Post-Hoc Supernatural Punishers

In the inaugural issue of Religion, Brain & Behavior, Jeffrey Schloss and Michael Murray examine the idea that belief in supernatural agents is adaptive because these agents are punishers: supernatural policeman if you will. This policing can have two effects. First, belief in supernatural punishment can enhance within group cooperation. Second, it can reduce cheating or free-riding. The former is characterized as “cooperation enhancement” or CE and the latter as “punishment avoidance” or PA. Schloss and Murray then ask what fitness-relevant feature of the ancestral environment might have selected for CE and PA.

Supernatural punishment theory is anchored in two sources: lab research and game theory. Although it frequently references ancestral environments, these environments are rarely if ever specified. When and where did humans begin believing in moralizing and punishing supernatural agents?

There are some hints. In their comment to Schloss and Murray’s target article, Aguair and Cronk observe that judgmental gods or policing spirits are historically recent: “Considerable evidence exists that such beliefs are rare among hunter-gatherer, smallscale, and egalitarian societies, and common among food producing, large-scale, and hierarchical societies.” Azim Shariff similarly comments: “If beliefs in omniscient and punitive gods were genetic adaptations rooted in our Pleistocene past, we would expect these beliefs to be psychological universals, or, at the very least, more prevalent in hunter-gatherer societies. Neither is true.”

In several of its guises, supernatural punishment theory is genealogical. It attempts, in other words, to explain what is known today as “religion” and account for it in evolutionary terms. While this is a perfectly reasonable endeavor, methodological care must be taken: “a purpose or point now found to be characteristic of morality and its institutions must not be uncritically read back into history as providing the key to understanding its origin” (Brandhorst 2010:24).

Friedrich Nietzsche, a genealogist of considerable skill, was sharply critical of ahistorical functionalism and utilitarian essentialism:

How have the moral genealogists reacted so far in this matter? Naïvely, as is their wont: they highlight some “purpose” in punishment, for example, revenge or deterrence, then innocently place the purpose at the start, as causa fiendi of punishment, and have finished.

[T]he origin of the emergence of a thing and its ultimate usefulness, its practical application and incorporation into a system of ends, are toto coelo separate.

No matter how perfectly you have understood the usefulness of any physiological organ (or legal institution, social custom, political usage, art form or religious rite) you have not thereby grasped how it emerged: uncomfortable and unpleasant as this may sound to more elderly ears,—for people down the ages have believed that the obvious purpose of a thing, its utility, form and shape are its reason for existence, the eye is made to see, the hand to grasp. — Genealogy of Morals (II:12)

We cannot simply assume that because supernatural watchers-punishers exist and have utility in post-Neolithic or complex societies, the same was true during the Paleolithic. Although the ethnographic and ethnohistoric hunter-gatherer record is an imperfect guide, it strongly suggests that supernatural watching-punishing is a recent invention.


Schloss, Jeffrey P., & Murray, Michael J. (2011). Evolutionary Accounts of Belief in Supernatural Punishment: A Critical Review. Religion, Brain & Behavior, 1 (1), 46-99 : 10.1080/2153599X.2011.558707

Brandhorst, Mario (2010). Naturalism and the Genealogy of Moral Institutions. The Journal of Nietzsche Studies, 40, 5-28

Did you like this? Share it:

4 thoughts on “Post-Hoc Supernatural Punishers

  1. Stewart Guthrie

    Excellent critique–demolition, really–of a cornerstone by the social-cohesionist camp in CogSciRel. Your Nietzsche excerpts on functionalism’s teleology should be regarded as the end of that story.

    Can’t help adding that functionalist teleology is another aspect of anthropomorphism. As Bacon pointed out, nature doesn’t have purposes; people do.

  2. Cris Post author

    Nietzsche is always salutary on matters of methodology. I have always been bothered by the large number of CSR theorists who simply project their functional findings back into prehistory as causes.

  3. Connor Wood

    Chris, under Nietzsche’s logic it would be hard to identify any evolved cultural or biological phenotype as an adaptation, wouldn’t it? For example, even eyes aren’t adaptations in an originalist sense. If the question of a phenotype’s origin is bound up with the question of adaptation, then the etiological story is always going to be non-adaptationist. Either a feature starts out as a byproduct of another phenotype, or it starts as a random mutation. So in assuming that the origin of phenomena like religious punishment must be far back in some timeless EEA far in our ancestral past in order to qualify as an adaptation, you’re rigging the game in your favor – push the origin far enough back, and nothing is properly an adaptation. But evolution didn’t stop when people started putting crops in the ground; any heritable cultural or biological feature that arose since the Iron Age could have altered the conditions of its host evolutionary niche to the point of feeding back on itself and becoming the target of selective pressure in its own right.

    I say all this despite being unimpressed with supernatural punishment arguments, which I think are often contradictory and conceptually muddled. This is more a general point about critiques of adaptationist frameworks.

Leave a Reply