claims obviously increase the dislike and distrust of scientists and science among conservatives, and the sense that the pronouncements of science are merely a liberal conspiracy. Second, the last thing that liberals in this country need at this time — and I write as a dyed-in-the-wool liberal — is more reasons to feel smugly superior. Third and most importantly, democracy is based on political discourse, and meaningful political discourse depends on, to some extent, taking what your opponent says seriously and engaging with it on that basis. If liberals believe that conservative opinions are atavistic remnants of attitudes that were adaptive when we were all living in caves or on the savannah, and can therefore be dismissed out of hand, then no serious discourse is possible.
If these negative views of conservatism were in fact entirely valid, then the scientific community would be in the difficult position of balancing the scientists’ commitment to truth against the good of society. However, since, as Duarte et al. demonstrate at length and in detail, they are certainly one-sided, often exaggerated, and sometimes false, there is no justification for it.
In this note, I want to add to the argument in Duarte et al. by making two further points. First, parallel to the contempt for conservatism, and related to it, is a pervasive contempt for religion in psychological studies of religious belief. This connection is explicit in works such as Kanazawa’s (2010) paper, “Why Liberals and Atheists are More Intelligent”. Second, the contemptuous views of both conservatism and religion are exacerbated by a lack of historical perspective and an uninterest in historical accuracy. I conclude with some general comments about the risks attendant in this kind of research and the caution that needs to be exercised.
Contempt for religion
Exhibit A here is the paper (Gervais and Norenzayan 2012) “Analytic Thinking Promotes Religious Disbelief.” The paper argues that religious belief is associated with system 1 cognitive processes (in the sense of dual-process theory (Kahneman 2013)) whereas disbelief is associated with system 2 cognitive process. They cite a number of experiments that, they argue, support that conclusion. The first they discuss is as follows: They gave subjects (college students) a questionnaire with statements of personal religious belief, such as “The devil exists”, and they gave the same subjects a number of brain-teaser math problems to solve, such as the following:
In a lake there is a patch of lily pads. Every day, the patch doubles in size. If it takes 48 days for the patch to cover the entire lake how long would it take for the patch to cover half the lake?
These problems were chosen so that there was an “intuitive” answer, which could be found by System 1 processes while finding the correct answer required System 2 processes. For the above problem the intuitive answer is 24 days whereas the correct answer is 47 days (since it doubles on the last day). They found that a high degree of religiosity correlates with low scores on the brain-teasers. Their explanation of these results is that both low scores on the brain-teasers and high degree of religiosity are to be expected reflect a greater preference for System 1 thinking, since religious belief is supported by System 1 characteristics such as “teleology, mind-body dualism, psychological immortality, and mind perception.”
Both the experiment and the analysis here are seriously flawed. First, the application of dual-process theory here is entirely invalid.1 The argument here is that the difference between “intuitive” religion and “analytic” disbelief is in some sense parallel to the “intuitive” wrong answer to the question about the lily pads and the “analytic” right answer. But there is no such parallel. For one thing, dual-process theory is a characterization of processes; it is not a characterization of beliefs. Moreover, it is not remotely plausible that System 1 processes are involved in getting the wrong answer to the lily pad question; after all, getting the wrong answer involves dividing 48 by 2, hardly an “intuitive” undertaking. The subject who gets the wrong answer has simply made a mistake. Making mistakes is a thing that all mathematicians do, even when they are thinking carefully about apparently simple problems (recall, for example, the debate about the “Monty Hall” problems, which some professional mathematicians, writing for publication, and having seen the correct answer, nevertheless got wrong.) 1Moreover, dual-process theory generally increasingly seems on shaky grounds (Evans, 2013).
Someone who makes a mistake on a tricky math problem is like a pianist who hits a wrong note. A pianist who hits a wrong note is not engaging in a different kind of cognitive process than one who plays the correct note; he is engaging in the same kind of process and executing it less well.2
In any case the only similarity between religious belief and the mathematical error being made in the lily pad question is the fact that Gervais and Norenzayan consider that religious belief is also an error. Certainly subjects who make the error in the lily pad problem are not being influenced by teleology, mind-body dualism, psychological immortality, or mind perception. As an outsider to the field, I am astonished that an argument based on a completely nebulous concept like “intuitiveness” could be considered to have any kind of validity.
One also suspects that there are framing effects in terms of the religious propositions that are posed, and that if other propositions had been posed, the negative correlation of religious answers to intelligence would be substantially weaker. The propositions that have been used to measure degree of religious belief are mostly quite unsophisticated in their formulation; and one thing that intelligent students learn very well is the importance of appearing intellectually sophisticated. (The propositions are also, not coincidentally, particularly characteristic of evangelical Christianity rather than other forms of religion.) For example, if the experimenters had asked for assent to the statement “The Bible is holy,” rather than “Angels exist,” or “When faced with an important decision, I try to act in accordance with the teachings of my religion” rather than “One should seek God’s guidance when making every important decision” they might well have gotten greater degrees of assent from a wider class of subjects.
In “Religion is natural”, Bloom (2007) takes belief in personal immortality as the hallmark of religion, and argues from there that religion entails a dualistic view of mind and body. Religion also posits noncorporeal agents such as God, which again entails mind/body dualism. He points out that children tend to have a dualistic view of mind and body:
When asked, in implicit and explicit ways, pre-school children will say they believe the brain is only responsible for some aspects of mental life, typically those involving deliberative mental work, such as solving math problems. But the brain is not essential for activities such as pretending to be a kangaroo, loving one’s brother, or brushing your teeth . . .This is done by people, not by brains.
Thus, religious belief is essentially a holdover from a childish error. Every step of this argument is flawed, to a greater or lesser degree.
1. Belief in personal immortality is common among religions but not universal. For instance, there is no mention of it, or barely any mention of it, in the Hebrew Bible.
2. Neither the supernatural beings that are the subject of religious worship nor the ghosts of the dead are necessarily non-material. The gods of polytheistic religions mostly were viewed as having bodies. In most descriptions of the immortal souls of the dead (e.g. the souls in Hades in the Odyssey or the souls in Hell in the Inferno) they are generally described as having at least many of the properties of material objects; e.g. they are spatially localized, have a human appearance, and suffer physical pain due to physical causes. In many religions, such as many forms of Hinduism, the most important form of immortality is reincarnation; this entirely avoids Bloom’s claim of dualism, since the consciousness is always embodied.
Religious philosophers have indeed argued against this view, but they have not always made much headway in changing the mind set of religious believers.
3. There is no mystery about why children think that the brain is associated with intelligent thought but not with other cognitive activities; that is how the word “brain” is commonly used. When their father tells them, “Use your brain!” he is not asking them to pretend to be a kangaroo, love their brother, or brush their teeth.
4. The change from a view of a physically material God to a purely spiritual one and from a view of survival after death as taking place in a material being to a spiritual one, used to be viewed as a major intellectual advance; it is ironic that religious believers are now being told that they would have been more sensible to have stuck with physical gods and physical ghosts. It is also ironic that when philosophers such as Daniel Dennett and David Chalmers speculate cheerfully about consciousness surviving death of the body by being uploaded into a computer or the like, that’s cuttting-edge epistemology; whereas when religious believers believe the same thing, they are holding onto childish illusions.
Contempt for history
It seems to me that a contempt for history, or at least an uninterest in it, is a contributing factor to both of these viewpoints.
This is well illustrated in (Hibbing, Smith, and Alford, 2014). The first paragraph begins as follows:
John Stuart Mill called it “commonplace” for political systems to have “a party of order or stability and a party of progress or reform”. RalphWaldo Emerson agreed, noting that “the two parties which divide the state, the party of conservatives and that of innovation, are very old, and have disputed the possession of the world ever since it was made” and he inferred that “this irreconcilable antagonism must have a corresponding depth of seat in the human condition.” The antagonism between two primal mindsets certainly pervades human history: Sparta and Athens; optimates and populares; Roundheads and Cavaliers; Inquisition and Enlightenment; Protagonus [sic] and Plato; Pope Urban VII and Galileo; Barry Goldwater and George McGovern; Sarah Palin and Hillary Rodham Clinton. . . . Is Emerson right in his claim that this division springs from a deep, possibly innate part of the human condition?”
Later in the article they write “. . .Emerson’s intuition was right. Politics might not be in our souls, but probably is in our DNA,” and they refer to the “ ‘irreconcilable differences’ that Mill, Emerson, and others have long suspected to be the basis of political beliefs.”
How can I write that this quote demonstrates an “uninterest in history”; surely it shows exactly the reverse? The problem is that the history is all wrong. First, they have completely misrepresented Mill and Emerson, neither of whom at all believed that political tendencies were innate. On the contrary, Emerson (1841) wrote, in the same speech that they are quoting above, “The war [between conservatism and liberalism] agitates every man’s bosom with opposing advantages every hour. ... It is the opposition of Past and Future, of Memory and Hope, of the Understanding and the Reason.” Emerson thus views conservatism and liberalism as universal natural tendencies, acting within everyone, each reflecting important aspects of the truth. Mill, in Utilitarianism (Hackett edition, ed. Sher, p.31) wrote “[I]f, as is my own belief, the moral feelings are not innate but acquired, they are not for that reason the less natural.” Mill strongly, and Emerson more weakly, also stress the importance of having both viewpoints represented in the political system; a viewpoint that is not much reflected in either Hibbing, Smith, and Alford or any of the other papers I discuss.
Second, their list of contrasts is both biased and peculiar. It is biased in that they have included mostly unpleasant or villainous conservatives and admirable leftists; they do not have “The Tsarists or the Bolsheviks; Chiang Kai-Shek or Mao Zedong; Lavoisier or Robespierre; Samuel Johnson or Jean-Jacques Rousseau.” Many of the individual comparisons are strange. The comparison between Protagoras and Plato is meaningless, since none of Protagoras’ political opinions have survived. Judging from the order, Hibbing, Smith, and Alford seem to view the Roundheads as the conservative party and the Cavaliers as the liberal party; the conventional wisdom is usually the other way around. One can, of course, argue it either way; on the one hand, the Roundheads closed the theaters, destroyed art, imposed Puritanical religion, and savaged Ireland; on the other hand, they were anti-monarchical, tolerant of the Jews, and at least in some cases such as Milton, enthusiasts both for free speech and for Galilean science. The truth is, the difference between the Roundheads and the Cavaliers does not at all align with the difference between Palin and Clinton in either direction; and neither do the divisions between the Roman optimates and populares, Sparta and Athens, Plato and any contemporary philosopher, or the Inquisition and the Enlightenment. (Whatever one thinks of Palin, her viewpoint bears no significant relation to that of the Inquisition.) There is certainly no reason whatever to suppose that the same genetic differences now supposedly found between conservatives and liberals would have been present in any of these earlier conflicts.
Of course, the actual politics of ancient Greece and Rome and so on does not matter to the question at hand, and neither do the political philosophies of Emerson or Mill. What is important about this is that Hibbing, Smith, and Alford cannot be bothered to look up even the text of the speech of Emerson’s that they are quoting; but they are perfectly willing to project their own theories onto these nineteenth-century liberal icons, who actually believed exactly the reverse; and willing to lump together a series of widely varying conflicts over two millenia into a single crude dichotomy.
More generally, it seems to me that the anti-conservative bias and, still more, the anti-religious bias in these social scientists draws on a parochial view that bien pensant liberal, secular thought of 2014 is a universal standard. If all your friends are liberals and areligious and you associate religion with the Inquisition, creationism, and Osama bin Laden, then papers like “Analytic Thinking Promotes Religious Disbelief” and “Why Liberals and Atheists are More Intelligent” are merely experimental confirmation of what you believed already, and you see no need to fuss over the details of the argumentation. If you associate religion with the intricate argumentation of the Talmud, the philosophy of Augustine, the architecture of the Great Mosque of Kairouan, the counterpoint of the Bach B Minor Mass, and so on, then such titles seem as inherently implausible as “Why Barbers and Tax Accountants are More Intelligent,” and you will anticipate that either the experimental setup or the argumentation is probably flawed.
Investigating why people believe X
The study of the psychological reasons why people hold one opinion or another on questions where reasonable people differ is a risky one, especially as regards questions where emotions run high. In particular, if the belief is one which the investigator does not himself hold and with which he holds no sympathy, the question, “Why do people believe X?” is apt to turn into the question “Why are people so stupid as to believe X?” I have noted above that these investigations study the questions of why people are conservatives and hold religious beliefs; not the equally valid questions of why people are liberals and atheists. Once you have asked “Why are people so stupid as to believe X?”, it is generally all too easy to find an explanation of why people are so stupid, generally an explanation that is even more insulting to the people involved. The explanation in turn reinforces your viewpoint; in view of this explanation, it is only to be expected that stupid people would believe X; hence X is, even more clearly than before, a stupid belief. Thus the study of the psychology of believing X infects the discussion as to whether X is true, which is an almost unrelated issue.
It is, of course, true that most religious beliefs and conservative political opinions are held for no very good reason. But that is just a special case of the fact that almost all beliefs of any kind, other than personal experience, are held for no very good reason. As William James (1897) wrote, “[W]e all of us believe in molecules and the conservation of energy, in democracy and necessary progress, in Protestant Christianity and the duty of fighting for ‘the doctrine of the immortal Monroe,’ all for no reasons worthy of the name.” With public knowledge, there is hopefully a reasonable correlation between the strengths of the foundations of a belief and its truth; in the case of personal beliefs, this correlation is vanishingly small. With few exceptions, the people who know the most about the Kennedy assassination and have the theories of the greatest explanatory power, are the conspiracy theorists.
I don’t do this kind of research myself, and therefore can hardly give much advice to those who do. All that I would suggest is that the scientist should start with an attitude of respect. The starting presumption should be, not that the belief in X is true, but that it is not stupid; and that, in all likelihood, in most cases, the psychological explanations of why some people believe X are no less and no more reasonable than the explanations that other people believe not X. In particular, the scientist should not begin his search for an explanation of why people believe X in their personality defects, or in their inheritance from the Pleistocene, or in holdovers from childhood, or in cognitive illusions; because if he looks for explanations in those places, he can almost certainly find them. We all have personality defects and ancestors in the Pleistocene; we were all once children, and we all suffer from cognitive illusions, whatever we believe. This kind of respect seems to me lacking in the literature I have reviewed.
Acknowledgements: Thanks to Gary Marcus for bringing Duarte et al. to my attention, and to Sam Fleischacker for supplying the quote from John Stuart Mill.
Footnotes
1. Moreover, dual-process theory generally increasingly seems on shaky grounds (Evans, 2013).
2. Incidentally, one reason that the answer “47 days” seems counter-intuitive is that it is physically impossible; the patch would have to expand from being about 700km in diameter on the 47th day to about 1000km in diameter on the 48th day. Thus, subjects are penalized if they draw on their correct knowledge of how biological systems actually grow. As is all too common with these kinds of brain teasers, getting the “right” answer depends on guessing which aspects of reality the questioner wants you to consider and which he wants you to ignore.
References
P. Bloom, “Religion is Natural,” Developmental Science, 10:1, 2007, 147-151.
J.L. Duarte et al., “Political Diversity Will Improve Social Psychological Science,” Behavior and Brain Science, to appear.
R.W. Emerson, “The Conservative” in Nature: Addresses and Lectures, 1841. http://www.emersoncentral.com/conservative.htm
J. St. B.T. Evans, “Dual-Process Theories of Deductive Reasoning: Facts and Fallacies,” in K. Holyoak and R.G. Morrison (eds.) The Oxford Handbook of Thinking and Reasoning, Oxford U. Press, 2012.
W.M. Gervais and A. Norenzayan, “Analytic Thinking Promotes Religious Disbelief”, Science, 336, 2012, 493-496
S. Kanazawa, “Why Liberals and Atheists are More Intelligent”, Social Psychology Quarterly, 73:1, 2010, 33-57.
J.R. Hibbing, K.B. Smith, J.R. Alford, “Differences in negativity bias underlie variations in political ideology.” Behavioral and Brain Sciences, 37, 2014, 297-350.
W. James, “The Will to Believe” in The Will to Believe, and Other Essays in Popular Philosophy, 1897.
D. Kahneman, Thinking Fast and Slow. Farrar, Straus, Giroux, 2013.
J.S. Mill, Utilitarianism, 1863.
C. Mooney, “Scientists are Beginning to Figure Out Why Conservatives are So ... Conservative,” 6 Mother Jones, July 15, 2014.
L. Stankov, “Conservatism and cognitive ability”, Intelligence, 37, 2009, 294-304.
G.D. Wilson, J. Ausman, and T. Mathews, “Conservatism and Art Preference,” Journal of Personality and Social Psychology, 25:2, 1973, 286-288.
