Content uploaded by Benjamin Grant Purzycki
Author content
All content in this area was uploaded by Benjamin Grant Purzycki on Oct 28, 2021
Content may be subject to copyright.
Appealing to the minds of gods: A novel cultural evolutionary
account of religious appeals and an empirical assessment using
ethnographic data from eight diverse societies
Theiss Bendixena,∗
, Coren Apicellab, Quentin Atkinsonc, Emma Cohend,e, Joseph
Henrichf, Rita A. McNamarag, Ara Norenzayanh, Aiyana K. Willardi, Dimitris
Xygalatasj, Benjamin Grant Purzyckia,∗
aDepartment of the Study of Religion, Aarhus University, DK
bDepartment of Psychology, University of Pennsylvania, USA
cDepartment of Psychology, University of Auckland, NZ
dInstitute of Cognitive and Evolutionary Anthropology, University of Oxford, UK
eWadham College, University of Oxford, UK
fHarvard University, USA
gSchool of Psychology, Victoria University of Wellington, NZ
hUniversity of British Columbia, CA
iDepartment of Psychology, The University of Texas at Austin, USA
jDepartment of Anthropology, University of Connecticut, USA
Abstract
While appeals to gods and spirits are ubiquitous throughout human societies past
and present, deities’ postulated concerns vary across populations. How does the
content of beliefs about and appeals to gods vary across groups, and what accounts
for this variation? With particular emphasis on locally important deities, we develop
a novel cultural evolutionary account that includes a set of predictive criteria for
what deities will be associated with in various socioecological contexts. We then
apply these criteria in an analysis of individual-level ethnographic free-list data on
what pleases and angers locally relevant deities from eight diverse societies. We
conclude with a discussion of how alternative approaches to cross-cultural variation
in god beliefs and appeals fare against our findings and close by considering some key
implications of our methods and findings for the cognitive and evolutionary study of
religion.
Keywords: human behavioral ecology, cognitive anthropology, cultural evolution,
∗Corresponding authors
Email addresses: tb@cas.au.dk (Theiss Bendixen), bgpurzycki@cas.au.dk (Benjamin Grant
Purzycki)
Preprint submitted to Religion, Brain and Behavior October 26, 2021
free-list method, gods’ minds, religious systems
1. Introduction
Various approaches address why humans are the only species on the planet
that entertains beliefs about gods, ghosts, and other spiritual beings (Geertz,2020;
Jensen,2019). Some emphasize evolved cognitive faculties such as agency detection,
anthropomorphism, and other mentalizing systems to account for beliefs in gods
(e.g., Andersen,2019;Bird-David,1999;Barrett,2000;Guthrie,1980,1995). Others
argue that gods are attention-grabbing and more memorable by virtue of their social
relevance inferred by our evolved moral cognition (Boyer,2000,2001) or suggest that
aspects of gods’ temperaments are projections of individuals’ personal attitudes and
characteristics (Johnson et al.,2015;Spiro and D’Andrade,1958). While many of
these approaches view beliefs as by-products of human cognition, others posit that
spiritual sanctions can minimize defection in cooperative dilemmas (Johnson,2005,
2016;Schloss and Murray,2011) and/or expand cooperation beyond kith and kin
(Norenzayan et al.,2016).
Two critical questions interwoven throughout these discussions are: How does the
content of beliefs about and appeals to gods vary across groups and what accounts for
this variation? While these questions have deep roots in the history of anthropolog-
ical thought (e.g., Evans-Pritchard,1965;Lang,1909;Swanson,1960;Tylor,1920),
the bulk of contemporary research mostly examines gods or religious traditions that
are explicitly interested in human morality (Baumard et al.,2015;Beheim et al.,
2021;Botero et al.,2014;Peoples and Marlowe,2012;Roes and Raymond,2003;
Skoggard et al.,2020;Snarey,1996;Watts et al.,2015). These studies primarily
rely on—often the same—society-level data, usually from coded ethnographies and
reports from travelers and missionaries (see Purzycki and Watts,2018).
When studies have examined individual-level beliefs, they have mostly focused
on moralistic gods’ role in cooperation (e.g., Atkinson and Bourrat,2011;Ge et al.,
2019;Lang et al.,2019;Purzycki et al.,2016b;McNamara and Henrich,2018;Willard
et al.,2020) rather than attended to deities concerned with other matters and their
ethnographic contexts. Yet, if religion can contribute to the evolution of cooper-
ation, we should expect that variation in religious appeals, beliefs, and practices
is partly attributable to variation in local threats to coordination and cooperation
(Bendixen and Purzycki,2020;Purzycki and McNamara,2016;Purzycki and Sosis,
2022). However, despite isolated exceptions (e.g., Atran et al.,2002;McNamara
et al.,2021;Purzycki,2011,2013,2016;Singh et al.,2021), there remains a dearth
of high-resolution, directly comparable, cross-cultural data with which to examine
2
how gods’ concerns systematically vary cross-culturally.
To initiate this inquiry, we present and examine individual-level ethnographic
free-list data on what pleases and angers deities across eight diverse field sites. In
Section 2, we first review contemporary work on cross-cultural variation in gods’ con-
cerns. Building on this, we develop a cultural evolutionary account of that variation
and outline a set of predictive criteria for what gods will evolve to care about given
local socioecological features and constraints (cf., Bendixen and Purzycki,2020).
We then present methods of our data collection and summarize key results (Section
3). In Section 4, we assess how our free-list data informs our cultural evolutionary
account. Finally, we discuss how alternative approaches fall short of accounting for
the present findings and consider important implications and cautionary notes of the
present work (Section 5).
Overall, we regard our contribution as taking the necessary theoretical and empir-
ical first steps toward a systematic and predictive account of cross-cultural variation
in beliefs about and appeals to gods and spirits. Given that these are first steps,
we hope to spark constructive discussion about i) our cultural evolutionary account
and its constituent elements; ii) the free-list data, its strengths, limitations, and
interpretation; iii) our site-by-site synthesis of data and ethnographic contexts; and
iv) the broader implications of our methods and findings for the cognitive and evo-
lutionary study of religion, including the cognitive and cultural processes that give
rise to variation and universals in supernatural appeals.
2. Social life and the minds of gods: A cultural evolutionary account
2.1. Variation in god beliefs and appeals
What do gods care about and why? While the subject of gods’ concerns has
been central to anthropological inquiry for over a century, its systematic empiri-
cal assessment has only recently begun. For example, Boehm (2008) surveyed 43
ethnographies covering 18 foraging societies, and found instances of supernatural
punishment of at least one behavior construed as “antisocial” and “predatory on
fellow band members” among all 18 groups. Other behaviors for which traditional
gods punished were violations of what Boehm calls “nonmoral taboos” which include
domains such as food, ritual, animals, sex, and life stages. In a general survey of
ethnographic cases, Purzycki and McNamara (2016) broke down gods’ concerns into
three broad categories: things people do or are implied to be directed toward other
people (e.g., moral conduct and virtuous qualities), toward the gods (e.g., ritual and
faith), and toward nature (e.g., preservation and maintenance). Recent site-specific
studies using contemporary social scientific methods corroborate these typologies
3
and also contextualize their relevance to interpersonal relationships (see below and
Atran et al.,2002;McNamara et al.,2021;Purzycki,2011,2013,2016;Shaver et al.,
2017;Singh et al.,2021).
As ongoing theory-building suggests (cf. Purzycki and McNamara,2016;Bendixen
and Purzycki,2020;Purzycki and Sosis,2011), these classes of concerns point to some
important roles that gods might play in local social ecologies. Yet, an encompass-
ing theory of gods’ concerns has yet to be assessed directly alongside systematically
collected, detailed, individual-level ethnographic data. Assuming that cultural tradi-
tions are contingent on cultural evolutionary historical processes (for recent survey,
see Kendal et al.,2018), we build upon these previous efforts to derive a constel-
lation of general predictive criteria for when we might expect gods to care about
particular behaviors. In doing so, we offer an account of what gods care about and
why they care about the things they do, thus allowing us to make predictions about
how religious traditions might continue to evolve (see Section 5).
2.2. Cultural evolution of gods’ concerns
Social life is replete with promises and perils, and the origin of cooperation and
coordination on human scales is often touted an evolutionary puzzle (Cronk and
Leech,2012;Henrich and Muthukrishna,2021;Richerson et al.,2016). Many evo-
lutionary mechanisms have been proposed for curbing selfishness and promoting
cooperation in human societies (e.g., Henrich and Henrich,2007), and among them
are supernatural punishment (Johnson,2005) and rituals (Sosis and Bressler,2003).
Indeed, as the work discussed in the previous section suggests, rather than being
projections of individual cognition, mundane desires or interests, cultural models of
gods’ concerns appear to revolve around behaviors that correspond to locally salient
threats to cooperation and coordination (Bendixen and Purzycki,2020;Purzycki
and Sosis,2011,2022;Rossano,2007).
Our account posits that, as individual acts, appeals to a watchful and puni-
tive deity are manipulative acts from a signaler to a receiver (see Cronk,1994a,b;
Rappaport,1994). These appeals include individually costly behaviors (e.g., “the
spirits get upset if you hunt in that part of forest” or “the gods demand expen-
sive sacrifices”) that—through threat of spiritual repercussions—can contribute to
the reduction of defection in cooperative ventures when enacted (Johnson,2016).
Supernatural appeals are often invoked when local challenges to cooperation and co-
ordination become salient and pressing (Purzycki et al.,2020) and as explanations of
maladies and misfortune (Boyer,2021;Fitouchi and Singh,2021). Given this, with
time cultural models of gods’ concerns should thus evolve to align themselves with
particular kinds of communal problems—and corresponding behaviors that might
4
mitigate these—that people face or have faced in the past. We refer to such prob-
lems as “god-problems” and predict that they have generally recurring features (cf.,
Bendixen and Purzycki,2020).
We predict that “god-problems” are first and foremost (a) social dilemmas that
are game-theoretic in nature, including problems with cooperation, coordination and
conflicts of interests (Atran et al.,2002;Lansing et al.,2017;Purzycki and Sosis,
2022;Shariff et al.,2014). Coordination and cooperation are critical aspects of
human social life, and concepts of deities that care about locally relevant issues may
be relatively more culturally and cognitively attractive within a community (Boyer,
2000,2001;Purzycki and Sosis,2022). As a result, god-problems are therefore also
predicted to be (b) materially and/or socially costly, and therefore (c) cognitively
salient.
Further, we expect god-problems to constitute a subset of local social dilemmas
that are (d) relatively difficult to police with secular means and/or (e) more con-
vincingly enforced by appeals to supernatural monitoring and punishment (Johnson,
2016;Norenzayan et al.,2016;Rossano,2007;Rossano and LeBlanc,2017). Although
empirical findings have been mixed overall, some studies have indicated that, at least
under certain circumstances, cues of being watched may curb rule-breaking (Bateson
et al.,2006;Dear et al.,2019;Nettle et al.,2013;Piazza et al.,2011;Northover et al.,
2017b,a). Appeals to supernatural agents may function as cues of being watched and
induce fear of supernatural sanctions under conditions where secular means and in-
stitutions are inefficient or unavailable (e.g., Endicott and Endicott,2014;Leeson
and Suarez,2015;Rossano,2010, p. 205-207) or where strict norm adherence is par-
ticularly critical for a community (Jackson et al.,2020;Roes and Raymond,2003;
Skoggard et al.,2020;Snarey,1996). Such appeals are likely particularly cognitive
and culturally attractive when a wide variety of frequently occurring maladies can
be interpreted as supernatural sanctions, a consistent pattern across ethnographic
reports (Hartberg et al.,2016). Lastly, since appeals to a deity might not be nec-
essary, effective, or convincing if the consequences of a behavior are transparent
and self-evident, god-problems likely have opaque pay-off structures (f) in that the
implications of widespread defection are non-obvious to individuals.
2.3. Illustration with two ethnographic examples
To illustrate these predictive criteria, let us compare two ethnographic case stud-
ies. First, consider that in the Tyva Republic localized spiritual entities known as
spirit-masters (cher eezi) are perceived to be particularly concerned with pollution,
littering and over-exploitation of natural resources (Purzycki,2011,2016, see also
below and Supplementary Section S5.1). Tradtionally, there is a widespread associa-
5
tion between spirit-masters and natural resource preservation and management (e.g.,
keeping rivers, land and sacred places clean; preventing excessive hunting). Spirits
are typically associated with areas rich in natural resources and specific sacred ar-
eas where spirit-masters reside and rituals are performed. Littering, pollution and,
more broadly, resource preservation and management, constitute a set of quintessen-
tial game-theoretic dilemmas (Colyvan et al.,2011;Hardin,1968): the collective
as a whole benefits maximally when everyone cooperates (e.g., no littering, no over-
exploitation) but individuals themselves are better off defecting on cooperation (e.g.,
littering wherever, over-exploiting public/common natural resources), when every-
one else cooperates. This fulfills criterion (a). It is also a costly dilemma, criterion
(b), in that everyone is worse off if resources deteriorate beyond repair. Maintain-
ing natural resources is also a very salient problem (c) to locals (Purzycki,2016).
Furthermore, littering, pollution and over-exploitation are often anonymous acts; in
many cases it is impossible to identify the perpetrator (d), and hence, appeals to
supernatural monitoring is likely a more effective means than secular alternatives
(e). Finally, since pollution and over-exploitation are inherently collective affairs in
that the severity of the problems depend on the accumulation of litter and/or over-
extraction of resources over time (i.e., any one person is unlikely to cause serious
damage on their own), the pay-offs of the dilemma may be considered non-obvious
for the involved individuals (f).
In contrast, consider Henrich and Henrich’s (2010) study of pregnancy-related
food taboos in Fiji. They found that the marine species that pregnant and breast-
feeding women are not supposed to consume are those carrying the highest levels of
toxins to which pregnant and lactating women and newborns are particularly vulner-
able. However, while these taboos appear to make adaptive sense and food taboos
in general are often supernaturally enforced (Meyer-Rochow,2009), local Fijians do
not regard the pregnancy-related taboos as something local deities are concerned
with. Henrich and Henrich (2010) suggest that “this is because compliance with the
taboo is pretty high (threats of social sanctions or of harm to one’s infant seem suf-
ficient to maintain them), so threats of supernatural sanctions may be unnecessary
to sustain the adaptive behavior” (suppl. mat., p. 36). In other words, the costs
of violating the taboo—although potentially high and salient (b and c)—are clear
and unambiguous with a predictable outcome likelihood (f) and relatively easy and
effective to enforce with secular social norms and stigma (d and e). Note in addition
that the interests of the actors involved (the women, their families, the community)
are aligned around the same outcome (or are at least not in any obvious conflict),
namely avoiding the consumption of toxic foods. It is therefore unclear whether the
Fijian pregnancy-related food taboos represent an actual social dilemma (a). Hence,
6
according to the criteria listed, we would not expect the Fijian pregnancy-related
food taboos to constitute a god-problem.
In sum, we submit that appeals to gods will include or indicate behaviors that
engage the kinds of challenges people face, in particular threats to coordination and
cooperation, real or ersatz1(Bendixen and Purzycki,2020;McNamara and Purzycki,
2020;Purzycki and Sosis,2022, ch. 10). In the next sections, we systematically
survey the content of religious appeals with freely listed ethnographic data from eight
diverse field sites about what pleases and angers locally relevant deities (Section 3)
and subsequently contextualize these cultural models of local gods’ concerns (Section
4). We expect the most salient items (i.e., the most frequently and earliest listed
responses; see Section 3.2) in these data to point to salient threats to cooperation,
coordination, and conflicts of interests—and behaviors that might address these—
within local social ecologies.
3. An empirical assessment of free-list data on deities’ likes and dislikes
3.1. Participants
The present data is part of The Evolution of Religion and Morality Project2
(Purzycki et al.,2022a) that was designed to explore and test hypotheses pertaining
to religious beliefs and cooperation. The main study consisted of a series of experi-
mental economic games and a battery of demographic and religiosity questions (see
Lang et al.,2019;Purzycki et al.,2016b,a,2018a, for further methodological details
and presentations of field sites) conducted in eight diverse field sites from around
the world (Table 1). We paid participants an initial fee of ∼25% of the local average
daily wage, and the participants kept their earnings in the experimental economic
games with a potential total of ∼50% of the local average daily wage. Participation
in the main study took a total of around 90 minutes.
1Note that our account does not assume that religious beliefs and behaviors are always beneficial
for a community (see e.g., Edgerton,1992). Just as cultural and ecological pressures can push a
group to beneficial behavioral patterns, so too can cultural and ecological pressures (including time
lag) push groups to sub-optimal traditions (e.g., Boyd and Richerson,1990,1992;Richerson and
Boyd,2005, chapter 5; for a case study, see le Guen et al.,2013). When we here focus less on
maladaptive aspects of religion, it is mainly for the reason that, all else being equal, it is the more
beneficial—or at least not extremely harmful—traditions that on the whole tend to survive long
enough to become objects of contemporary scientific inquiry.
2The full protocol, summary of methods, and descriptions of the cultural samples are available
at:
https://github.com/bgpurzycki/Evolution-of-Religion-and- Morality.
7
Table 1: Summary of sample size, demographics, and items listed. Aside from mean age and
standard deviations, values reflect cross-variable sample size and, in parentheses, average number of
items listed across domains. Number of individuals for which demographics are available and who
completed at least one of the free-list tasks (N) and sub-sample sizes per free-list sub-domain (MG:
moralistic god, LG: local god, PO: police). Note that there are mismatches between demography
and free-list data, since some participants completed the demographic survey but not the free-lists
and vice versa. Some participants completed the demographic survey but did not report their age.
Site NAge Females MGpleases MGangers LGpleases LGangers POpleases POangers
Coastal Tanna 42 35.3 (14.4 ) 21 42 (2.2) 42 (3.0) 41 (1.2) 37 (1.3) 42 (2.5) 41 (3.7)
Hadza 68 39.8 (12.1 ) 31 58 (1.6) 57 (1.7) 47 (1.3) 47 (1.6) 69 (1.2) 69 (1.3)
Inland Tanna 73 37.4 (16.2 ) 36 74 (1.6) 74 (1.7) 74 (1.7) 74 (1.3) 72 (1.3) 68 (2.1)
Lovu Fiji 75 44.6 (16.9 ) 52 80 (2.7) 80 (2.8) – – 79 (3.1) 77 (3.3)
Maraj´o, Brazil 73 34.5 (13.2) 39 73 (2.5) 76 (2.6) 58 (1.8) 60 (1.5) 68 (2.1) 65 (2.4)
Mauritius 77 36.2 (14.9 ) 26 80 (3.5) 69 (2.4) 52 (2.4) 48 (1.6) 80 (3.2) 83 (3.4)
Tyva Republic 72 34.3 (13.0 ) 52 73 (2.6) 73 (2.7) 72 (2.4) 72 (2.2) 73 (4.0) 71 (3.8)
Yasawa Fiji 67 38.4 (16.1) 37 105 (3.1) 105 (3.0) 105 (2.2) 105 (2.2) 105 (5.0) 105 (4.9)
Total 547 37.7 (14.9 ) 294 585 (2.6) 576 (2.5) 449 (1.9) 443 (1.8) 588 (2.9) 579 (3.2)
3.2. Methods
Prior to the main study of which the present data was a part, we conducted
preliminary ethnographic interviews where participants listed gods and spirits that
were important in their communities. For each listed deity, we asked follow-up
questions about their knowledge breadth, how punitive they were, and how concerned
they were with moral norms (i.e., interpersonal social behaviors that benefit or harm
others). Taking these ratings, we selected one deity that was the most knowledgeable,
morally concerned, and punitive (the “moralistic deity” henceforth) and another
locally important deity that was relatively less associated with these qualities (the
“local deity” henceforth). We used these deities to design the main project with a
new sample of individuals. Table 2details the deities we selected and the working
languages across our eight field sites.
Table 2: Deities selected, language used in, and primary economy in each field site. Note
that no local gods were identified for the Lovu Fiji sample.
Site Economy Moralistic deity Local deity Language of study
Coastal Tanna Hort./Hunting Christian god Tupunus Bislama
Hadza Hunting Haine Ishoko Hadzane/Swahili
Inland Tanna Hort./hunting Kalpapen Tupunus Navhaal
Lovu, Fiji Wage labor Shiva — Fiji-Hindi/English
Mauritius Wage labor Shiva spirit (nam) Mauritian Creole
Maraj´o, Brazil Wage labor Christian god St. Mary Portuguese
Tyva Republic Wage labor/herding Buddha-Burgan spirit-masters (cher eezi) Tyvan
Yasawa, Fiji Fishing/farming Christian god ancestor spirits (kalou-vu) Bauan Fijian
8
An ideal method for soliciting naturalistic, discrete, and quantifiable ethno-
graphic data is the free-list method (Smith,1993;Smith and Borgatti,1997). The
method entails participants listing items that represent their knowledge about some
topic. Compared to other social scientific instruments such as pre-fabricated item
response scales, free-listing also ensures cultural relevance and validity (a particu-
larly pressing issue in the study of “indigenous” religions; see Maarif,2019), since
the content is fully dictated by participants. As such, it is maximally useful among
non-numerate populations.
After processing (see Supplementary Section S2), listed items are analyzed ac-
cording to their frequency and order of mention to generate Smith’s S, an index of
cognitive-cultural salience3. As such, the free-list method can reveal variation within
and across topics, individuals and groups. Among many other questions and items
(see Purzycki et al.,2016a), we asked participants to freely list:
1. the kinds of things the moralistic deity cares about or like
2. the kinds of things the moralistic deity dislikes.
3. the kinds of things the local deity cares about or likes.
4. the kinds of things the local deity dislikes.
5. up to 5 things that the police likes.
6. up to 5 things that the police dislike.
We designed these lists to be capped at five items per domain due to time, but some
did offer more. We probed participants about deities in counterbalanced fashion
(items 1-4), followed by questions about the police (items 5-6)4. In line with the
prediction that some features of social life are more likely to be associated with
supernatural rather than secular concern, questions about the police were included
as a contrast to the two kinds of deities in order to directly compare cognitive and
cultural models across secular and supernatural agents.
We coded these free-list data in two ways (see Supplementary Section S2 for
further notes and procedures). The first method was general; two independent coders
coded the data with the following twelve-category rubric drawn from Purzycki and
McNamara (2016):
3Specifically, the Smith’s Sof a listed item is calculated simply as S=Σ(n+1−k)
n
N, where nis the
number of items an individual lists, kis the order number in which an item was listed, and Nis the
total sample size of the specific task. See Supplementary Section S4 for more details.
4We also asked participants to list up to 5 behaviors that make someone a good/virtuous/moral
person and up to 5 behaviors that make someone a bad/immoral person. For an empirical report
on the results of these questions, see Purzycki et al. (2018b).
9
1. Morality: generalized behaviors that have a benefit or cost to other people
(e.g., hurting, being generous, sharing, etc.)
2. Virtue: individual qualities that may or may not have social ramifications
(e.g., hard-working, kind, bad conscience, etc.)
3. People: in reference to the quality, and/or the state of people (e.g., people,
people stay in good health, live beings, happy, etc.)
4. Etiquette: conventional social behaviors that have no immediate cost or ben-
efit to others (e.g., shaking hands, wearing the proper clothes, etc.)
5. Substance Use/Abuse: Items that involve the use of illicit substances
6. Religion: any non-ritual or non-behavioral item concerned with the supernat-
ural (e.g., faith, devotion, loving god, etc.)
7. Ritual: any behavior or object used in ritual devoted to the supernatural (e.g.,
praying, meditation, offerings, sacrifices, not participating in ritual, etc.)
8. Ecology: any behavior or object affecting non-human relationships (e.g., pol-
lution, keeping sacred places clean, gardening, etc.)
9. Food: food items (e.g. yam, milk, etc.)
10. Miscellaneous: miscellaneous items, items that cross-cut categories, etc.
11. D/K: I don’t know, not sure, etc.
12. Specific: Items that are specific to a culture (e.g., bel’ leaf, artysh, etc.). These
were subsequently re-coded into one of the other codes after consultation with
field researchers.
Inter-coder reliability was generally quite high across domains (Supplementary Table
S1). In cases of inter-coder conflicts, B.G.P. selected the one code of the two that
best reflected the coding rubric (Supplementary Table S2).
The same individuals also coded the free-list data in a more specific, bottom-up
fashion. As these specific codes are by definition more subjective and unconstrained
in nature, there was much variation and inconsistency in granularity, labels, and
(mis)spellings across the coders. We subsequently cleaned such entries and lumped
together semantically similar items (e.g., what we coded as “No stealing” was initially
coded by one assistant as “No stealing – Burglers [sic]”, “No stealing – Robbery”,
“No Stealing – Thieves”). We report the specific codes from one coder (NC), who
developed the most fine-grained coding scheme (see Supplementary Section S2).
We conducted salience analyses (Supplementary Section S4) of the coded free-
list data using the AnthroTools package (Purzycki and Jamieson-Lane,2016) for R
(R Core Team,2021). Once the free-list data were analyzed and summarized, T.B.
examined the ethnographic literature for each field site in order to contextualize the
data and evaluate the predictive criteria we detailed earlier. Importantly, the criteria
were developed prior to the site-by-site ethnographic examinations (cf., Bendixen
10
and Purzycki,2020), which goes some way towards ensuring that our account is not
“over-fit” to the data and field sites at hand.
3.3. Results
3.3.1. Cultural models of agents’ concerns
Our key findings can be summarized as follows: (1) participants systematically
responded that their deities and the police are angered and pleased by human be-
havior; (2) the moralistic gods and the police converge on similar themes of moral
concern but also diverged in predictable ways; (3) while the moralistic gods and the
police are primarily and unambiguously moralistic, the local gods also consistently
exhibit some salient moral concern; and (4) compared with the moralistic gods and
the police, which are generally similar across cultures, the local gods are associated
with unique site-specific concerns. We elaborate on these findings in the following.
Since much of contemporary research emphasizes the punitive aspects of deities, we
focus here on what angers these agents (see Supplementary Figure S2 for panels of
what pleases these agents).
Figure 1reports Smith’s Sfor the general codes across sites and agents. Tables
3and 4report the most salient global and site-specific codes across agents (see Sup-
plementary Section S6.1 for expanded salience tables of the general codes). Globally,
both the moralistic gods (MO) and the police (PO) converge on disliking breaches of
Morality (SMG = 0.64, SPO = 0.78) and Virtue (SMG = 0.26, SPO = 0.11) and both
are pleased by social harmony (People: SMG = 0.16, SPO = 0.15). However, there
are predictable divergences; in contrast to the police, moralistic gods are pleased by
religious thought (Religion: SMG = 0.11) and behavior (Ritual: SMG = 0.30), while
the police are more often associated with disliking Substance Use/Abuse (“Drugs”
in graphs and tables: SPO = 0.16)5. Notably, local deities (LG) are also angered
(SLG = 0.19) and pleased (SLG = 0.11) by (im)moral behaviors, but typically to a
lesser degree than moralistic gods and the police, thus confirming our design and
operationalization; according to our design and definitions, “moralistic gods” are
indeed more moralistic than our “local gods”.
The conceptual overlap between the moralistic deities and the police is further
nuanced when examining the specific codes. Aggregated across all sites, the moral-
istic deities and the police share the same top four items in terms of what they are
angered by: stealing, violence, lies, and disobedience (in general and toward the
law) (see Table 5). These are unambiguously moral items. As was to be suspected,
5For the police, note also the high salience of “Don’t know” among the Hadza and Inland Tanna,
two sites that are generally unfamiliar with a formal and organized police force.
11
0.0 0.2 0.4 0.6 0.8 1.0
D/K Ecology Etiquette Food Misc. Morality People Religion Ritual Drugs Virtue
Smiths' S
What angers the moralistic gods?
0.0 0.2 0.4 0.6 0.8 1.0
D/K Ecology Etiquette Food Misc. Morality People Religion Ritual Drugs Virtue
What angers the local gods?
Smiths' S
************
0.0 0.2 0.4 0.6 0.8 1.0
Coastal Tanna
Hadza
Inland Tanna
Lovu Fiji
Marajó
Mauritius
Tyva Republic
Yasawa Fiji
D/K Ecology Etiquette Food Misc. Morality People Religion Ritual Drugs Virtue
Smiths' S
What angers the police?
Figure 1: Smith’s Sof general codings for what angers the moralistic gods, the local gods and the police. Note that
no local deities were identified for Lovu Fiji, hence the asterisks. D/K = Don’t know; Misc. = Miscellaneous; Drugs = Substance
Use/Abuse.
12
however, there are also some notable differences. In particular, the police are an-
gered by concrete crimes, such as rape and murder, whereas the moralistic deities
are displeased by more abstract, religious transgressions such as sin, swearing, and
general “bad behavior”. In terms of what pleases moralistic deities and the police,
there is some, but less overlap. Among the top three specific codes, obedience is
again a shared concern but the police are primarily pleased by behaviors and qual-
ities that directly support human social interactions, including abiding by the law,
not stealing, honesty, discipline, and no violence. In contrast, the moralistic deities
are again associated with more general themes, usually with religious connotations,
including prayer, general “good behavior”, human welfare, faith, and truth. Taken
together, the specific codes support the inference from the general codes in that the
moralistic deities and the police are both very similar in their moral concern but
also each associated with domain-specific items, namely law and crime for the police
and religious devotion for the moralistic deities. The local gods, even though they
are clearly associated with moral behaviors like the other two agents, are consider-
ably more cross-culturally diverse, exhibiting unique signatures of localized concerns
(see Figure 1and Table 4). As such, we now turn to a finer-grained site-by-site
contextualization of the free-list data on local gods’ concerns.
Table 3: Global Smith’s Sof and number of participants who listed the most salient
general codes of what pleases and angers the moralistic gods (MG), the local gods (LG)
and the police (PO). Only Smith’s S≥.10 is reported. “Substance Use/Abuse” is abbreviated
as “Drugs”. “Don’t know” excluded.
MGpleases MGangers LGpleases LGangers POpleases POangers
Morality Morality Ritual Religion Morality Morality
(0.31, 220) (0.64, 413) (0.29, 152) (0.23, 112) (0.55, 372) (0.78, 484)
Ritual Virtue Drugs Morality Virtue Drugs
(0.30, 212) (0.26, 201) (0.18, 91) (0.19, 94) (0.30, 239) (0.16, 137)
Virtue – Ecology Ritual People Virtue
(0.28, 218) –(0.12, 57) (0.19, 101) (0.15, 130) (0.11, 106)
People – Morality Ecology Misc. –
(0.16, 119) –(0.11, 70) (0.15, 73) (0.14, 107) –
Religion – Food – – –
(0.11, 92) –(0.11, 52) – – –
13
Table 4: Cross-cultural Smith’s Sof and number of participants who listed the most
locally salient general codes of what pleases and angers the moralistic gods (MG),
the local gods (LG) and the police (PO). “Substance Use/Abuse” is abbreviated as “Drugs”.
“Don’t know” excluded.
Group MGpleases MGangers LGpleases LGangers POpleases POangers
Coastal Tanna People Morality Ecology Ritual Morality Morality
(0.56, 25) (0.90, 41) (0.16, 7) (0.22, 8) (0.52, 25) (0.95, 41)
Hadza People Morality People Morality People Morality
(0.31, 22) (0.49, 30) (0.29, 15) (0.47, 23) (0.21, 15) (0.35, 24)
Inland Tanna Ecology Morality Food People Morality Morality
(0.34, 26) (0.35, 29) (0.56, 43) (0.22, 17) (0.22, 17) (0.64, 45)
Lovu Fiji Morality Morality – – Morality Morality
(0.57, 53) (0.80, 68) – – (0.77, 73) (0.96, 76)
Maraj´o, Brazil Virtue Morality Ritual Morality Morality Morality
(0.36, 32) (0.63, 56) (0.45, 27) (0.39, 25) (0.46, 35) (0.77, 52)
Mauritius Ritual Morality Morality Ritual Morality Morality
(0.68, 62) (0.70, 51) (0.40, 25) (0.40, 21) (0.61, 58) (0.82, 75)
Tyva Republic Morality Morality Ritual Ecology Morality Morality
(0.31, 29) (0.61, 51) (0.57, 46) (0.69, 53) (0.53, 50) (0.77, 67)
Yasawa Fiji Virtue Morality Drugs Religion Morality Morality
(0.58, 76) (0.68, 87) (0.71, 79) (0.73, 84) (0.90, 103) (0.96, 104)
Table 5: Global Smith’s Sof and number of participants who listed most salient spe-
cific codes of what pleases and angers the moralistic gods (MG) and the police (PO)
(excludes “don’t know”). Only Smith’s S≥.05 is reported.
MGpleases MGangers POpleases POangers
Prayer Stealing Law Abiding Stealing
(0.13, 102) (0.18, 142) (0.15, 127) (0.44, 315)
Obedient Lies No Stealing Violence
(0.07, 55) (0.16, 120) (0.12, 97) (0.22, 184)
Behaviour - Good Violence Obedience Disobedience - Law
(0.06, 49) (0.11, 89) (0.07, 52) (0.13, 110)
People Disobedient Honesty Lies
(0.06, 36) (0.07, 57) (0.07, 47) (0.11, 101)
Faith Sin Discipline Murder
(0.05, 41) (0.07, 49) (0.06, 48) (0.10, 92)
Truth Swearing No Violence Disobedience
(0.05, 39) (0.06, 50) (0.05, 49) (0.06, 48)
– Behaviour - Bad – Rape
–(0.06, 42) –(0.06, 62)
– Murder – –
–(0.06, 49) – –
14
4. Local gods in ethnographic context
By contextualizing the most culturally salient facets of local deities’ concerns at
each field site, we can assess aspects of our cultural evolutionary account. Here, we
restrict our discussion to the most salient (i.e, Smith’s S≥.05) specific codes (see
Supplementary Tables S12-S15 for expanded salience tables). We provide broader
ethnographic discussions of the local spirits in Supplementary Section S6. Recall our
account, which predicts that god-problems—the things that people associate deities
with—will tend to be: (a) game-theoretic social dilemmas, that are (b) costly and/or
(c) salient, (d) difficult to monitor and enforce with (appeals to) secular alternatives
and/or (e) more convincingly and effectively enforced with supernatural appeals,
and/or (f) where the consequences of norm deviation are opaque. Importantly,
the free-list data can neither directly confirm nor reject our account since we only
measure here are individual appeals. However, we do argue that the data bolster
significant aspects of it; as we treated the data as appeals, our subsequent site-by-site
ethnographic inquiries yielded some important insights.
4.1. Tyva Republic
Consistent with previous research (Purzycki,2011,2013,2016), spirit-masters
of the Tyva Republic dislike pollution and destruction of the natural environment
and are primarily pleased by “no pollution” (see Section 2.3) and ritual devotion
(e.g.,sang salyr/sanctification, bowing, and various food offerings). Throughout In-
ner Asia, people make offerings at cairns and other places of ritual significance that
are strategically located in the landscape, such as at territorial borders and natural
springs. Adhering to local rituals signals trustworthiness among Tyvans (Purzycki
and Arakchaa,2013) and violating ritual and resource obligations are believed to
result in bad luck. Taken together, appeals to spirits and their associated behaviors
may align people toward the problem of curbing instances of territorial trespass-
ing and resource over-exploitation (Section 2.3) and thereby reduce costly conflict
between neighboring camps (Purzycki,2010;Purzycki and Arakchaa,2013). The
Tyvan free-list data, then, fit with the criteria for god-problems discussed earlier, in
that territorial defense, disputes, and pollution can be construed as social dilemmas
(a) with salient and real costs to people involved (b and c). Conceivably, since terri-
tories cover vast and sparsely inhabited land and pollution is typically anonymous,
these dilemmas are difficult to police with secular means (d), such as patrolling, and
the pay-offs are likely opaque (f) (e.g., pollution accumulates over time; not per-
forming rituals saves time, effort and material goods, but being caught trespassing
or not adhering to ritual prescripts is potentially very costly) without appeals to
supernatural monitoring (e).
15
4.2. Tanna
In both of the Tanna sites, food crops, gardening, and garden rules are recur-
ring themes of concern for Tupunus, the spiritual force of the local sacred garden
system. On Tanna, indigenous garden rules and taboos regulate who can enter the
sacred crop gardens and what should be done in the gardens at various times of the
year (Bonnemaison,1984,1991;Flexner et al.,2018;Kouha,2015). Importantly,
Tupunus is angered by taboo violations and punishes perpetrators with sickness
and bad luck (Atkinson,2018;Nehrbass,2011). In line with the god-problem cri-
teria, ethnographic sources suggest that garden taboos revolve around conceivable
threats to coordination and cooperation, particularly ensuring proper cultivation
and distribution of collective resources (Bonnemaison,1991, p. 75-76, 86) and their
maintenance (Flexner et al.,2018, p. 258). Resource management constitutes a
costly (b) and salient (c) set of game-theoretic dilemmas (a) (Hardin,1968;Ostrom,
2009;Rogers,2020) with opaque pay-offs (f) in that resource mismanagement is a
cumulative problem (see Section 2.3). In lieu of dedicated secular institutions (d;
indeed, on Tanna, the police are not thought to care about these matters; see Tables
S14-S15), it might be the case that defection is more effectively enforced by spirit
beliefs and appeals (e).
4.3. Yasawa, Fiji
Yasawans free-list the ancestor spirits, kalou-vu, as primarily pleased by “kava”
rituals and their own “worship”. Kava is a pepper plant with sedative properties
that can be prepared into a mildly narcotic substance and, according to one ethnog-
rapher, “kava stands metaphorically at the center of Fijian public life” (Tomlinson,
2007, p. 1066). Throughout Polynesia, kava is consumed at nightly drinking cer-
emonies traditionally associated with ancestor worship (Shaver,2015;Shaver and
Sosis,2014;Tomlinson,2004;Turner,2012). Consistent with the god-problem crite-
ria, the religious system involving kalou-vu beliefs and kava drinking is closely linked
to central (c) cooperative affairs (a); kava ceremonies are arenas for a host of salient
(b) social activities, including competition for status among males (Shaver,2015;
Turner,2012), forging social bonds and coordinating communal projects (Tomlinson,
2004), as well as dissolving disagreements and aligning economic interests (Tomlin-
son,2007); activities, with inscrutable long- and short-term costs and benefits (f),
where appeals to spirits (e) are perhaps particularly potent at disincentivizing de-
fection above and beyond locally available alternatives (d).
However, interpreting the free-listed dislikes of the ancestor spirits among Ya-
sawans is complicated by the presence of a particular form of Christianity. The
ancestor spirits are perceived as angered by such things as “faith”, the “word” and
16
“grace” of God, and “truth”, themes closely related to Christianity. As traditional
ancestor worship and kava-drinking are viewed as illegitimate and sometimes associ-
ated with witchcraft from the perspective of the local Christian churches (McNamara,
2012, p. 17-18; Tomlinson,2004;Turner,2012), the free-list responses reveal a form
of cultural competition between Christianity and traditional ancestor beliefs (Mc-
Namara and Henrich,2018;McNamara et al.,2021) where Christian churches have
successfully demonized the kalou-vu. We see a similar feature in Mauritian responses
about nam spirits.
4.4. Mauritius
Nam spirits, the local gods of the Mauritius sample, are similar to the Western
concept of the soul (Kundt´ova Klocov´a et al.,2022;Xygalatas et al.,2018). When a
person dies, the nam leaves the body of the deceased. If the death was undramatic
and natural, and if the bereaved honor the deceased with prayer and ritual offerings,
the nam will peacefully journey to the realm of the spirits. However, if the death
was sudden, unexpected or violent, the spirit is hindered in its transition and can get
trapped between the world of the living and the dead. In these cases, rituals, prayer,
and offerings toward these spirits is critical lest they transform into jab, evil spirits
that retaliate ritual neglect with misfortune, illness, or death (Colwell-Chanthaphonh
and de Salle-Essoo,2014;Sussman,1981). Therefore, nam spirits, in the form of jab,
are sometimes associated with black magic and sorcery (Kundt´ova Klocov´a et al.,
2022;Xygalatas et al.,2018). The ambivalent nature of these entities may explain
why nam are free-listed as pleased with predominantly immoral things, such as “bad
behavior”, “fear”, and “revenge”. Similar to the case of Yasawan ancestor spirits,
nam are displeased with such things as “prayer”, “good people” and “God”, which
might reflect cultural antagonism between local religious systems, since sorcery is
illegal under Mauritian law (Kundt´ova Klocov´a et al.,2022;Xygalatas et al.,2018).
However, nam spirits like “bad behavior” but also (the act of) “loving”. Further,
“prayer” is something they both like and dislike. These seeming contradictions may
be a result of distinct and conflicting cultural models of nam as either a good-willed
spirit or a force of evil in the form of jab (cf., Kundt´ova Klocov´a et al.,2022), and,
again, may reflect cultural competition between different local belief systems in that
nam like prayer directed towards themselves but dislike prayer dedicated to other
“rival” deities. Note that for these reasons this particular ethnographic context
complicates assessing our predictive criteria (a through f).
4.5. Hadza
Among the Hadza, Ishoko, represented by the Sun, was selected as the local
deity. Ishoko is free-listed as primarily concerned with (im)moral deeds (e.g., in-
17
sults, loving, murder, sharing, stealing, violence) and virtuous states and qualities
(e.g., peace, good heart) as well as ritual acts (e.g., singing). Morality constitutes
a straightforward set of god-problems, in that (im)moral actions almost always in-
volve social dilemmas (a) with substantial (b, c) cost/benefit trade-offs (Alexander,
1987;Curry et al.,2019;Purzycki et al.,2018b) that might be easier to coordinate
with appeals to an intervening supernatural agent (e), particularly when secular in-
stitutions are weak or absent (d; note that many participants of our Hadza sample
responded “Don’t know” with regards to what the police dislike, indicating a general
unfamiliarity with formal law-enforcement, see Figure 1).
The relatively high moral salience of Ishoko is worth highlighting. As a cul-
tural group, the Hadza is often regarded as a prototypical hunter-gatherer society
(e.g., Peoples and Marlowe,2012;Wright,2010), but no other site in our sample
scores a higher Smith’s Sof Morality for something that upsets a local god than
the Hadza6(see Figure 1). This finding supports a classic view, namely that the
presence of moralistic deities are not limited to large-scale, complex societies (e.g.,
Evans-Pritchard,1965;Malinowski,1992). Complimentary analyses further indicate
that the moralistic concerns of Ishoko (and Haine, the “moralistic god” of our Hadza
sample) cannot be fully accounted for by external influence from moralizing world
religions and missionary activities (Purzycki et al.,2022b;Stagnaro et al.,2022).
4.6. Maraj´o, Brazil
In Maraj´o, Brazil, St. Mary (Our Lady of Nazareth – Nossa Senhora de Nazar´e)
was the selected local deity. It is clear from the free-lists that St. Mary resembles a
moralistic deity; she is first and foremost concerned with (im)moral behavior (e.g.,
loving, violence, sin, murder) as well as ritual and religious acts (e.g., prayer, faith,
worship) dedicated to her (see Table 4). While ethnographic details of this field site
do not lend themselves to a fine-grained examination of the predictive criteria, as
noted earlier, (im)moral behaviors are generally costly (b) and pertinent (c) social
dilemmas (a) that often require third-party policing (d), such as appealing to a
6Note that some participants considered Haine and Ishoko to be identical entities, an observation
that might account for the high moral salience of Ishoko as well as the general similarity between
Haine and Ishoko (see e.g., Table 4). In cases where participants said that Haine and Ishoko are
the same, the free-list data from Haine were duplicated to Ishoko, a decision made by the local field
research team. However, to assess how this decision impacts the main results, in the Supplements
we analyze and plot the salience of the general codes separately for those Hadza participants who
said that Ishoko and Haine are “different” or the “same” (Figures S3 and S4). Analyzing these two
groups of participants separately does not substantially change the main results. See Sections S3,
S7, and Figures S3 and S4 for further discussion.
18
watchful and punitive deity (e).
The moralistic deity for Maraj´o was the Christian God, and since St. Mary is
central to Catholicism, the primary religious denomination at this field site, it is likely
that the close conceptual link between St. Mary and the Christian God influences
adherents’ conception of the former’s concerns. Indeed, other work with some of
the present participants found that ratings of St. Mary as a punitive and moralistic
figure predicts increased prosociality in anonymous economic games (Cohen et al.,
2018), suggesting that St. Mary shares important features with moralizing deities in
general (Lang et al.,2019;Purzycki et al.,2018a,2022b). In this sample, however, the
general codes suggest that St. Mary is distinct from the Christian God in at least one
way, namely, that Ritual is more salient for her (SLikes = 0.45; SDislikes = 0.19) than
for the Christian God (SLikes = 0.27; SDislikes = 0.08; see Table 4and Supplementary
Tables S4-S7). This salient association is likely due to the fact that St. Mary is
the patron saint of the sampled village of Pesqueiro and residents throw religious
festivals in her honor (see Cohen et al.,2018;Purzycki et al.,2016a), suggesting that
the free-list data reflect a key feature of local cultural tradition.
5. Discussion
We began with asking the following questions: How does the content of beliefs
about and appeals to gods vary across groups, and what accounts for this variation?
Despite a growing contemporary literature on moralizing aspects of deities, rela-
tively few studies have systematically investigated the many other concerns that
deities are associated with and their ethnographic contexts. To initiate this inquiry,
we first presented a novel cultural evolutionary account of god beliefs and appeals,
which includes a set of predictions, derived from prior theoretical, experimental, and
ethnographic evidence, for when we might expect people to appeal to the supernat-
ural. We then reported individual-level free-list data on a diverse set of deities’ likes
and dislikes across eight societies and contextualized the cultural models of local
deities’ concerns in light of our account. As noted throughout, the degree to which
the free-list data and the ethnographic literature allowed a direct assessment of the
predictive criteria varied across sites. However, we argue that the data are at least
consistent with our account.
Several key findings are worth highlighting, which provide preliminary answers to
our guiding questions: First, we show that rather than random, arbitrary, universal,
or idiosyncratic, individuals’ appeals to locally relevant gods’ concerns are i) con-
stituent parts of shared cultural models that ii) systematically point to iii ) human
behaviors associated with site-specific contexts that iv ) mediate interpersonal rela-
19
tionships. In other words, even though gods could have cared about mythical ideas
or idiosyncratic things that individuals transparently project onto the gods, appeals
to gods’ concerns point to human behaviors that share a constellation of features that
reflect local pressing challenges to social life, referred to herein as “god-problems”.
Second, while the moralistic gods and the police are closely aligned in terms of gen-
eral moral salience, supporting a popular notion that deities of world religions come
to resemble law-enforcing entities (e.g., Johnson,2016;Norenzayan,2013), the spe-
cific content of their moral concerns also predictably diverge—the police are more
often concerned with concrete crimes (e.g., rape and murder), whereas the moralis-
tic gods are concerned with personal and collective displays of devotion and general
“good behavior”. Third, supporting a classical view that deities of even smaller-scale
societies are often morally salient and relevant (e.g., Evans-Pritchard,1965;Lang,
1909;Malinowski,1936) local gods also exhibit stable moral concern, though not
typically to the extent of moralistic gods or the police. In this final section, we first
discuss how alternative approaches to accounting for cross-cultural variation in god
beliefs and appeals fare in the face of these findings and close by considering some
key implications and cautionary notes of the present work.
5.1. Alternative accounts of variation in god beliefs and appeals
Consider first the view that god concepts are “catchy” because gods are in-
tuitively thought to be morally interested (Boyer,2000,2001). This is a helpful
perspective in that it predicts that gods will revolve around consequential aspects
of social life, a prediction we confirmed. However, while deities might in general be
intuitively moralistic (Purzycki et al.,2012,2022b), we found that there was consid-
erable cross-cultural variation in the explicit appeals people employ, suggesting that
the local context might play a more important role in the evolution of god beliefs and
appeals. By way of illustration, consider the idea that gardening is a “catchy” god
concern on Tanna (Section 4.2), but not in any other of our sites. This particularism
makes sense in light of the ethnographic context but is not obviously accounted for
by claims about beliefs being “catchy” for their general social or moral salience per
se.
Other approaches (e.g., Johnson et al.,2015;Spiro and D’Andrade,1958) in-
vestigate variants of the “projection hypothesis”, namely that gods’ concerns are
projections of believers’ own concerns, temperaments, and interests. While this sug-
gestion might help address some very specific aspects of beliefs (e.g., when elites use
appeals to gods as a means for social control; Ellwood 1918;Swanson 1960, chap-
ter 9), it is limited in a few important ways (Bendixen and Purzycki,2020). First,
the many things that deities are not associated with are arguably as revealing as
20
the things that they are associated with, and overall we do not find that deities care
about all manner of personal whims and preferences. As we showed, representational
models of deities are, overall, culturally shared rather than idiosyncratic. Second,
many god-problems such as commitments to costly ritual routines and temporal or
spatial prohibitions on the use of natural resources, are costly to individuals and
therefore are inconsistent with immediate, projected self-interest. Third, individuals
do not simply project their own explicit moral values onto local gods; there is little
overlap in what people say their deities care about and what they say make a person
“good” and “bad” (see Purzycki,2016;Purzycki et al.,2018b). Contrary to some
implied variants of the projection hypothesis, then, individuals’ personal concerns
and whims are rather separate from those of the gods.
Many current accounts of god beliefs emphasize various moral aspects of deities
(e.g., Johnson,2005,2016;Norenzayan et al.,2016;Schloss and Murray,2011).
While participants attributed all gods, “moralistic” and “local” alike, with some
moral concern (see Purzycki et al.,2022b), they are also culturally specialized in that
they are each associated with a locally limited but globally diverse set of concerns
that are often not explicitly moralistic. This finding is consistent with our cultural
evolutionary account that beliefs about and appeals to deities reflect locally salient
challenges and dilemmas in the socioecological landscape. However, what the present
work shows is that even if deities are not explicitly concerned with general moral
prescriptions and prohibitions, they are unambiguously relevant to the costs and
benefits of social life (Teehan,2016). And so, if we are to conceive of morality as
cooperation (Alexander,1987;Curry et al.,2019), all of these deities would clearly
be construed as “moralistic”. This too calls for a closer examination of local deities
and their associations and in-context roles in mediating human relations.
5.2. Implications and cautionary notes
While we have argued that the present work shows that appeals to gods and
spirits reflect threats to coordination and cooperation in the local socioecological
landscape, we have not demonstrated that these beliefs and appeals actually mo-
tivate corresponding behaviors that mitigate such threats. Even though particular
religious systems exhibit clear features of adaptive self-organization (e.g., Bird et al.,
2013;Lansing et al.,2017;Sosis and Bressler,2003) and many other cases are inter-
preted as such (e.g., Angsongna et al.,2016;Connors,2000;Leeson and Suarez,
2015;Rossano,2007;Rappaport,1968;Reynolds and Tanner,1995;Strassmann
et al.,2012), it remains unclear exactly how beliefs about and appeals to deities
21
are implicated in such systems (Purzycki and Sosis,2022).7There is experimen-
tal evidence across 15 diverse field sites showing that higher individual ratings of
gods’ general monitoring and punitive tendencies predict fairer and more generous
behavior towards co-religionists in behavioral economic games (Lang et al.,2019).
However, crucially, this work shows no clear relationship between how much money
individuals allocate to others in these experiments and how moralistic they think
their gods are.
Nonetheless, consistent with our account, some detailed ethnographic case stud-
ies employing contemporary social scientific methods suggest that god beliefs and
appeals can in fact have tractable behavioral consequences and curb defection in co-
operative dilemmas. For instance, among the Mentawai on Siberut Island, Indonesia,
the local spirit, Sikameinan, punishes people with illness or accidents for violating
meat sharing norms. Sikameinan can be appeased by hosting costly ceremonies,
which among other elements involves sharing meat with community members. As
such, beliefs about the punitive Sikameinan seems to motivate exactly the kind of
behavior that the spirit is associated with (Singh et al.,2021). To take another
example, in the Maya lowlands of Guatemala, the native Itza’ Maya traditionally
profess beliefs that local spirits protect the rain forest, for instance by punishing vi-
olations of the spirits’ “preferences” for certain species and patches of land8(Atran
et al.,2002). These beliefs appear to have co-evolved with the subsistence practices
of the Itza’, which are determined as more productive and sustainable compared to
neighboring groups’ across a range of physical measurements. Thus, spirit beliefs
among the Itza’ appear to contribute to the management of forest resources (see
also Atran et al.,1999;le Guen et al.,2013).
Finally, since social life is rife with a variety of challenges and dilemmas—modeled
as cooperative “games” (Bednar and Page,2007;Smaldino and Lubell,2014)—that
7Caution is also generally warranted in inferring functional behavior from cultural beliefs in lieu
of rigorous analysis and evidence (e.g., Elster,1983,2015;Smith and Wishnie,2000). A related
methodological issue pertains to the non-trivial task of drawing rich inferences from free-list data;
since free-list responses in isolation are often subject to several interpretative etic outcomes, insight
into the local emic context is crucial for valid conclusions. For instance, in the Mauritius case,
many participants listed food-related items (e.g., milk, bel’leaf, flowers, water) as something Shiva
likes (see Supplementary Table S10), but since certain foodstuffs play a prominent role in Hindu
ritualistic tradition, these responses were coded as Ritual according to the general coding rubric.
8These preferences “represent a synthesis of experience accumulated over generations. Violations
of spirit preferences can lead to accidents, falling ill, or worse. It matters little if the supernatural
threat is real or not: if people believe in it, the threat of punishment becomes a real deterrent”
(le Guen et al.,2013, p. 781).
22
might shift and change over time, we expect that that when communities face novel
challenges and dilemmas, appeals to gods’ concerns will co-evolve with behaviors
that address these problems (cf., Jensen,2019, ch. 7; Purzycki and Sosis,2022, ch.
10; Sørensen,2004). Some experimental and ethnographic evidence support this pre-
diction. For example, believers are more likely to emphasize that greed angers God
after receiving no return in a trust game (Purzycki et al.,2020). In Tyva, alcoholism
is a severe social ill, and drinking is a steady component of what displeases or angers
spirit-masters (Purzycki,2016) and Buddha (Purzycki and Holland,2019)9(see Fig-
ure 1and Supplementary Tables S5 and S7). In Taiwan, the traditional sea goddess,
Mazu, was recently co-opted as the patron deity of the local anti-nuclear movement
after an oracle of Mazu accurately “foresaw” the non-operation of a newly con-
structed power plant (Shih,2012). And across the globe, “eco-spiritual” movements
have recently emerged within world religions in the wake of increased sociopolitical
attention on environmentalism (e.g., Sponsel,2014). Additionally, when religious
systems compete—such as at our field sites in Mauritius and Yasawa, where spirit
worship is suppressed as illegal or illegitimate—local deities and spirits appear highly
concerned with ritualized commitments to themselves and neglect of or antagonism
toward rival deities (see Figure 1, Sections 4.3-4.4, and Supplementary Sections S5.3-
5.4); a pattern that makes sense from a cultural evolutionary perspective, in that god
concepts that did not evolve to demand constant attention, particularly when under
pressure, would likely be outcompeted by more persistent strains. All of these cases
provide hints of the ways in which religion evolves, but of course, we are only just
beginning to make sense of these evolutionary processes (Bendixen and Purzycki,
2020). In particular, cross-cultural longitudinal studies would allow researchers to
track religious appeals and behaviors across changing demographic, existential, cul-
tural, social, and ecological circumstances and thereby disentangle the various forces
at play in the cultural evolution of gods’ minds (Bendixen and Purzycki,2021).
9Purzycki and Sosis (2022, ch. 10) frame the potential costs and benefits of social drinking as a
social dilemma.
23
Acknowledgments
This work was made possible by the Cultural Evolution of Religion Research Con-
sortium, funded by a SSHRC partnership grant and the John Templeton Foundation
(awarded to A.N. and J.H.), and the Max Planck Society. T.B. and B.G.P. thank the
Aarhus University Research Foundation for generous support. We thank Adam Bar-
nett, Nicholas Chan, and Tiffany Lai for coding the data. We thank Martin Lang,
Manvir Singh, and the Religion, Cognition, and Culture research unit at Aarhus
University for critical comments on earlier drafts.
Author Contributions
T.B. and B.G.P led project development, data checks, analysis, and writing. B.G.P.
managed the greater project of which this was a part, compiled data, and organized
coding of the free-list data and T.B. led analysis and ethnographic background re-
search. J.H., A.N., and B.G.P. designed the Evolution of Religion and Morality
Project. C.A., Q.A., E.C., R.A.M., B.G.P., A.K.W., and D.X. collected data.
Data Availability
All data and code to reproduce the present study is available at: https://github.
com/tbendixen/cross-cultural-free-list-project. The main project reposi-
tory including raw data, full protocols, and related materials is available at: https:
//github.com/bgpurzycki/Evolution-of-Religion-and-Morality.
References
Alexander, R.D., 1987. The Biology of Moral Systems. Aldine Transaction, New
Brunswick.
Andersen, M., 2019. Predictive coding in agency detection. Religion, Brain &
Behavior 9, 65–84.
Angsongna, A., Ato Armah, F., Boamah, S., Hambati, H., Luginaah, I., Chuen-
pagdee, R., Campbell, G., 2016. A systematic review of resource habitat taboos
and human health outcomes in the context of global environmental change. Global
Bioethics 27, 91–111.
Atkinson, Q.D., 2018. Religion and expanding the cooperative sphere in kastom and
christian villages on Tanna, Vanuatu. Religion, Brain & Behavior 8, 149–167.
24
Atkinson, Q.D., Bourrat, P., 2011. Beliefs about God, the afterlife and morality
support the role of supernatural policing in human cooperation. Evolution and
Human Behavior 32, 41–49.
Atran, S., Medin, D., Ross, N., Lynch, E., Coley, J., Ek’, E.U., Vapnarsky, V., 1999.
Folkecology and commons management in the Maya Lowlands. Proceedings of the
National Academy of Sciences 96, 7598–7603. doi:10.1073/pnas.96.13.7598.
Atran, S., Medin, D., Ross, N., Lynch, E., Vapnarsky, V., Ek’, E.U., Coley, J.,
Timura, C., Baran, M., 2002. Folkecology, Cultural Epidemiology, and the Spirit
of the Commons: A Garden Experiment in the Maya Lowlands, 19912001. Current
Anthropology 43, 421–450. doi:10.1086/339528.
Barrett, J.L., 2000. Exploring the natural foundations of religion. Trends in Cognitive
Sciences 4, 29–34.
Bateson, M., Nettle, D., Roberts, G., 2006. Cues of being watched enhance coopera-
tion in a real-world setting. Biology Letters 2, 412–414. doi:10.1098/rsbl.2006.
0509.
Baumard, N., Hyafil, A., Morris, I., Boyer, P., 2015. Increased affluence explains
the emergence of ascetic wisdoms and moralizing religions. Current Biology 25,
10–15. doi:10.1016/j.cub.2014.10.063.
Bednar, J., Page, S., 2007. Can Game(s) Theory Explain Culture?: The Emergence
of Cultural Behavior Within Multiple Games. Rationality and Society 19, 65–97.
doi:10.1177/1043463107075108.
Beheim, B., Atkinson, Q.D., Bulbulia, J., Gervais, W., Gray, R.D., Henrich, J.,
Lang, M., Monroe, M.W., 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, E29–E34.
doi:10.1038/s41586-021-03655-4.
Bendixen, T., Purzycki, B.G., 2020. Peering into the minds of gods: What cross-
cultural variation in gods’ concerns can tell us about the evolution of religion.
Journal for the Cognitive Science of Religion 5, 142–165. doi:10.1558/jcsr.
40951.
Bendixen, T., Purzycki, B.G., 2021. Competing forces account for the stability and
evolution of religious beliefs. The International Journal for the Psychology of
Religion 31, 307–312. doi:10.1080/10508619.2020.1844969.
25
Bird, R.B., Tayor, N., Codding, B.F., Bird, D.W., 2013. Niche construction and
Dreaming logic: aboriginal patch mosaic burning and varanid lizards (Varanus
gouldii) in Australia. Proceedings of the Royal Society B: Biological Sciences 280,
20132297. doi:10.1098/rspb.2013.2297.
Bird-David, N., 1999. “Animism” Revisited: Personhood, Environment, and Rela-
tional Epistemology. Current Anthropology 40, S67–S91. doi:10.1086/200061.
Boehm, C., 2008. A biocultural evolutionary exploration of supernatural sanctioning,
in: Bulbulia, J., Sosis, R., Harris, Erica, Genet, R, Wyman, K. (Eds.), Evolution
of Religion: Studies, Theories, and Critiques. Collins Foundation Press, Santa
Margarita, CA, pp. 143–152.
Bonnemaison, J., 1984. The tree and the canoe: Roots and mobility in vanuatu
societies. Pacific Viewpoint 25, 117–152. doi:https://doi.org/10.1111/apv.
252002.
Bonnemaison, J., 1991. Magic Gardens in Tanna. Pacific Studies; Laie, Hawaii 14,
71–89.
Botero, C.A., Gardner, B., Kirby, K.R., Bulbulia, J., Gavin, M.C., Gray, R.D., 2014.
The ecology of religious beliefs. Proceedings of the National Academy of Sciences
111, 16784–16789. doi:10.1073/pnas.1408701111.
Boyd, R., Richerson, P.J., 1990. Group selection among alternative evolutionarily
stable strategies. Journal of Theoretical Biology 145, 331 – 342. doi:https:
//doi.org/10.1016/S0022-5193(05)80113-4.
Boyd, R., Richerson, P.J., 1992. Punishment allows the evolution of cooperation
(or anything else) in sizable groups. Ethology and Sociobiology 13, 171 – 195.
doi:https://doi.org/10.1016/0162-3095(92)90032-Y.
Boyer, P., 2000. Functional origins of religious concepts: ontological and strategic
selection in evolved minds. Journal of the Royal Anthropological Institute 6, 195–
214. doi:10.1111/1467-9655.00012.
Boyer, P., 2001. Religion Explained: The Evolutionary Origins of Religious Thought.
Basic Books.
Boyer, P., 2021. Deriving Features of Religions in the Wild. Human
Nature URL: https://doi.org/10.1007/s12110-021-09410-y, doi:10.1007/
s12110-021-09410-y.
26
Cohen, E., Baimel, A., Purzycki, B.G., 2018. Religiosity and resource allocation
in Maraj´o, Brazil. Religion, Brain Behavior 8, 168–184. doi:10.1080/2153599X.
2016.1267029.
Colwell-Chanthaphonh, C., de Salle-Essoo, M., 2014. Saints and evil and the wayside
shrines of mauritius. Journal of Material Culture 19, 253–277. doi:10.1177/
1359183514540066.
Colyvan, M., Justus, J., Regan, H.M., 2011. The conservation game. Biological
conservation 144, 1246–1253.
Connors, S.M., 2000. Ecology and religion in Karuk orientations toward the land,
in: Harvey, G. (Ed.), Indigenous religions : A companion. Cassel, London, pp.
139–151.
Cronk, L., 1994a. Evolutionary theories of morality and the manipulative use of
signals. Zygon 29, 81–101.
Cronk, L., 1994b. The use of moralistic statements in social manipulation: A reply
to roy a. rappaport. Zygon 29, 351–355.
Cronk, L., Leech, B.L., 2012. in: Meeting at Grand Central, Princeton University
Press.
Curry, O., Whitehouse, H., Mullins, D., 2019. Is it good to cooperate? testing the
theory of morality-as-cooperation in 60 societies. Current Anthropology 60.
Dear, K., Dutton, K., Fox, E., 2019. Do ‘watching eyes’ influence antisocial behavior?
a systematic review & meta-analysis. Evolution and Human Behavior 40, 269–280.
Edgerton, R.B., 1992. Sick societies: Challenging the myth of primitive harmony.
Free Press.
Ellwood, C.A., 1918. Religion and social control. The Scientific Monthly 7, 335–348.
Elster, J., 1983. Explaining technical change: A case study in the philosophy of
science. CUP Archive.
Elster, J., 2015. Explaining social behavior: More nuts and bolts for the social
sciences. Cambridge University Press.
27
Endicott, K.L., Endicott, K.M., 2014. Batek Childrearing and Morality, in: Narvaez,
D., Valentino, K., Fuentes, A., McKenna, J.J., Gray, P. (Eds.), Ancestral Land-
scapes in Human Evolution: Culture, Childrearing and Social Wellbeing. Oxford
University Press, Oxford.
Evans-Pritchard, E.E., 1965. Theories of primitive religion. Oxford University Press.
Fitouchi, L., Singh, M., 2021. Supernatural punishment beliefs as manipulated expla-
nations of misfortune. URL: psyarxiv.com/vejac, doi:10.31234/osf.io/vejac.
Flexner, J.L., Lindstrom, L., Hickey, F., Kapere, J., Lindstrom, L., Hickey, F.,
Kapere, J., 2018. Kaio, kapwier, nepek, and nuk: Human and non-human agency
and ‘conservation’ on Tanna, Vanuatu, in: Verschuuren, B., Brown, S. (Eds.),
Cultural and Spiritual Significance of Nature in Protected Areas, pp. 251–262.
doi:10.4324/9781315108186-17.
Ge, E., Chen, Y., Wu, J., Mace, R., 2019. Large-scale cooperation driven by
reputation, not fear of divine punishment. Royal Society Open Science 6,
190991. URL: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6731744/,
doi:10.1098/rsos.190991.
Geertz, A.W., 2020. How did ignorance become fact in American religious studies?:
A reluctant reply to Ivan Strenski. Studi E Materiali Di Storia Delle Religioni 86,
365–403.
le Guen, O., Iliev, R., Lois, X., Atran, S., Medin, D.L., 2013. A garden experiment
revisited: Inter-generational change in environmental perception and management
of the Maya Lowlands, Guatemala: A garden experiment revisited. Journal of the
Royal Anthropological Institute 19, 771–794. doi:10.1111/1467-9655.12064.
Guthrie, S.E., 1980. A cognitive theory of religion. Current Anthropology 21, 181–
203. doi:10.1086/202429.
Guthrie, S.E., 1995. Faces in the clouds: A new theory of religion. Oxford University
Press, New York.
Hardin, G., 1968. The Tragedy of the Commons. Science 162, 1243–1248. doi:10.
1126/science.162.3859.1243.
Hartberg, Y., Cox, M., Villamayor-Tomas, S., 2016. Supernatural monitoring and
sanctioning in community-based resource management. Religion, Brain & Behav-
ior 6, 95–111.
28
Henrich, J., Henrich, N., 2007. Why Humans Cooperate: A Cultural and Evolution-
ary Explanation. Oxford University Press, Oxford, UK.
Henrich, J., Henrich, N., 2010. The evolution of cultural adaptations: Fijian food
taboos protect against dangerous marine toxins. Proceedings. Biological Sciences
277, 3715–3724. doi:10.1098/rspb.2010.1191.
Henrich, J., Muthukrishna, M., 2021. The Origins and Psychology of Hu-
man Cooperation. Annual Review of Psychology 72, 207–240. doi:10.1146/
annurev-psych-081920-042106.
Jackson, J.C., Caluori, N., Abrams, S., Beckman, E., Gelfand, M.J., Gray, K., 2020.
Tight cultures and vengeful gods: How culture shapes religious belief. doi:10.
31234/osf.io/n74fx.
Jensen, J.S., 2019. What is religion? 2. ed., Routledge.
Johnson, D.D.P., 2005. God’s punishment and public goods. Human Nature 16,
410–446.
Johnson, D.D.P., 2016. God is watching you: How the fear of god makes us human.
Oxford University Press, USA.
Johnson, K.A., Okun, M.A., Cohen, A.B., 2015. The mind of the lord: Measuring
authoritarian and benevolent god representations. Psychology of Religion and
Spirituality 7, 227.
Kendal, R.L., Boogert, N.J., Rendell, L., Laland, K.N., Webster, M., Jones, P.L.,
2018. Social learning strategies: Bridge-building between fields. Trends in cognitive
sciences 22, 651–665.
Kouha, C., 2015. A comparison between the God of the Bible and the Tannese
primal gods: An apolegetic to educate Tannese Christians. Melanesian Journal of
Theology 31.2, 65.
Kundt´ova Klocov´a, E., Lang, M., Kundt, R., Xygalatas, D., 2022. Cigarettes for the
dead: Effects of sorcery beliefs on parochial prosociality in Mauritius. Religion,
Brain & Behavior .
Lang, A., 1909. The making of religion. Longmans, London.
29
Lang, M., Purzycki, B.G., Apicella, C.L., Atkinson, Q.D., Bolyanatz, A., Cohen, E.,
Handley, C., Kundtov´a Klocov´a, E., Lesorogol, C., Mathew, S., et al., 2019. Moral-
izing gods, impartiality and religious parochialism across 15 societies. Proceedings
of the Royal Society B 286, 20190202.
Lansing, J.S., Thurner, S., Chung, N.N., Coudurier-Curveur, A., Karaka¸s, a., Fe-
senmyer, K.A., Chew, L.Y., 2017. Adaptive self-organization of Bali’s ancient
rice terraces. Proceedings of the National Academy of Sciences 114, 6504–6509.
doi:10.1073/pnas.1605369114.
Leeson, P.T., Suarez, P.A., 2015. Superstition and Self-Governance. New Thinking in
Austrian Political Economy 19, 47–66. doi:10.1108/S1529-213420150000019003.
Maarif, S., 2019. Indigenous religion paradigm: Re-interpreting religious practices
of indigenous people. Journal of Studies in Philosophy 44, 103–121. doi:http:
//doi.org/10.15068/00155157.
Malinowski, B., 1936. The Foundations of Faith and Morals. London: Oxford
University Press UK.
Malinowski, B., 1992. Malinowski and the Work of Myth. Princeton University
Press, Princeton. URL: https://www.degruyter.com/document/doi/10.1515/
9781400862801/html. publication Title: Malinowski and the Work of Myth.
McNamara, R.A., 2012. When does it matter that God is watching?: Differential
effects of large and small gods on cheating as a function of material security in
Yasawa, Fiji. Ph.D. thesis. The University of British Colombia.
McNamara, R.A., Henrich, J., 2018. Jesus vs. the ancestors: how specific religious
beliefs shape prosociality on Yasawa Island, Fiji. Religion, Brain & Behavior 8,
185–204.
McNamara, R.A., Purzycki, B.G., 2020. Minds of gods and human cognitive con-
straints: socio-ecological context shapes belief. Religion, Brain & Behavior 10,
223–238. doi:10.1080/2153599X.2019.1678510.
McNamara, R.A., Senanayake, R., Willard, A.K., Henrich, J., 2021. God’s mind on
morality. Evolutionary Human Sciences 3, E6. doi:10.1017/ehs.2021.1.
Meyer-Rochow, V.B., 2009. Food taboos: their origins and purposes. Journal
of Ethnobiology and Ethnomedicine 5, 18. URL: https://doi.org/10.1186/
1746-4269-5-18, doi:10.1186/1746-4269-5-18.
30
Nehrbass, K., 2011. Dealing with Disaster: Critical Contextualization of Misfortune
in an Animistic Setting: Dealing with Disaster. Missiology 39, 459–471. doi:10.
1177/009182961103900404.
Nettle, D., Harper, Z., Kidson, A., Stone, R., Penton-Voak, I.S., Bateson, M., 2013.
The watching eyes effect in the Dictator Game: It’s not how much you give,
it’s being seen to give something. Evolution and Human Behavior 34, 35–40.
doi:10.1016/j.evolhumbehav.2012.08.004.
Norenzayan, A., 2013. Big gods: How religion transformed cooperation and conflict.
Princeton University Press.
Norenzayan, A., Shariff, A.F., Gervais, W.M., Willard, A.K., McNamara, R.A.,
Slingerland, E., Henrich, J., 2016. The cultural evolution of prosocial religions.
Behavioral and brain sciences 39.
Northover, S.B., Pedersen, W.C., Cohen, A.B., Andrews, P.W., 2017a. Artificial
surveillance cues do not increase generosity: Two meta-analyses. Evolution and
Human Behavior 38, 144–153. doi:10.1016/j.evolhumbehav.2016.07.001.
Northover, S.B., Pedersen, W.C., Cohen, A.B., Andrews, P.W., 2017b. Effect of
artificial surveillance cues on reported moral judgment: Experimental failures to
replicate and two meta-analyses. Evolution and Human Behavior 38, 561–571.
doi:10.1016/j.evolhumbehav.2016.12.003.
Ostrom, E., 2009. A General Framework for Analyzing Sustainability of Social-
Ecological Systems. Science 325, 419–422. doi:10.1126/science.1172133.
Peoples, H.C., Marlowe, F.W., 2012. Subsistence and the evolution of religion.
Human Nature 23, 253–269.
Piazza, J., Bering, J.M., Ingram, G., 2011. “Princess Alice is watching you”: Chil-
dren’s belief in an invisible person inhibits cheating. Journal of Experimental Child
Psychology 109, 311 – 320. doi:http://dx.doi.org/10.1016/j.jecp.2011.02.
003.
Purzycki, B.G., 2010. Spirit masters, ritual cairns, and the adaptive religious system
in tyva. Sibirica 9, 21–47.
Purzycki, B.G., 2011. Tyvan cher eezi and the socioecological constraints of super-
natural agents’ minds. Religion, Brain & Behavior 1, 31–45.
31
Purzycki, B.G., 2013. The minds of gods: A comparative study of supernat-
ural agency. Cognition 129, 163 – 179. doi:http://dx.doi.org/10.1016/j.
cognition.2013.06.010.
Purzycki, B.G., 2016. The evolution of gods’ minds in the Tyva Republic. Current
Anthropology 57, S88–S104. doi:10.1086/685729.
Purzycki, B.G., Apicella, C., Atkinson, Q.D., Cohen, E., McNamara, R.A., Willard,
A.K., Xygalatas, D., Norenzayan, A., Henrich, J., 2016a. Cross-cultural dataset
for the evolution of religion and morality project. Scientific Data 3, 160099. doi:10.
1038/sdata.2016.99.
Purzycki, B.G., Apicella, C., Atkinson, Q.D., Cohen, E., McNamara, R.A., Willard,
A.K., Xygalatas, D., Norenzayan, A., Henrich, J., 2016b. Moralistic gods, super-
natural punishment and the expansion of human sociality. Nature 530, 327–330.
doi:10.1038/nature16980.
Purzycki, B.G., Arakchaa, T., 2013. Ritual Behavior and Trust in the Tyva Republic.
Current Anthropology 54, 381–388. doi:10.1086/670526.
Purzycki, B.G., Finkel, D.N., Shaver, J., Wales, N., Cohen, A.B., Sosis, R., 2012.
What does god know? supernatural agents’ access to socially strategic and non-
strategic information. Cognitive Science 36, 846–869. doi:10.1111/j.1551-6709.
2012.01242.x.
Purzycki, B.G., Henrich, J., Apicella, C., Atkinson, Q.D., Baimel, A., Cohen, E.,
McNamara, R.A., Willard, A.K., Xygalatas, D., Norenzayan, A., 2018a. The
evolution of religion and morality: a synthesis of ethnographic and experimental
evidence from eight societies. Religion, Brain & Behavior 8, 101–132.
Purzycki, B.G., Holland, E.C., 2019. Buddha as a God: An Empirical Assess-
ment. Method & Theory in the Study of Religion 31, 347–375. doi:10.1163/
15700682-12341453.
Purzycki, B.G., Jamieson-Lane, A., 2016. Anthrotools: An R package for cross-
cultural ethnographic data analysis. Cross-Cultural Research 51, 51–74.
Purzycki, B.G., Lang, M., Henrich, J., Norenzayan, A., 2022a. The evolution of
religion and morality: Reflections and looking ahead. Religion, Brain & Behavior
X.
32
Purzycki, B.G., McNamara, R.A., 2016. An ecological theory of gods’ minds, in:
De Cruz, H., Nichols, R. (Eds.), Cognitive Science of Religion and Its Philosophical
Implications. Continuum, New York, pp. 143–167.
Purzycki, B.G., Pisor, A., Apicella, C., Atkinson, Q.D., Cohen, E., Henrich, J.,
McElreath, R., McNamara, R.A., Norenzayan, A., Willard, A.K., Xygalatas, D.,
2018b. The cognitive and cultural foundations of moral behavior. Evolution and
Human Behavior 39, 490–501.
Purzycki, B.G., Sosis, R., 2011. Our gods: Variation in supernatural minds, in:
Essential building blocks of human nature. Springer, pp. 77–93.
Purzycki, B.G., Sosis, R., 2022. Religion Evolving: The Dynamics of Culture, Cog-
nition and Ecology. Equinox.
Purzycki, B.G., Stagnaro, M.N., Sasaki, J., 2020. Breaches of Trust Change the
Content and Structure of Religious Appeals. Journal for the Study of Religion,
Nature and Culture 14, 71–94. doi:10.1558/jsrnc.38786.
Purzycki, B.G., Watts, J., 2018. Reinvigorating the comparative, cooperative ethno-
graphic sciences of religion. Free Inquiry 38, 26–29.
Purzycki, B.G., Willard, A.K., Klocov´a, E.K., Apicella, C., Atkinson, Q., Bolyanatz,
A., Cohen, E., Handley, C., Henrich, J., Lang, M., Lesorogol, C., Mathew, S.,
McNamara, R.A., Moya, C., Norenzayan, A., Placek, C., Soler, M., Weigel, J.,
Xygalatas, D., Ross, C.T., 2022b. The Moralization Bias of Gods’ Minds: A
Cross-Cultural Test. Religion, Brain and Behavior , 25.
R Core Team, 2021. R: A Language and Environment for Statistical Computing. R
Foundation for Statistical Computing. Vienna, Austria.
Rappaport, R.A., 1968. Pigs for the Ancestors: Ritual in the Ecology of a New
Guinea People. Yale University Press.
Rappaport, R.A., 1994. On the evolution of morality and religion: a response to lee
cronk. Zygon 29, 331–349.
Reynolds, V., Tanner, R., 1995. The Social Ecology of Religion. 2. ed., Oxford
University Press, New York.
Richerson, P., Baldini, R., Bell, A.V., Demps, K., Frost, K., Hillis, V., Mathew,
S., Newton, E.K., Naar, N., Newson, L., et al., 2016. Cultural group selection
33
plays an essential role in explaining human cooperation: A sketch of the evidence.
Behavioral and Brain Sciences 39. doi:10.1017/S0140525X1400106X.
Richerson, P.J., Boyd, R., 2005. Not by genes alone: How culture transformed
human evolution. University of Chicago Press Chicago.
Roes, F.L., Raymond, M., 2003. Belief in moralizing gods. Evolution and Human
Behavior 24, 126–135. doi:10.1016/S1090-5138(02)00134-4.
Rogers, A.R., 2020. Beating your neighbor to the berry patch. bioRxiv doi:10.1101/
2020.11.12.380311.
Rossano, M., 2010. Supernatural selection: How religion evolved. Oxford University
Press.
Rossano, M., LeBlanc, A., 2017. Why add the supernatural? Religion, Brain &
Behavior 7, 375–377.
Rossano, M.J., 2007. Supernaturalizing Social Life: Religion and the Evo-
lution of Human Cooperation. Human Nature 18, 272–294. doi:10.1007/
s12110-007-9002-4.
Schloss, J.P., Murray, M.J., 2011. Evolutionary accounts of belief in supernatural
punishment: a critical review. Religion, Brain & Behavior 1, 46–99.
Shariff, A., Purzycki, B.G., Sosis, R., 2014. Religions as Cultural Solutions to Social
Living, in: Cohen, A.B. (Ed.), Culture Reexamined: Broadening Our Under-
standing of Social and Evolutionary Influences. APA Books, Washington, DC, pp.
217–238.
Shaver, J.H., 2015. The evolution of stratification in Fijian ritual participation.
Religion, Brain & Behavior 5, 101–117. doi:10.1080/2153599X.2014.893253.
Shaver, J.H., Fraser, G., Bulbulia, J., 2017. Charismatic signaling: How religion
stabilizes cooperation and entrenches inequality. The Oxford Handbook of evolu-
tionary psychology and religion .
Shaver, J.H., Sosis, R., 2014. How Does Male Ritual Behavior Vary Across the
Lifespan? Human Nature 25, 136–160. doi:10.1007/s12110-014-9191-6.
Shih, F.L., 2012. Generating power in Taiwan: Nuclear, political and religious power.
Culture and Religion 13, 295–313. doi:10.1080/14755610.2012.706229.
34
Singh, M., Kaptchuk, T.J., Henrich, J., 2021. Small gods, rituals, and cooperation:
The mentawai water spirit sikameinan. Evolution and Human Behavior 42, 61–72.
doi:https://doi.org/10.1016/j.evolhumbehav.2020.07.008.
Skoggard, I., Ember, C.R., Pitek, E., Jackson, J.C., Carolus, C., 2020. Resource
Stress Predicts Changes in Religious Belief and Increases in Sharing Behavior.
Human Nature 31, 249–271. doi:10.1007/s12110-020-09371-8.
Smaldino, P.E., Lubell, M., 2014. Institutions and cooperation in an ecology of
games. Artificial Life 20, 207–221. doi:10.1162/ARTL_a_00126.
Smith, E.A., Wishnie, M., 2000. Conservation and Subsistence in Small-Scale So-
cieties. Annual Review of Anthropology 29, 493–524. doi:10.1146/annurev.
anthro.29.1.493.
Smith, J.J., 1993. Using ANTHOPAC 3.5 and a Spreadsheet to Compute a Free-List
Salience Index. CAM 5, 1–3. doi:10.1177/1525822X9300500301.
Smith, J.J., Borgatti, S.P., 1997. Salience Counts—And So Does Accuracy: Cor-
recting and Updating a Measure for Free-List-Item Salience. Journal of Linguistic
Anthropology 7, 208–209. doi:10.1525/jlin.1997.7.2.208.
Snarey, J., 1996. The natural environment’s impact upon religious ethics: A cross-
cultural study. Journal for the Scientific Study of Religion , 85–96.
Sosis, R., Bressler, E.R., 2003. Cooperation and commune longevity: A test of the
costly signaling theory of religion. Cross-cultural research 37, 211–239.
Spiro, M.E., D’Andrade, R.G., 1958. A cross-cultural study of some supernatural
beliefs. American Anthropologist 60, 456–466.
Sponsel, L.E., 2014. Spiritual ecology: Is it the ultimate solution for the environ-
mental crisis? Choice 51, 1339–13448.
Sørensen, J., 2004. Religion, evolution and an immunology of cultural systems.
Evolution and Cognition 10, 61–73.
Stagnaro, M.N., Stibbard-Hawkes, D.N.E., Apicella, C.L., 2022. Do religious and
market-based institutions promote cooperation in hadza hunter-gatherers? .
Strassmann, B.I., Kurapati, N.T., Hug, B.F., Burke, E.E., Gillespie, B.W., Karafet,
T.M., Hammer, M.F., 2012. Religion as a means to assure paternity. Proceed-
ings of the National Academy of Sciences 109, 9781–9785. doi:10.1073/pnas.
1110442109.
35
Sussman, L.K., 1981. Unity in diversity in a polyethnic society: The maintenance
of medical pluralism on mauritius. Social Science and Medicine. Part B: Medi-
cal Anthropology 15, 247 – 260. doi:https://doi.org/10.1016/0160-7987(81)
90051-X.
Swanson, G.E., 1960. The birth of the gods: The origin of primitive beliefs. Univer-
sity of Michigan Press.
Teehan, J., 2016. Religion and Morality: The Evolution of the Cognitive Nexus,
in: Liddle, J.R., Shackelford, T.K. (Eds.), The Oxford Handbook of Evolutionary
Psychology and Religion.
Tomlinson, M., 2004. Perpetual Lament: Kava-Drinking, Christianity and Sen-
sations of Historical Decline in Fiji. The Journal of the Royal Anthropological
Institute 10, 653–673.
Tomlinson, M., 2007. Everything and Its Opposite: Kava Drinking in Fiji. Anthro-
pological Quarterly 80, 1065–1081. doi:10.1353/anq.2007.0054.
Turner, J.W., 2012. Listening to the ancestors: Kava and the Lapita Peoples. Eth-
nology 51, 31–53.
Tylor, E.B., 1920. Primitive culture: Researches into the development of mythology,
philosophy, religion, language, art, and custom. London: Murray. URL: http:
//archive.org/details/primitiveculture01tylouoft.
Watts, J., Greenhill, S.J., Atkinson, Q.D., Currie, T.E., Bulbulia, J., Gray, R.D.,
2015. Broad supernatural punishment but not moralizing high gods precede the
evolution of political complexity in Austronesia. Proceedings of the Royal Society
of London B: Biological Sciences 282, 20142556. doi:10.1098/rspb.2014.2556.
Willard, A.K., Baimel, A., Turpin, H., Jong, J., Whitehouse, H., 2020. Re-
warding the good and punishing the bad: The role of karma and afterlife be-
liefs in shaping moral norms. Evolution and Human Behavior 41, 385–396.
doi:10.1016/j.evolhumbehav.2020.07.001.
Wright, R., 2010. The evolution of God: The origins of our beliefs. Hachette UK.
Xygalatas, D., Kotherov´a, S., Maˇno, P., Kundt, R., Cig´an, J., Klocov´a, E.K., Lang,
M., 2018. Big gods in small places: The random allocation game in mauritius.
Religion, Brain & Behavior 8, 243–261. doi:10.1080/2153599X.2016.1267033.
36
