Shook (2017) argues that if god-beliefs are “innate,” one is obligated to be skeptical about them by virtue of their mutually incompatible plurality and nativist origin. Second, Shook suggests that even if god-beliefs are not innate, it is still epistemically vicious to believe in gods. Shook also raises concerns about using theology to motivate or interpret scientific inquiry. This response essay clarifies the character of the theories offered in the cognitive science of religion (
Suppose someone argued that everyone is irrational or intellectually vicious to believe their moral convictions are true on the grounds that (1) people hold a variety of beliefs about moral truths and (2) scientists have reason to think that moral beliefs are, in large part, a product of the operation of evolved cognitive systems in human minds. Suppose, similarly, that someone argued that everyone who believes that other humans have minds should be skeptical about such beliefs on the grounds that (1) people have a diversity of beliefs about minds and what they are and (2) scientists have reason to think that belief in minds is importantly predisposed because of human biological endowment. Similar suppositions could be generated concerning beliefs about grammatical constructions in language, whether and in what way my spouse or mother loves me, and how humans get sick from unseen pathogens. My guess is that, in all of these cases, we would be unimpressed by the argument for irrationality or intellectual vice in holding these beliefs. Belief diversity—even clear disagreements about the facts across individuals—coupled with some kind of nativist account for the origins of the beliefs does not mean that our beliefs are not rational to hold or that they are epistemically suspect. Humility may be demanded, but not full-bodied skepticism. Yet, an argument of this structure is what Shook offers in his article.
In Shook’s article, I detect one primary argument plus two subsequent important assertions. First, Shook argues that if god-beliefs are “innate,” one is obligated to be skeptical about them by virtue of their mutually incompatible plurality and their nativist origin. Second, Shook suggests that even if god-beliefs are not innate (a possibility that he rightly rejects), it is still epistemically vicious to believe in gods. Third, if I understand him correctly, Shook means to suggest that using theology to motivate (or interpret?) scientific inquiry amounts to “interference” in science and is out-of-bounds. In this essay I briefly evaluate these claims, concluding Shook’s primary assertions are not adequately argued or substantiated. Before evaluating these claims, however, I make a few notes of clarification concerning cognitive science of religion (
Points of Clarification
The Naturalness of Religion Thesis is not Equivalent to Claiming that Religion or God-beliefs are Innate
If innateness is at all a useful concept, it means something like nearly inevitable due to human biological endowment, including genetics. It does not mean, for instance, present at birth, because many ordinary features of being humans, such as having a complete skeleton made of bone, walking up right, and sexually reproducing, are not present at birth. Shook begins by arguing from the claim that “religion” is innate and spends some time arguing that religion and god-beliefs are not innate. Shook asserts that I have endorsed and argued for the innateness of religious thought, but this is mistaken. In my 2012 book, Born Believers, the word “innate” only occurs once and that is in a quote of another scholar; and in my 2011 book, Cognitive Science, Religion, and Theology, I explicitly criticize the scientific utility of the concept of “innateness” and adopt instead Robert McCauley’s concept of maturational naturalness (McCauley 2011), a concept that captures typical cognitive development that arises by virtue of being an ordinary member of our species developing under ordinary human environmental conditions including socialization. I use the term natural as short-hand for McCauley’s “maturationally natural.” Natural does not entail goodness or correctness, and unnaturalness does not imply badness or falsehood.
Jesse Bering is one of the few current lead scientists in
Cognitive Naturalness may be Socially Tuned Up
A key feature of at least some cognitive naturalness is that it may develop through social interaction. For instance, language use is natural in humans. Nevertheless, language development may require exposure to language used in human social interaction in order to mature. Aspects of social intelligence, such as theory of mind or perspective-taking, may develop similarly. Furthermore, human (and other primate) relational attachments systems may require early social interaction. That social input is required does not mean a trait is not natural for humans—humans are naturally social animals. Hence, comparing religious thought (such as belief in gods) to talking, playing, singing, and artistry is not confused or misleading. All of these common human forms of expression may have strong natural bases and likewise require social input to develop properly. Some aspects of some of these modes of cultural expression may have stronger claims to naturalness than cultural elements typically termed “religious,” but that is an open empirical question.
The Naturalness of Religion Thesis is not Owned by Theists
One may get the impression from Shook’s article that the naturalness of religion thesis is a theological intrusion into science by partisan theists—such an impression would be almost entirely backwards. Some of the earliest book-length explorations of the cognitive naturalness of religion can be found in Lawson and McCauley (1990), Guthrie (1993), and Boyer (1994). Other prominent works advancing a naturalness of religion thesis include Boyer (2001), Atran (2002), Pyysiäinen (2003), Tremlin (2006), Bloom (2007), Bering (2011), and McCauley (2011). Most of these scholars have revealed in their writings or public presentations that they are not theists. In fact, Bering, who offers the most nativist treatment of the bunch, is most explicitly atheistic.
The Naturalness of Religion Thesis is not Viciously Circular
Nevertheless, circular arguments for god-belief’s [sic] innateness remain common. They usually boil down to this: Humanity’s long-standing interest in god-belief (which hasn’t been proven) points to its innateness in all members of our species, while humanity’s universal belief in gods (which isn’t accurate) is best explained by our ancient facility with conceiving them.
Setting aside the parentheticals for now (see below), Shook has misrepresented the basic argument structure of naturalness theses, and, in spite of the alleged commonness of such circular arguments, has failed to cite any examples as cases to back up this characterization. Reasoning about naturalness typically proceeds along the following lines (e.g., see books cited above and Barrett 2004, 2007):
- Historically and currently, belief in superhuman beings (i.e., gods) are extremely widespread within and across human groups, many of which bear similarities to each other without any known common history. This apparent fact requires explanation.
- To explain cross-cultural regularities in cultural expression it is best to appeal to cross-culturally recurrent features of human experience (e.g., environmental, biological, cognitive, social, etc.). That is, cultural particulars would not easily be able to explain cross-cultural patterns.
- Given that we are considering beliefs—a product of human cognitive systems—a good place to look for explanations (even if only partial) for human god-beliefs is human cognitive regularities.
- Are there any human cognitive regularities (e.g., patterns of thought, ways of conceptualizing the world, etc.) that would generate, select for, or stabilize ideas and beliefs about gods?
- Among other types, evidence for such cognitive regularities may come from:
- comparisons with other similar species;
- cognitive developmental studies;
- cross-cultural anthropological and psychological studies; and
- experimental studies that use cognitive load, implicit cognitive measures, brain-scanning, and other techniques that can help peer “beneath” culturally-acquired modes of thinking to more natural impulses (e.g., see Bering 2011 for examples).
If these different sorts of evidence converge, then our confidence increases concerning the existence of the postulated natural cognition.
If insufficient evidence for natural cognition underwriting god-beliefs is available, then perhaps other factors (e.g., solving social problems that emerge from group living over a certain size) should be given more attention in place of cognitive ones.
There is nothing viciously circular about such a research strategy.
“Natural religion” is not the same as any Primal Religion from which others have Arisen or the same as an Infantile Religion Found in any Children
Shook expresses concern that I have proposed “the core of pure and true religion” to be under the label of natural religion in my book, Born Believers (2012). He misunderstands me as saying first, that the list of features I suggest for natural religion are the same as a primal ancestral religion from which contemporary religions have descended, and second, that I “credited naïve children with innate beliefs about gods … that accurately generate the pure religion.” Apparently my expositions in Born Believers and a similar one in my 2011 book were unclear. I regard the list of features that I provided as examples of belief tendencies that appear to support cultural expression that might be regarded as “religious.” Belief candidates—including theologies and other religious propositions—that are closer to these features will be easier for individuals to communicate and use and will be more likely to become shared and stable in a population. Hence, this list could be thought of as an ingredients list for building religions with a reasonable likelihood of cultural success. Nowhere do I mean to assert that any primordial religion had all of these features and from it all existent religions developed. Nowhere do I want to say that individual children hold an articulated set of religious beliefs that match these features. I have never asserted that what I have called “natural religion” is any core of pure and true religion, whatever that means.
The Degree of Naturalness of a Form of Cultural Expression is not Equivalent to how Ancient it is, although the Two Aspects are Related
Shook writes that “it should be seriously doubted whether religiosity should be categorized with walking and talking, or even singing and artistry,” on the grounds that, “religious expression takes advantage of far older human capacities for singing, dancing, ritual, artistry, and narrative.” If my naturalness thesis were an innateness thesis, this critique would be apt, with some softening: the evidence for the relative age of these various forms of cultural expression (excluding walking) has no consensus. As it is, my view of naturalness does not depend upon it being more ancient than other forms of cultural expression. I have argued that in addition to McCauley’s other heuristic features of maturational naturalness (McCauley 2011), picking out cultural forms as more or less natural is possible by considering how ancient a form is (Barrett 2011). If, for instance, it is only a few hundred years old (e.g., chess playing, knowing multiplication tables up to tens), then even fluent excellence in the area is likely the result of expertise due to special personal or cultural factors and not the result of ordinary maturation. However, the age of a trait is only heuristic and relative age is not the issue. Adult lactose metabolism may be maturationally natural for a sizable human population, but it is not nearly as ancient as burying the dead with grave goods. It does not follow, then, that burying the dead with grave goods is more natural than adult lactose metabolism.
Cultural Expression, Including Religious Expression, can Scientifically be Described as More or Less Natural
Shook thinks that describing some religions as more or less natural “won’t make sense from a scientific standpoint”—but perhaps he thinks that “natural” is evaluative in terms of goodness, rationality, or correctness. Again, following McCauley’s distinction, naturalness refers here to how closely undergirded particular beliefs and practices are by maturationally natural cognitive systems. From this scientific (not theological) perspective, the kind of spontaneous dancing that preschoolers do is more natural than ballet, and belief in ghosts is arguably more natural than beliefs in a triune God.
Humans are not Capable of Conceiving and Believing Almost Anything
Shook writes: “An undeniable fact is that humans, in general across the species, are capable of conceiving and believing almost anything.” I deny this “undeniable fact.” Human conceptual flexibility is considerable, but we have essentially no evidence that humans can conceive of and believe almost anything. We do, however, have evidence that humans find some ideas very challenging to conceive and believe, particularly given that belief is usually predicated upon being able to conceive it. Consider what it would mean to only remember things that never happened or what it would mean for matter to not have four dimensions (height, breadth, depth, and time) but twenty-five. Not only are some thoughts just hard to think, in a given culture, weaker constraints on the conception of and belief in ideas may greatly reduce the range of concepts that are likely to become widespread. We are simply better at thinking (including believing) some things over others (McCauley 2011; Sperber and Hirschfeld 2004).
Evaluating the Arguments
With these clarifications in mind, I turn to Shook’s primary concerns.
Skepticism, Innateness, and God-beliefs
Shook’s first argument—skepticism concerning god-beliefs—has two components. The first part argues that skepticism regarding god-beliefs follows if god-beliefs are innate. The second part argues that skepticism regarding god-beliefs also follows from god-beliefs not being innate but arising in some other way (e.g., socially conditioned). Shook presents this first part of his argument thusly:
- “If religion is innate, then all people are normally born to be religious.”
- “A person normally born to be religious will usually place belief in god(s).”
- “Humans have long been placing belief in a vast variety of extraordinary beings.”
- “The innateness for religion accounts for the way that humans are naturally and normally born to place belief in gods.”
- “It is not true that all of those seemingly credible unnatural powers and agents are actually real just as believers take them to be.”
- Religion’s innateness provides for most or all people normally and naturally placing their belief in unreal gods.
Shook takes (6) to follow from the previous five and to entail skepticism regarding god-beliefs. He refers to (6) as a “failure” of “religion,” a “skeptical result,” and suggests that god-belief must seek other grounds to be “epistemically reasonable.”
Concerning (1), as mentioned above, the “innateness” of religion is not as common a claim in the
The second reason for rejecting, or at least modifying, proposition (1) (and proposition (2) and (4)), is the ambiguous phrase, “born to be religious.” Such phrasing suggests a teleological direction, goal, or purpose or behind the (alleged) innateness of “religion.” Consider, “If body hair is innate, then all people are normally born to grow body hair.” Even if body hair is innate, growing body hair is not a human purpose. A more defensible version of Shook’s first proposition would be: “If religion is innate, then all people who share the relevant biological endowment are normally going to become religious.” As offered, Shook’s first premise should be rejected; and if he believes the rest of his argument to depend upon it, we may conclude now that he has not made his case for god-belief skepticism.
Premise (2) reads: “A person normally born to be religious will usually place belief in god(s).” If my modification of Shook’s “born to be” to “going to become” is acceptable, then proposition (2’) is: “A person normally going to become religious will usually place belief in god(s).” This version appears true if “religious” is taken to including belief in the existence of one or more superhuman beings or gods. Though definitions of “religion” are contentious, including gods as a component of these definitions is common enough to accept proposition (2’) for the sake of argument.
Premise (3) is: “Humans have long been placing belief in a vast variety of extraordinary beings.” If “long been” means “for hundreds or thousands of years,” this proposition is well-supported. If it is meant to mean a longer period such as tens (or even hundreds) of thousands of years, then the evidence is much thinner, but still likely enough.
Premise (4) is: “The innateness for religion accounts for the way that humans are naturally and normally born to place belief in gods.” If Shook takes “innateness” to mean that some trait’s expression is biologically determined in some strong sense, then it is obviously true that were such a trait innate, its innateness would account for the “natural and normal” expression. Propositions (1) and (2) seem to say as much, so why (4)? Shook’s subsequent discussion of the plurality of god-beliefs (which is assumed in (5)) tempts me to think (4) may mean: “The innateness for religion accounts for the way that humans are naturally and normally going to believe in each and every god(s).” That is, what (4) adds to (1) and (2) is that any given god-belief—and not just being religious or believing in some sort of god—is what is “innate.” Perhaps, however, this precision is not what Shook is proposing. It would be a very peculiar view of innateness to suggest that belief in Allah but not the ancestors is “innate” in some people, but the contrary would be true in others, so peculiar that I have never heard such a proposal by any scholar. Charity demands that I assume Shook is aware of such a peculiarity, and so by “innate” he must mean something softer than biological determinism of each particular god-belief that a human might have.
If I have Shook’s meaning right, then, the “innateness” of god-beliefs is not entirely biologically determined. Presumably, these beliefs may have some native disposition that makes them likely in people, but the particular beliefs—just which superhuman beings are believed to exist—arise through cultural learning mechanisms. Such a view of what Shook calls innateness is much closer to my version of the naturalness of religion thesis; and since it appears that Shook is primarily arguing against me, I take this as good evidence that I am on the right track in understanding Shook.
Premise (5) reads: “It is not true that all of those seemingly credible unnatural powers and agents are actually real just as believers take them to be.” Given that any two members of the same church congregation could characterize God differently from each other, (5) seems a safe claim—but I would take it farther. Humility suggests that it is not true that any (let alone all) gods are real “just as believers take them to be.” Perhaps, however, Shook means by (5) that not all of the gods and powers that have been posited by religious believers throughout history really exist at all. Though such a metaphysical claim may be scientifically inscrutable, I am prepared to accept even this interpretation of (5).
Shook’s conclusion (6) reads: “Religion’s innateness provides for most or all people normally and naturally placing their belief in unreal gods.” This conclusion fails to follow from the five premises in several ways. First, Shook’s premises have not established that “most or all” people believe in unreal gods, only that most people believe in gods and that not all of these gods really exist, at least in the way that they are understood to exist. It may be that no gods exist, one god exists, or an innumerable number of gods exists, and it may seem that the great diversity of god-beliefs must be false because different gods are attributed the same exclusive properties. If one person thinks god X is the creator of the cosmos and a second person believes god Y to be the exclusive creator of the cosmos, both cannot be correct. Nevertheless, it could be that god X and god Y both exist but one is the cosmic creator and the other is not (or neither are). It could also be that X and Y are different monikers for the same god. None of these various possibilities are ruled out by Shook’s argument.
Shook’s conclusion also does not follow because it is not the case that innateness, in a biologically deterministic sense, accounts for each and every god-belief in terms of their full range of properties. To argue so would be comparable to arguing that because language is “innate” in some sense, then “innateness provides for” each and every particular language (e.g., Hindi, Mandarin, Spanish), which is false.
If one takes a more expansive sense of “innate” that includes social and cultural input tuning up or filling in the particulars of natural biases and tendencies, then proposition (6) could at best be saying that some combination of “innate” tendencies and other social-cultural factors “provides for most or all people normally and naturally placing their belief in” gods, some of which are not real. Such a conclusion, however, does not compel religious believers to be skeptical about their particular god-beliefs. In fact, an enormous range of beliefs would fit this same profile, not just god-beliefs. Hence, in order to reach his skeptical conclusion, Shook is either depending upon a definition of “innate” that is wildly implausible or one that is broad enough to make us doubt the deliverances of our minds concerning whether or not our friends and spouses love us, other humans really have mental states, and causation is real, among other common, non-religious beliefs.
Skepticism, Non-innateness, and God-beliefs
Shook argues that whether or not god-beliefs are innate, they are epistemically dubious. Apparently even if one rejects “innateness” as an explanation of god-beliefs (as I do), one is still required to be skeptical regarding god-beliefs due to the broad diversity of mutually-incompatible god-beliefs that have been held. His reasoning appears to be that, given that the vast majority of god-beliefs are false, the probability of having a true god-belief is so small that one should be skeptical about their god-belief(s) because regardless of the evolutionary-cognitive-social-cultural complexes that give rise to god-beliefs, the diversity of outputs shows that the complexes cannot deliver consistent, true beliefs. Of course, such an argument—if one were inclined to accept it—would not only lead to skepticism concerning god-beliefs but huge ranges of beliefs. Given the diversity of moral judgments that have been entertained, wouldn’t one likewise be required to be a moral skeptic? Given the diversity of broadly metaphysical beliefs—beliefs in gods, afterlife, reincarnation, souls, conscious minds, free will, and atheism—wouldn’t one be required to likewise remain skeptical about those metaphysical beliefs? Indeed, it appears that Shook’s concern with belief diversity would lead to skepticism in all scholarship including philosophical and scientific. Apparently, the causal factors that lead to belief in any of these domains are too unreliable to be trusted, and so we should be skeptical regarding these scholarly claims, including this essay and Shook’s target article. That would mean, however, that Shook’s argument is self-defeating and we have no obligation to accept its conclusions. For more thorough discussions of the epistemological implications of cognitive and evolutionary studies of religion, I refer readers to the edited volumes by Schloss and Murray (2009) and Trigg and Barrett (2014).
No Place for Theology in Relation to Science?
If I understand him correctly, Shook is uncomfortable with using theology to motivate or interpret scientific inquiry, regarding it as “interfering” in science. Perhaps the clearest statement of this discomfort is in his abstract, but it pops up occasionally throughout the article. Such a topic is too great for a fair treatment here. I only observe that just as a-theological reflections may generate empirically testable hypotheses, so may theological ones; and it is not obvious—and Shook does not construct a clear and compelling argument to the contrary—that only a-theological reflections should be permitted.
I have encouraged theologians to consider interpretations of
One may think that the case for avoiding theologically motivated scientific exploration is stronger—but why? The experience of any given scientist, whether naturalist, Buddhist, Christian, Hindu, Jew, or Muslim, can lead to testable hypotheses. Some will be fruitful and some not so. No worldview has cornered the market on good or bad hypothesis generation, and in the domain of religious expression, both outsiders’ and insiders’ perspectives are valuable. Even if the hypotheses come from decidedly theological motivations, what follows? Shook seems to think data or theory distortions are inevitable. I suggest that regardless of one’s worldview, data and/or theory distortions are inevitable, which is why a community of scholars committed to honestly and transparently pursuing truth is important for scientific progress. What is the alternative? To exclude all people with theological convictions from participating in science? To psychoanalyze scientists in order to discover which of their hypotheses may have had theological origins? Such McCarthyism is either chilling or comical. Shook’s own attempt to get my theological motivations right has led him to variously suggest that I am a liberal Protestant and a deist with great sympathy for animism and polytheism. A quick look at Wikipedia will reveal that I am a fairly orthodox Evangelical Christian.
Even if our own personal biases get in the way of the best data analyses and interpretations, which they inevitably will, our scientific colleagues will call us out if we are unfair and then we can collectively move inquiry forward. Particularly in the scientific study of religion, instead of censuring or excluding religious scholars, I would rather see a diverse range of contributors who genuinely and robustly pose hypotheses from their particular viewpoints, pursue data collection and analysis using the good and broadly accepted techniques, and interpret the findings as fairly as they can muster. Indeed, one of the strengths of
Barrett, Justin L. (2011). Cognitive Science, Religion, and Theology: From Human Minds to Divine Minds. West Conshohocken, PA: Templeton Press.
Barrett, Justin L. (2007). Cognitive Science of Religion: What Is It and Why Is It? Religion Compass, 1(6), pp. 768-786. doi: 10.1111/j.1749-8171.2007.00042.x.
Bering, Jesse M. (2011). The Belief Instinct: The Psychology of Souls, Destiny, and the Meaning of Life. New York, NY: W. W. Norton & Company.
Bersaglieri, T.; Sabeti, P. C.; Patterson, N.; Vanderploeg, T.; Schaffner, S. F.; Drake, J. A.; Rhodes, M.; Reich, D. E.; Hirschhorn, J. N. (2004). Genetic Signatures of Strong Recent Positive Selection at the Lactase Gene. The American Journal of Human Genetics 74 (6), pp. 1111-1120. doi:10.1086/421051. pmc 1182075. pmid 15114531.
- Search Google Scholar
- Export Citation
( ; Bersaglieri, T. ; Sabeti, P. C. ; Patterson, N. ; Vanderploeg, T. ; Schaffner, S. F. ; Drake, J. A. ; Rhodes, M. ; Reich, D. E. Hirschhorn, J. N. ). 2004 Genetic Signatures of Strong Recent Positive Selection at the Lactase Gene. The American Journal of Human Genetics 74( 6), pp. 1111- 1120. doi:10.1086/421051.. pmc1182075. pmid15114531
Boyer, Pascal. (1994). The Naturalness of Religious Ideas: A Cognitive Theory of Religion. Berkeley, CA: University of California Press.
Clark, Kelly J., & Justin L. Barrett (2010). Reformed Epistemology and the Cognitive Science of Religion. Faith & Philosophy 27(2), pp. 174-189. doi: 10.5840/faithphil201027216.
Lawson, E. Thomas, and Robert N. McCauley. (1990). Rethinking Religion: Connecting Cognition and Culture. Cambridge, MA: Cambridge University Press.
Schloss, Jeffrey & Michael Murray (Eds.) (2009). The Believing Primate: Scientific, Philosophical, and Theological Perspectives on the Evolution of Religion. New York, NY: Oxford University Press.
Sperber, Dan, & Lawrence A. Hirschfeld. (2004). The Cognitive Foundations of Cultural Stability and Diversity. Trends in Cognitive Sciences 8 (1), pp. 40-46.
Trigg, Roger & Justin L. Barrett (eds.) (2014). The Roots of Religion: Exploring the Cognitive Science of Religion. Farnham, Surrey: Ashgate Publishing.