The Psychology and Cultural Evolution of Religion Predicts Bias in Philosophy of Religion

cross-posted at Prosblogion


“What is the first business of philosophy? To part with self-conceit. For it is impossible for anyone to begin to learn what he thinks that he already knows.” —Epictetus, Discourses, Book 2, Ch. 17.

At HECC and CERC we study religion–its cultural evolution and transmission, its psychology, its emotional and cognitive makeup and more. But you might think that studies about how the plebs experience religion are fundamentally different than studies about how sophisticated academics experience religion. In a recent paper Paul Draper and I diagnose bias not just in religious persons but in philosophers of religion. If the guiding hypotheses of CERC about the psycho-social purposes of religion are accurate, extensive biases even amongst elite thinkers are what we would expect.

Before knickers get in twists, we are not arguing in this work that religious belief is epistemically unwarranted or unjustified. We aren’t crafting a debunking argument based on spurious causal sources of religious belief. Matter of fact religion is a sine qua non of modern life. It would be not only silly but wrongheaded to attempt to banish religion from the world. Religiosity covaries with lots of positive outcomes. (You wouldn’t want to live longer and be healthier, would you?) In the words of the Doobie Brothers, internal and external forms of religiosity are alright with me. More than alright. Besides, folks in and out of the phil religion biz have narrowed the meaning of ‘religion’ leaving out intriguing options like Ietsism and the religious implications of the Simulation Argument. At least those who want to “commit it then to the flames”, à la Hume in the immortal conclusion to his Enquiry Concerning Human Understanding, should first get straight one what they want to burn.

This doesn’t mean we pull punches. Religious philosophers of religion appear quite biased. Of course all philosophers are biased–me not least amongst us–but the cultural evolution and psychology of religion gives us special reason to think that religious philosophers of religion will be a few standard deviations higher than the mean. This is ruining philosophy of religion. Symptoms of the bias include partisanship, polemicism, narrowness of focus and the persistent evaluation of arguments in phil rel with in-house criteria, criteria that appeal to doctrinal commitments or some such. Examples of each abound. Here are two.

Partisanship: Even a philosopher like Richard Swinburne, who demands evidence for his religious views (and has very precise standards of what counts as evidence), takes partisanship for granted. He says that philosophical theists endeavor to provide “cogent arguments for the existence of . . . God” (2001, 3) while philosophers who are atheists “endeavor to show that there is no God” (2001, 5). The ubiquity of this attitude in the discipline suggests readers won’t find this worthy of mention.


Polemicism: In philosophy of religion opponents are labeled as “enemies” (Peter van Inwagen 2006, 6, referring to those who use the fact of evil to “attack” theism) and an argument is called “triumphant” by someone who admits that it fails to establish its conclusion (Alvin Plantinga 1974, 111). We strongly suspect that rousing emotional language suffused with violent connotations is not used with as much frequency elsewhere in philosophy. ‘Strongly suspect’ only because we don’t have the data yet. I’m working on it.

But if these are the symptoms, what is the diagnosis? We present a diagnostic hypothesis in three parts.

First, all human beings appear to be cognitively biased with respect to their preferred beliefs. Since Lord, Ross and Lepper (1979), cognitive psychologists have done untold numbers of experiments documenting the effects of cognitive bias on preferred beliefs. Taber and Lodge (2006) improve on earlier work and present participants with arguments for and against both gun control and affirmative action. In addition to testing hypotheses about prior attitude effects, confirmation bias, and disconfirmation bias, they also tested hypotheses about attitude polarization, attitude strength, and a sophistication effect. Unsurprisingly, in the case of confirmation bias, it was found for all groups examined that proponents of an issue sought out more supporting than opposing arguments. What struck me was the following result: this effect was significantly more pronounced for sophisticated respondents, who selected like-minded arguments 70-75%. (‘Sophistication’ is a measure of intelligence, testing participants’ political knowledge in this case.) Some of the most ‘sophisticated’ religious believers are theologians, philosophers of religion and pastors. Could it be that similar effects would be found amongst them? Yes but studies testing for this have not been carried out. Helen De Cruz’s forthcoming research on the psychology behind the design argument is something to look forward to in this connection. (Click here for a peek.)

“I think it clearly and abundantly evident that true religion lies very much in the affections.” So says Jonathan Edwards in A Treatise Concerning Religious Affections. If religion was selected for by cultures for psycho-social outcomes like heightened pro-sociality amongst the in-group, I couldn’t agree more with Edwards, which brings us to the second component of our diagnosis. Religious believers, including philosophers of religion, are likely to exhibit bias due to affective triggers. Edwards’ point has been well-documented by psychologists studying the covariance of rates of emotion and critical thinking. The upshot is that “those high in emotional conviction generated more refutational arguments [for preferred beliefs] than did those low in emotional conviction” (Edwards and Smith 1996, 17-18). Many religious philosophers of religion, having committed their whole lives to a body of religious doctrine, have strong emotions about their religious beliefs. Prior commitments associated with strong attitudes and emotions are strongly correlated in these and other studies with higher rates of cognitive bias and attitude polarization.

Third, religiosity as it is typically experienced by philosophers of religion is intimately tied to a suite of additional socio-psychological factors that lead to more bias. I refer to factors having to do with effects of priming with supernatural agency concepts, religious motivations for pro-sociality, religious means of reducing free-ridership and in-group cheating; and very positive features of religion, including religiosity’s covariance with empathy, honesty, and willingness to help. Religious people in groups are bound together more tightly–no doubt from the outside, it would appear often, enviably tightly–in ways that non-religious people are typically not. Pro-social bonds such as these appear to trump selfish, first-person motivations and lead to sacrifices of behavior for the good of the group.

But these same bonds covary with negative cognitive and behavioral outcomes too. Some studies indicate religiosity of a certain kind covaries with aggression against the out-group and bias against outsiders. (See John Teehan’s recent book for discussion of how this plays out in the history of between-group conflict in the major monotheistic faiths.) This bias is especially prominent against atheists today, as a forthcoming study Jen Wright and I have done about the social costs of atheism confirms. We also find it amongst religious philosophers of religion, who are in ‘combat’ with atheists. Who is likely to recognize a mistake or form the belief that an opponent’s argument or objection is ultimately convincing if that involves conceding defeat at the hands of a spiritual enemy in what Plantinga has called a “battle for men’s souls” (1991, 16)?

Together these three sources of bias function as conflicts of interest that set philosophers of religion apart from other groups of philosophers—from, say, four-dimensionalist metaphysicians. When a four-dimensionalist comes to give up that theory, she does not get fired or pressured to resign from her job, as would most religious philosophers of religion at religious institutions, were they to change their minds about theism. John Sanders is by accounts a good man, not a mere example. Religious biases affect lives. The sort of emotional convictions and social bonds that religious philosophers of religion have about things like God’s existence or Jesus’s forgiveness of sins have no parallel in the case of four-dimensionalists.

We identify some reasons for optimism in the paper-length version … but I am cynical. You might think that declaring one’s bias would alone assist in resolving the problem, for example. But not only doesn’t openly declaring a conflict of interest prevent bias, studies indicate it makes it worse!  But more unfortunate in our (and Epictetus’s) opinion is the way that the methods in philosophy of religion used to negotiate the terms of the debate themselves are subject to bias. One Christian philosopher wrote recently,

As Christian scholars we are of course free to entertain all manner of “what if” questions, some heterodox, some heretical. . . . While we’re free to entertain such thoughts, I believe we are constrained by our faith to answer them in certain ways. If it seems to me that a particular claim is well-argued but it contradicts a significant tenet of the faith . . . , then I should seek to refute rather than defend it. (Gary DeWeese 2012, 7)

Patients, heal thy selves.


Almost forgot that philosophy’s case of anosognosia makes that a bit difficult.

Humm. Paul, we didn’t talk about this. What are we charging for co-pays? I’d better get malpractice insurance!

Leave a Reply