Some thoughts on supernatural agency beliefs

I am pleased to have been given by the ICCI the opportunity to advertise two notes here, each about four pages long, that I have recently written and put on SSRN. Both try to shed light on supernatural agency beliefs. The abstract to the first one reads:

Should we account for belief in supernatural agents in terms of benefits it might provide to the human believers? Or is it just a by-product of the human cognitive architecture? Perhaps neither. A different perspective, no longer human-centric, shines through in an observation like “admittedly, the Argument from Design must have been quite convincing before Darwin”. We can go farther in this direction.

Full text can be found here. This is the abstract to the second piece:`

Can the root of animism be illustrated as follows? The sun, cliffs, fountains, trees, etc; and in our time the planet Earth as seen in the darkness of space: they have all lasted a long time despite perceived threats to their existence, and are thus attributed a will to survive.

Full text here. What follows now is a shortened and adapted version of the first note.

In their 2009 BBS target paper, The evolution of misbelief, Ryan McKay and Daniel Dennett investigate a contentious subclass of human misbeliefs, namely (I quote from the abstract) “misbeliefs best conceived as design features [of the belief formation system]. Such misbeliefs […] would have been systematically adaptive in the evolutionary past.” There is an acknowledgement to the “Explaining Religion” collaborative project, so one motivation must have been supernatural agency misbelief. Accordingly, section 11 of the paper deals with this case and reviews the literature. After discussing a “supernatural punishment hypothesis” that indeed regards certain supernatural belief as a design feature, they conclude “the currently dominant evolutionary perspective on religion remains a by-product perspective”.

Design feature or by-product? The question pervades modern scientific assessments of supernatural agency beliefs. But perhaps the key to a different perspective can be found in David Dunning’s commentary: “Focusing on the individual’s internal cognitive architecture, McKay & Dennett (M&D) provide an incomplete analysis [though he also calls it thoughtful and stimulating] because they neglect the crucial role played by the external environment in producing misbeliefs [emphasis mine] and determining whether those misbeliefs are adaptive.” And: “Even a perfectly rational human organism could come to hold the types of misbeliefs that M&D discuss, because the environment much more frequently provides people with incomplete or misleading data than M&D anticipate.” Dunning does not say anything specific about the case of supernatural agency, using other examples instead, but below I will take up his line and try to blame (clear mis-)belief in certain supernatural agents, namely rain gods and similar deities, on a misleading environment.

How is such a perspective different? Puzzled by a perceived “bug” in the human belief-formation system, we are told by design-feature theories “it’s not actually a bug, it’s a feature!”, whereas by-product theories usually try to show us how the bug makes sense within a larger scheme. By contrast, the alternative perspective denies the need to grapple with an internal bug. Instead, the philosophical root of the puzzling situation is a misleading external world.

For what it’s worth, let me add a piece of personal experience here. Once my laptop’s battery would not charge any more. So I got a brand new one. However, that one wouldn’t charge either! What was broken was apparently the charging facility of the laptop. I was quite upset with myself — I may not know much about tech stuff, but still, how could I not have considered this possibility beforehand? Now, you could say that was simply stupid, and perhaps we should indeed stop the analysis right there, but we could also ask: is overlooking environmental context when assigning blame somehow an easy mistake to make? Plus, it would become even easier if there was a revealing new angle available from which to recognise and examine certain idiosyncrasies in laptop batteries. (The new angle is meant to refer to evolutionary psychology, of course — but I would then have examined the battery rather than just replaced it — well, I guess the whole analogy is quite a strained one anyway.)

The original note on which this post is based says nothing about laptop batteries, but it does in turn include a bit more on the following three works, to be just briefly mentioned here:

• Stewart Guthrie’s 1993 book, Faces in the Clouds: A new theory of religion [1]. This well-known by-product theory can be illustrated by a sentence (p. 6) that McKay & Dennett also quote: “it is better for a hiker to mistake a boulder for a bear than to mistake a bear for a boulder”.

• Robin Horton’s 1993 book, Patterns of thought in Africa and the West [2], a collection of some of his major essays over the years. Horton employs the perspective I am advocating: in the introduction, on p. 13, he characterises his approach as contextualist as much as (what it is better-known for) intellectualist. Actually, regarding Horton, I would like to use the occasion to ask readers an additional question. He sees substantial similarities between the thought of almost all religions and that of science. In the postscript he calls it the “Similarity Thesis”. Among the obstacles to its acceptance in his discipline, social anthropology, he notes two that stem from non-scientific motivations: a scrupulous refusal to impute mistakes to non-Westerners (thus scholars may deny that people take their rain gods for real), and a romantic determination to find in traditional cultures what has supposedly been lost in the science-dominated West. But 20 years on, we have the new cognitive science of religion, and I don’t get the impression that it suffers from such obstacles. So has Horton’s thesis swept the field by now, then? Not at all. Newer writers don’t even seem to notice him much. I have studied Horton’s book closely earlier this year and find it brilliant, so I wonder why this is the case — and might even make my own attempts, as best I can, to answer objections to Horton in the comments.

• Bernard Fontenelle’s late-17th-century essay, De l’origine des fables. Despite what I just said, I would disagree slightly with Horton, at least in emphasis, on the reason most cultures use what he calls the “personal idiom” — gods and spirits — in their theories. My own take on it will follow right below. Fontenelle, however, gets the emphasis right in my opinion; and otherwise his account, while less elaborate (he devotes only about one page to this), is remarkably similar to Horton’s: gods are explanations of the world by analogy, the results of a “principle so natural that even today our philosophy has none other” (English translation, quoted by Guthrie on p. 21). It would be quite astonishing if it were to turn out that the possibly oldest treatment of the topic hits the mark best, notwithstanding all the later discussions through the centuries.

So where does the personal idiom come from?

(The rest of this post is copied verbatim, except for a few omissions, from the original note.)

Dunning calls the environment out on the “incomplete or misleading data” it provides. I would put it in our case as follows. The agentive interpretation, attributing desires and other mental states, is clearly apt for a lot of events, namely those caused by humans and animals; and one might not even entertain the thought that the environment would mix in other events, such as rain and wind, where this interpretation just can’t work well any more — however hard one may try to get the mental states right. No desires at all are involved in the weather’s actions, but this is far less clear than that desires are involved when humans and animals act, at least for someone without a specific education or cultural context.

For an analogy, let’s replace the ability to cause events with something else typical of humans and animals: the tendency to leave footprints. Imagine a world where sometimes, without any agent involvement, shapes appear in the soil that look like footprints (although perhaps much bigger). People inhabiting such a world could surely be forgiven for mistakenly assuming these lookalikes to be footprints, left by agents not unlike themselves or animals.

They might see through the hoax if given a chance to directly observe a track of footprint-lookalikes coming into existence without anyone in sight. Or they might then instead perceive an invisible agent walking there. But notice that, in our original case in our own world, supposed agents causing weather events need not be invisible outright. A rain god could be merely hidden from view, or too far away to see.

One might criticise the footprints analogy for relying on one specific sort of shape, out of many possible, when the case at hand involves all sorts of events rather than just a specific sort. Again, then, let’s replace the ability to cause events with something else typical of humans and (less so this time) animals: the ability to create design. All sorts of design, so the criticism doesn’t apply here. Imagine a world where sometimes, without any agent involvement, appears what looks like design … but this is now reminiscent of our own world! Think of eyes, or birds’ wings. The world exhibits apparent design not created by humans or animals; and unsurprisingly, beliefs in “intelligent design”, in some form or other, have been widespread at all times.

I should note that Guthrie, on pp. 186–187, invokes his account of anthropomorphism to help explain the human feeling of a designer behind the works of nature, citing Hume on the need for such help. But otherwise it would not have occurred to me that an intellectualist approach might be insufficient to account for the beliefs. Provided, of course, that Darwin’s notion of design by natural selection is not available as a viable alternative to design by agent. Likewise, in our original case the argument depends on there not being any notion of natural event-causation. That is, on Nietzsche being correct when stating “Now man believed originally that wherever he saw something happen, a will had to be at work in the background as a cause, and a personal, willing being. Any notion of mechanics was far from his mind.”

I read this, originally from The Gay Science, when it appeared last year as the first part of a larger quotation in a post by Cris Campbell on Nietzsche and theory of mind. We should be alert for counterexamples, though. Have certain happenings always been seen as unconnected to agency? That would present a challenge to our parsimonious approach. It would show that people had the option, after all, to cut agency out of the picture, as in today’s mechanistic view of geological or weather events; and yet they didn’t exercise it as often as they “should” have done. Why not? At this point, to answer that, it may make more sense to invoke, as Guthrie does, a human cognitive bias or feature.

But do we have to answer it? Does such a counterexample exist? I conclude the post with two caveats regarding what would qualify. First, while I have singled out desires among mental states (cf. Dennett’s concept of intentional stance, singling out beliefs and desires), it is not enough for an event to be not thought of as desired by an agent to happen. Agents may cause events not only on purpose, but also out of carelessness. And as side effects. When there is a new footprint, it is clear that its creation must have been a small event when it happened. But everyone knows, presumably, that most likely this event wasn’t desired by the agent who caused it.

Secondly, of course all people have concepts of physics. For example, everyone knows gravity. However, even if the downward trajectory of an object is seen as unfolding by the necessity of physics, the perceived cause of the event could still be a (supposed) agent who dropped or hurled the object in the first place. What if it was pushed off a cliff by the impact of another object? That would be physics as well. Yet who set the other object in motion, then? Only examples qualify that lack any perceived contribution from agents with desires and other mental states.

---

[1] Guthrie, S. E. (1995). Faces in the clouds: A new theory of religion. Oxford University Press.

[2] Horton, R. (1997). Patterns of thought in Africa and the West: Essays on magic, religion and science. Cambridge University Press.