In 2019, Whitehouse et al. ignited a relative firestorm of controversy around the Big Gods Hypothesis of social complexification. The Big Gods Hypothesis holds that belief in moralizing big gods allowed, or perhaps even drove the significant increases in the scale and complexity of human societies that we see in many places throughout the past ten thousand years or so.
A strictly Darwinian argument on this would say that once there is a belief in gods that monitor the behaviour of people towards each other, and who enforce sociability by punishing the antisocial, larger and more complex groups can exist.
By contrast, Whitehouse et al. claimed to show in a study published in Nature, that significant increases in social complexity in our species’ recent past systematically precede the appearance of big gods, and even of most forms of moralizing supernatural punishment. They suggest that perhaps greater complexity actually drove the growth of big gods.
Beheim et al. (2021), among others, were quick to point out a few issues with Whitehouse et al. (2019), the most serious of which was that the original paper treated the absence of evidence of big gods at a certain time and place as evidence of their absence, instead of treating their status as unknown. This eventually led to the retraction of the Nature paper.
The retake
Recently, Religion, Brain & Behaviour published a “retake article” on the original study by Whitehouse et al. (2022), an expanded analysis of the data by Turchin et al (2022a), a set of comments on the expanded analysis from a range of scholars, and a response by Turchin et al (2022b). I dove fairly deeply into all this new material, and made several false starts at going over the nuances of the arguments back and forth, of specialist discussions of whether and to what degree there is or isn’t evidence of big gods at time x in place y, only to realize that we are in fact still dealing with very simple, very fundamental, and very significant disagreements about what we can know about the past, and how we should treat missing data in analyses.
There are two basic questions here: 1) Does historical and archaeological evidence of moralizing gods precede evidence of increased scale and complexity of socio-political groups? And 2) Is there a causal relationship between big gods and big societies? To answer either of those questions, we need data, and I don’t think the data exists at the moment. I don’t know if it ever will.
Whitehouse et al. (2022) explain that the error in their original paper consisted of “the mislabeling of some data on the presence of Big Gods as unknown prior to the dates of first appearance, rather than absent or inferred absent.” They claim that in most of the cases in which the presence of big gods was recorded as unknown in their original data, they actually did had evidence of the absence of big gods at a certain time in a certain place, so that big gods should have been recorded as absent or inferred absent, rather than as unknown.
They further claim that when they correct that mislabeling, their original result still holds, and that the “largest increases in social complexity do indeed precede Big Gods in world history and that Big Gods did not contribute to the evolution of sociopolitical complexity as predicted by the Big Gods hypothesis.”
The main problem remains
I have trouble with both claims. This may be a failure of creativity on my part, but I confess that I struggle to imagine what evidence of absence would consist of in the case of big gods. The threshold for reasonable evidence of absence of a religious belief or concept in the archaeological or historical record, either as genuinely personally held belief or as organizing principle in a social network, is absurdly high.
I just don’t think there is a point, at this moment, at which we can say we have evidence of the absence of big gods in the archaeological record. We can certainly have no evidence of their presence. That’s not an issue.
But unless you have, signed and in triplicate, a positive statement from some high priest that they actively reject and condemn their unenlightened neighbour’s ridiculous belief in a moralizing big god, you’re dealing with unknown, not with absent. And even then, I would still be cautious and put an asterisk on it as evidence of absence. Who knows, maybe the priest was under duress.
Having grown up in a fairly conservative old-testament big-god-fearing catholic sort of environment, and having been duly Communed and Confirmed into it, I would struggle to figure out which of my own relatives really believed what, and more importantly, to what degree it impacted or structured their behaviours, social, antisocial, or otherwise. So I very much doubt my ability to read these things into a fragmentary, degraded, and transformed archaeological record.
I’ll also mention that I just spent the better part of a year evaluating appeals for exemptions from the University’s COVID restrictions, many of them religiously based, which even further eroded my confidence in my ability to decypher material traces of beliefs and their impacts on behaviour.
At the end of the day, remember that we are not looking here for things people say or write about what they believe, or for symbols they carve in stone about how they think the divine impacts their lives. Those definitely leave traces that we can find archaeologically and historically. Rather, we are in fact looking for the ways in which their beliefs about the supernatural or the divine actually structure their societies, whether they decide to tell us or not. There’s a big difference in archaeological visibility between those two.
Where Whitehouse et al. and Turchin et al. err on the side of absent, I unapologetically err on the side of unknown. They even specify that in order to address this criticism, they instituted in their project a practice according to which investigators had the opportunity to argue for a coding of “unknown” to be applied for a particular case, indicating unambiguously that they favour absent as a default where there is no positive evidence of presence.
Conversely, in my view, the onus should strongly be on those wanting a coding of “absent” to prove their case. “Unknown” should be the default state when there is no positive evidence of presence, and positive evidence of absence should be required for a declaration of no big god.
In archaeology, we learn from the material we find, not from the material we don’t find. When we find what we’re expecting, we don’t learn much, because it is supposed to be there. When we find what we’re not expecting, we learn a lot, because there must be a reason it’s there, and we clearly don’t know what that reason is. When we don’t find what we’re expecting, we learn little, because what we’re looking for might have been there at some point, but perhaps isn’t anymore. When we don’t find what we’re not expecting, we learn nothing much of value.
In their revised analysis, Whitehouse et al. address the question of absent vs unknown by forming a committee of experts to review cases. Their experts are definitely bona fide experts in their areas and widely respected. That’s fine, but it doesn’t really matter.
I want to emphasize that I fully respect their expertise, but even that is not enough to get me to soften my stance on the very serious problem of evidence of absence in the archaeological record, especially when dealing with archaeological remains of beliefs.
The method used by the committee of experts who decide on absence, is still basically to have the best possible overall grasp of the available information, and to render an impressionistic judgment. Which is fine for what it is, but not very helpful for the sort of cut and dried quantitative result the authors are after. They’re still doing a basic lag analysis, and they still base it on presence/absence of features in the system.
If you don’t find a cache of curated obsidian tools at the bottom of a storage pit in a sealed stratigraphic context, you might be getting closer to evidence of absence (but only slightly). If you’re not finding evidence of a religious belief, you are faced with absence of evidence.
Even if I grant the assertion that Whitehouse et al. do have positive evidence of the absence of big gods in the past, I still have to ask myself: What would become archaeologically and historically visible first? Evidence of increased scale and complexity of a society? Or evidence of a belief in moralizing supernatural punishment?
It can take a lifetime, or even several generations, to clearly formulate a concept that is active in our lives and affects us. How long did it take ancient Egyptian social and religious theorists to formulate the concept of Maat (order), which Whitehouse et al. identify as the first appearance of big gods in their sample? How long had Maat been a defining feature of ancient Egyptian society before someone named and recorded it? How long did it take after Marius for Roman citizens to realize they weren’t living in a functional republic anymore? We simply don’t know, and I would argue we have no way of finding out with any degree of certainty.
Urban agglomerations and large irrigation projects are more difficult to hide than religious beliefs. Big societies are much more archaeologically visible than big gods. Even if Whitehouse et al. and Turchin et al. can somehow read the absence of big gods in the past, I would therefore not rely on a determination of which came first, especially not one that is based on first appearance in the archaeological or historical record, because existence in a society, does not necessarily and automatically mean appearance in it’s record in the form in which that record is available to us in the present.
Which, interestingly, is sort of where the expanded analysis by Turchin et al. ends up: “the strong correlation between sociopolitical complexity and moralizing religion is a result of shared evolutionary drivers, including intense military competition aided by increasing agricultural productivity.”
Despite its professed defense of the original result, the expanded analysis by Turchin et al is much more nuanced, much more cautious, much more ambivalent about what comes first and what drives what, and comes to more plausible conclusions.
Predictably then, the expanded study didn’t make any headlines. In the end, behind all the sturm und drang of claim, counterclaim, news coverage and retraction, that difference in headline-making potential between the original and the revised studies may be the real problem in this whole saga of the moralizing gods.
References
Beheim, B., Atkinson, Q. D., Bulbulia, J., Gervais, W., Gray, R. D., Henrich, J., Lang, M., Monroe, Willis M., Muthukrishna, M., Norenzayan, A., Purzycki, B. G., Shariff, A., Slingerland, E., Spicer, R., & Willard, A. K. (2021). Treatment of missing data determined conclusions regarding moralizing gods. Nature, 595(7866), E29–E34. https://doi.org/10.1038/s41586-021-03655-4
Peter Turchin, Harvey Whitehouse, Jennifer Larson, Enrico Cioni, Jenny Reddish, Daniel Hoyer, Patrick E. Savage, R. Alan Covey, John Baines, Mark Altaweel, Eugene Anderson, Peter Bol, Eva Brandl, David M. Carballo, Gary Feinman, Andrey Korotayev, Nikolay Kradin, Jill D. Levine, Selin E. Nugent, Andrea Squitieri, Vesna Wallace & Pieter François (2022a) Explaining the rise of moralizing religions: a test of competing hypotheses using the Seshat Databank, Religion, Brain & Behavior, DOI: 10.1080/2153599X.2022.2065345
Peter Turchin, Harvey Whitehouse, Jennifer Larson, Enrico Cioni, Jenny Reddish, Daniel Hoyer, Patrick E. Savage, R. Alan Covey, John Baines, Mark Altaweel, Eugene Anderson, Peter Bol, Eva Brandl, David M. Carballo, Gary Feinman, Andrey Korotayev, Nikolay Kradin, Jill D. Levine, Selin E. Nugent, Andrea Squitieri, Vesna Wallace & Pieter François (2022b) Big Gods and big science: further reflections on theory, data, and analysis, Religion, Brain & Behavior, DOI: 10.1080/2153599X.2022.2065354
Harvey Whitehouse, Pieter François, Patrick E. Savage, Daniel Hoyer, Kevin C. Feeney, Enrico Cioni, Rosalind Purcell, Jennifer Larson, John Baines, Barend ter Haar, Alan Covey & Peter Turchin (2022) Testing the Big Gods hypothesis with global historical data: a review and “retake”, Religion, Brain & Behavior, DOI: 10.1080/2153599X.2022.2074085
Whitehouse, H., François, P., Savage, P. E., Currie, T. E., Feeney, K. C., Cioni, E., Purcell, R., Ross, R. M., Larson, J., Baines, J., Haar, B. t., Covey, A., & Turchin, P. (2019). Retracted Article: Complex societies precede moralizing gods throughout world history. Nature, 568(7751), 226–229. https://doi.org/10.1038/s41586-019-1043-4
ArcheoThoughts 
There are two basic questions here: 1) Does historical and archaeological evidence of moralizing gods precede evidence of increased scale and complexity of socio-political groups? And 2) Is there a causal relationship between big gods and big societies? To answer either of those questions, we need data, and I don’t think the data exists at the moment. I don’t know if it ever will.
…
I have trouble with both claims. This may be a failure of creativity on my part, but I confess that I struggle to imagine what evidence of absence would consist of in the case of big gods. The threshold for reasonable evidence of absence of a religious belief or concept in the archaeological or historical record, either as genuinely personally held belief or as organizing principle in a social network, is absurdly high.
This is such an important point to stress, and I am reminded of the long and venerable line of scholars who have grappled with the question of how exactly the polytheism of ancient Judea, for example, with a pantheon of only a handful of, it is believed, mostly Canaanite deities, gradually transformed into henotheism and then into outright monotheism. We know that Yahweh, the storm god, and El, the creator god become (theistically and ritualistically) one and the same at some pre-Exile point, with the vaguely more ethno-culturally Israelite (as opposed to Canaanite) Yahweh emerging “victorious”. Is this a “big God” triumphing? Not exactly, it seems.
Other traditional (Canaanite) deities like Asherah and Baal remained “in use” and only gradually began “losing out” for reasons not entirely clear (though scholars have launched some interesting conjectures) with Asherah becoming only a ritualistic prop by the time of the late monarchy (10th c. BC and after the dissolution of the united kingdom) – in the form of the “asherah pole”- while the latter became – in the Hebrew Biblical tradition – Yahweh’s quintessential adversary; a “demon” requiring child sacrifices and worshipped only by “bad” kings such as Ahab or “weak” ones like the aged and “henpecked” Solomon. Yet even this did not arguably mark the start of a classic “moralizing big god” as taken for granted by today’s Abrahamic religions.
How did this process change the nature of Yahweh himself then? How, in other words, did Yahweh become God? For one, it seems, he ceased being a mere weather god and assumed (why?) the cosmogonic omnipotence of the now absent El. He also ceased being an uncomplicatedly anthropomorphic tribal deity requiring violent (and sometimes human) sacrifice while glorying in the death of “his” enemies (more on this point below). Instead Yahweh became the “jealous” but ultimately just judge passing swift and fair retribution or reward without any “dionysian” excesses. Lastly and perhaps most importantly, Yahweh ceased being a mortal (if renewable) and sexual god (characteristics associated with the “demon” Baal) among other gods. He became instead a sexless, timeless, deathless, and peerless “inconceivable” mystery transcending and yet continuously adjusting human reality. It is of course not at all clear how exactly this metamorphosis occurred or over what specific time span, but Mark Smith, for example, argues (compellingly enough, I find) that the priestly caste played a likely decisive role:
“Given the priestly insistence on the impurity of death and sexual relations, it is difficult to resist the suggestion that the presentation of Yahweh generally as sexless and unrelated to the realm of death was produced precisely by a priesthood whose central notions of holiness involved separation from the realms of impurity, specifically sexual relations and death. For the priesthood there were several levels of cultic purity, and the deity represented the epitome of this hierarchy.” (Smith, The Early History of God)
And this segues back into the “issue” of an ascendant, if not quite yet unrivalled, Yahweh requiring violent (and sometimes human) sacrifice while glorying in the death of “his” enemies–is this a “moralizing big god”? On the surface, with a hindsight morphed by thousands of years of “orthodox” and ever-increasingly-moral(izing) Abrahamic religion, the answer is maybe more of a no. A god, even a “big god” that glories in gore and sacrifice, particularly gore that isn’t always meant to achieve some clear social purpose (ensuring the harvest, punishing impurity etc), has seemingly not achieved “full morality”. But even leaving aside the vast collection of similarly bloody gods of other ancient (and undisputed) civilisations (Aztecs, Phoenicians), perhaps things aren’t that clear cut even for Abrahamic religion (which, in its Christian iteration, has long been touted as the most morally refined, leading to the vast and still rippling mess that we find ourselves in, in Canada as much as elsewhere where colonialism has long been society’s main driving force). And here, even though I like how your post keeps things anchored in the realm of measurable (or at least identifiable) evidence, I feel like the cultural and philosophical criticism of someone like Nietzsche, for example, could come in useful. As he put it in his Genealogy of Morals, among other works, the good is very often a triumphant incarnation of what had previously been considered bad and despicable. And vice versa of course. Hence the (much misread and misinterpreted) concept of “slave morality”, i.e. the morality that a ruling class perceives as proper to literal slaves/subalterns becoming, through cultural and social tectonic shifts, the morality of a new ruling class, not always entirely divorced from the old one (socially, ethnically), but with a very different ideology to justify its stronghold on society. Morality is very, very fluid and not at all a linear progression from “primitive/immoral” to “civilization/moral”. So not only is actual evidence lacking, but perhaps the very question “when was the time when moralizing big gods did not yet exist” is so loaded as to defy all sense.
I agree with your very measured and principled “agnosticism” (if I may call it that) on this question.
LikeLike
Thanks. Very interesting discussion. And you may certainly call it my agnosticism, That’s exactly what it is.
LikeLike
I agree that its critical thinking 101 that agnosticism is the default response to all claims and not something that needs defending. And by definition, beliefs do not leave a distinctive archaeological trace.
The better documented a society is, the more variety we see in its religious discourse and practices. One person can believe there are a world full of gods, another can declare that they are all aspects of the Lord, but both can build the same sacred architecture or attend the same feast and throw the bones and cups into a midden for archaeologists to excavate.
LikeLike
Precisely.
LikeLike