Content uploaded by Manvir Singh
Author content
All content in this area was uploaded by Manvir Singh on Jul 11, 2020
Content may be subject to copyright.
This is an accepted manuscript. Cite as: Singh, M., Kaptchuk, T. J, & Henrich, J. (in press) Small
gods, rituals, and cooperation: The Mentawai water spirit Sikameinan. Evolution & Human Behavior.
Small gods, rituals, and cooperation
The Mentawai water spirit Sikameinan
Manvir Singh1*, Ted J. Kaptchuk2, and Joseph Henrich1
1Department of Human Evolutionary Biology, Harvard University
2 Program in Placebo Studies, Beth Israel Deaconess Medical Center, Harvard Medical School
*Corresponding author: manvirsingh@fas.harvard.edu
Abstract
Cognitive and evolutionary research has overwhelmingly focused on the powerful deities of
large-scale societies, yet little work has examined the smaller gods of animist traditions. Here, in
a study of the Mentawai water spirit Sikameinan (Siberut Island, Indonesia), we address three
questions: (1) Are smaller gods believed to enforce cooperation, especially compared to bigger
gods in larger-scale societies? (2) Do beliefs in these deities encourage people to engage in
behavior that would otherwise be perceived as costly? and (3) Does ritual reinforce beliefs in
these deities? Drawing on interview responses, data from healing ceremonies, and ethnographic
observation, we show that Sikameinan is believed to punish people who violate meat-sharing
norms and that people ‘attacked’ by Sikameinan pay shamans to conduct healing rituals. The
public nature of rituals, involving prestigious individuals apologizing to Sikameinan for the
patient’s stinginess, reinforce onlookers’ beliefs about Sikameinan. The most widely shared
beliefs about Sikameinan are represented in rituals while beliefs not represented vary
considerably, indicating that ritual may be potent for cultural transmission. These results suggest
that moralizing supernatural punishers may be more common than suspected and that the trend
in the cultural evolution of religion has been an expansion of deities’ scope, powers, and
monitoring abilities.
Keywords: animism; belief; cooperation; punishment; religion; ritual
p. 2
1. Three unresolved questions
Worshiped by the majority of living humans (Pew Research Center, 2017), punitive supernatural
agents are ubiquitous and important; they seem to promote cooperation beyond the provincial
circle regulated by kinship and reciprocity (Purzycki et al., 2016; Shariff, Willard, Andersen, &
Norenzayan, 2015); and they are associated with the emergence and maintenance of political
complexity around the world (Watts, Greenhill, et al., 2015). Yet despite evidence that beliefs in
these gods enabled humans’ unique levels of cooperation (Norenzayan, 2013; Norenzayan et al.,
2016), three basic questions remain unresolved:
First, are supernatural agents believed to enforce cooperation in small-scale societies,
especially compared to the supernatural forces of modern world religions? According to a
canonical anthropological view, the answer is “no”. Tylor (1874/1920, p. 360) endorsed this
position when he wrote that “savage animism is almost devoid of that ethical element which to
the educated modern mind is the very mainspring of practical religion.” Instead, in his appraisal,
people in such societies behave morally because of the pressure of “tradition and public opinion,
comparatively independent of the animistic belief and rites which exist beside them.” Recently
however, researchers have returned to the ethnographic record and begun to scrutinize it more
closely (Boehm, 2008; Purzycki & Sosis, 2019; Purzycki et al., 2020; Watts, Greenhill, et al.,
2015). Boehm (2008), for instance, coded the ethnographies of 18 forager societies and found
supernatural punishment in every society examined. Interestingly though, these punitive forces or
gods typically focused on only a few behaviors, often local taboos, and usually lacked the broad,
moralistic scope of supernatural punishers in modern world religions, such as karma or the
Christian god (White & Norenzayan, 2019). Better understanding the domains, nature, scope,
and intensity of supernatural punishment among hunter-gatherers and horticulturalists is crucial
not only for properly characterizing the cultural evolution of prosocial religion, but also for
understanding the forces supporting human cooperation in small-scale societies, especially given
humans’ exceptional levels of cooperation towards non-kin (Hill et al., 2011).
1
The second key question is: do people really believe in supernatural punishment? Just as
people believe that invisible viruses can endanger them and thus buy hand sanitizer to protect
themselves (Wiener-Bronner, 2020), do they hold mental representations of supernatural
punishers that encourage them to engage in behaviors that would otherwise be perceived as
costly? Experimental work suggests that, for believers in world religions, the answer is “yes”.
Across diverse societies, participants who rated gods as more punitive and knowledgeable were
more cooperative with distant co-religionists in economic games (Lang et al., 2019; Purzycki et
al., 2016). Moreover, a meta-analysis of 25 experiments showed that whereas religious priming
induced prosocial behavior among religious participants, it failed to affect non-religious
participants, indicating that the effects of priming depended on participants’ beliefs (Shariff et
1
Although recent work on the cultural evolution of religion has focused on the role of supernatural punishment in
supporting behaviors favorable to the scaling up of human societies (often the meaning of “morality”), this work has
not seriously engaged with the question of how much supernatural punishment exists in small-scale societies or on
specifically which behaviors are punished (Norenzayan, 2013; Peoples & Marlowe, 2012; Roes & Raymond, 2003).
Norenzayan et. al. (2016), for example, write “Whereas there is little dispute that foraging societies possess beliefs in
supernatural agents, these spirits and deities are quite different from those of world religions, with only limited
powers and circumscribed concerns about human morality.”
p. 3
al., 2015). In lab settings, supernatural belief seems to motivate people to engage in behavior that
would otherwise be considered costly.
Despite this growing evidence, many researchers remain skeptical (Baumard & Boyer,
2013; Boyer, 2018, 2019). They point to evidence showing that religious beliefs affect “reflective”
beliefs (those that participants report) but not “intuitive” beliefs (those that affect decision-
making) (Barrett, 2000). Moreover, most existing empirical evidence comes from experimental
rather than naturalistic data, leaving it unclear to what extent beliefs in supernatural punishment
affect real-world behavior (although see Shariff and Rhemtulla, 2012; Barro and McCleary,
2006; Edelman, 2009). These gaps are even more pronounced given that almost all
aforementioned studies have been conducted with participants in large-scale societies, with little
research investigating whether supernatural belief affects naturalistic behavior in small-scale
settings.
The final unresolved question is: does observing or participating in ritual reinforce
religious belief? Does, for instance, watching a healing ritual promote the belief that a particular
supernatural agent causes illness? Scientists have investigated many social effects of ritual,
including enabling coordination (Chwe, 2001), transmitting norms (Rossano, 2012), enhancing
group identification (Clingingsmith, Khwaja, & Kremer, 2009; Whitehouse & Lanman, 2014),
and maintaining political hierarchy through demonstrations of power (Watts, Sheehan,
Atkinson, Bulbulia, & Gray, 2016). Researchers have also described how people’s social contexts
and cognitive architecture contribute to religious belief (Boyer, 2001; Gervais & Henrich, 2010;
Lanman & Buhrmester, 2017). But these lines of research remain disconnected. With the
exception of research on children’s belief (Kapitány, Nelson, Burdett, & Goldstein, 2019;
Woolley, Boerger, & Markman, 2004), very little empirical work has examined the interaction
between belief and ritual (although see Barth, 1975). This gap is all the more striking given that
some scholars hypothesize that ritual is a critical mechanism for transmitting religious belief
(Henrich, 2009; Sosis, 2006; Whitehouse, 1995, 2002, 2004). Without ritual, some propose, the
transmission of such beliefs would be impeded by cognitive adaptations for protecting against
misinformation.
These questions remain unanswered partly because of a paucity of targeted data on
religious practice in small-scale societies. To help fill these gaps, we here present data on a
punitive small god: the Mentawai people’s water spirit, Sikameinan. As a belief system that
includes shamanism, ancestor spirits, and an animist worldview (Loeb, 1929b, 1929c; Schefold,
1988; Singh & Henrich, 2020), Mentawai religion shares features with the traditional religions
of many small-scale societies, including those of hunter-gatherers (Boyer, 2019; Peoples, Duda,
& Marlowe, 2016; Rossano, 2007; Singh, 2018).
We learned about Sikameinan after M.S. first began conducting fieldwork in Mentawai
in 2014. Through informal interviews in the Sarereiket region and Schefold's (1982, 1988)
reports about the Sakuddei clan (which originally migrated from Sarereiket), we learned that
people claim that a water spirit attacks individuals who refuse to share and that, in response,
shamans remove the water spirit from people’s houses. These findings inspired the much more
comprehensive investigation reported here. Synthesizing interview responses across four regions,
behavioral data on healing ceremonies, and primary ethnographic observation, we address the
three questions just outlined while presenting a rich ethnographic case study of a punitive small
god in an animist tradition.
p. 4
2. Ethnographic context
2.1. General background
The Siberut Mentawai live in the river valleys of Siberut Island (Indonesia), the largest island of
the Mentawai Archipelago (4,030 km2; about 150 km west of Sumatra) (Tulius, 2012) (Figure
1). The island is covered by at least 11 major rivers, each branching into dozens of smaller
waterways. Given the distance between rivers and a history of headhunting, people infrequently
traveled to other rivers, driving cultural differences among people living in distinct river valleys
(Schefold, 2007). We refer to the set of communities who speak a common dialect and live in
the same river valley as a cultural region. The research reported here draws on interviews and
observations conducted in four cultural regions of southern Siberut: (1) Sabirut (villages of
Maileppet and Muntei), (2) Sarereiket (villages of Buttui and Ugai), (3) Silaoinan (village of
Salappa), and (4) Taileleu (village of Taileleu) (Figure 1).
Figure 1. Siberut Island, the largest island of the Mentawai Archipelago (Indonesia). The
different study sites are colored. The legend shows the villages studied with the cultural region in
parentheses. Indonesia is light gray in the inset, while other countries are dark gray.
Buttui-Ugai (Sarereiket)
Muntei-Maileppet (Sabirut)
Salappa (Silaoinan)
Taileleu (Taileleu)
Rivers 0 3 612 18 24
p. 5
The Mentawai are organized into patrilineal clans (known as uma), traditionally residing
either in longhouses (also known as uma) or small houses nearby (Schefold, 1988; Tulius, 2012).
Following settlement programs by the government, most people shifted to villages (barasi),
which host schools, mosques, churches, and clinics. Settlement villages alter residence patterns
by placing families in close proximity and positioning clans much closer than they traditionally
lived. Nevertheless, many families maintain several residences, traveling between a house in the
forest, where they tend to pigs and conduct ceremonies, and a house in the settlement village.
Some families, typically those of shamans, spend most or all of their time in these forest
residences. Families with school-age children, meanwhile, tend to remain in settlement villages.
The Mentawai living on Siberut are rainforest horticulturalists. The staple of their diet is
starch extracted from the sago palm, but they also cultivate taro, cassava, and a wide assortment
of fruit trees, including coconuts, bananas, durian, rambutan, and mangos. They raise pigs and
chickens, as well as ducks, geese, cows, and buffalo, although those in the latter category are
recent introductions and, with the exception of ducks, rarely owned by Mentawai people. They
supplement their diet with foraged foods, such as deer, ferns, shrimp, frogs, primates, wild pigs,
and sago grubs.
Market foods are ubiquitous on the island and, in the case of rice, noodles, sugar, tea, and
coffee, highly preferred. People seek out wage labor opportunities to acquire these goods, pay for
school fees, and build larger houses, among other ends. In the villages of Ugai and Buttui, where
M.S. has to date conducted 11 months of fieldwork, the main means of acquiring cash, in
descending frequency, include: selling agricultural products (e.g., bananas, cacao), working on
development projects (e.g., road-building), finding opportunities off-island (e.g., in Sumatra),
and, most enviably, working as a government official or teacher. Several families (~5) also host
tourists interested in exotic homestays (“tribal tours”). Still, hunting and subsistence farming—in
particular, sago horticulture—serve as the basis for the Mentawai people’s livelihoods, especially
for people living in the island’s interior. As of 2011, 75% of individuals in Matotonan (a large
village in Sarereiket) received a quarter or less of their food and drink from the market, while
94% reported that they would not be able to survive without sago (Indigenous Education
Foundation, 2012).
2.2. Sharing meat
As will become clear, the domain of cooperation most relevant to the punitive spirit Sikameinan
is meat sharing. Meat sharing is regulated by a set of norms (Schefold, 1982). People are
expected to share meat in at least four circumstances: (1) when inviting help in a collaborative
endeavor, such as building a house, making a canoe, or dragging a canoe down from the forest to
the village (Figure 2A); (2) when sharing the spoils of a collaborative hunt (Figure 2B); (3) when
hosting a ceremony, such as a healing ceremony or after building a house (Figure 2C); and (4)
after receiving a large package of meat, such as from the three previous contexts as well as from a
bride payment, a penalty for a crime, or a solitary hunt (Figure 2D). With some exceptions,
people are expected to distribute portions that are equal, not only in size but also in the kind of
meat, such as fat, lungs, and skeletal muscle, included in each portion (Figure 2).
Meat sharing is, to some extent, altruistic. In the first three contexts just listed, sharing is
enforced through mechanisms such as reputation or reciprocity, because other people know
p. 6
whether a person shared. But the fourth context—sharing large packets—is vulnerable to
cheating. People can take advantage of norms about sharing with clan members and close
friends. They can hide their own prizes while accepting the gifts of others. Or they can lie about
how much they received, handing out smaller or fewer shares. In many cases, meat sharing
subject to exploitation.
Figure 2. Four contexts of meat sharing (photographs taken in 2017; credit M. Singh). (A)
People who request help in a collaborative effort compensate helpers with food, including meat.
This picture shows individuals divvying up portions of taro and pig to share with men who
helped drag canoes from the forest, where they were made, to the settlement village. (B) People
who cooperatively hunt share the rewards. This pictures shows the meat of a hunted boar divided
among all of the families who helped hunt and butcher it in the forest. (C) People share meat
during ceremonies. This picture was taken during a ceremony for inaugurating a new house. It
shows portions of meat for both attendees and sharing partners who could not attend. (D)
People share meat when they receive a large packet, including from any of the previous three
contexts as well as from bride payments, penalties of crimes, and solitary hunts. This picture
shows a man posing with several monkeys he had just hunted. Later, he divided up the meat to
share.
2.3. Religion in Mentawai
p. 7
The indigenous Mentawai religious system is known as Arat Sabulungan. Arat Sabulungan is
animist: Every thing is believed have a soul (Delfi, 2013; Loeb, 1929b; Schefold, 1988). Pigs,
plants, wild deer, tiny rats, new canoe motors, even the different materials composing a house –
all are animated (although not necessarily alive) (Reeves, 2000). All have spiritual essences that
can be inadvertently disturbed, and which must be appeased to protect against misfortune.
Humans have souls, too, and a person is said to fall ill when their soul wanders too far from their
body. A person can also purportedly become ill if they (1) are attacked by sorcery, (2) violate
taboos, (3) inadvertently encounter spirits in the forest or while bathing, (4) are scratched by the
forest spirit Silakikio, (5) are attacked by the water spirit Sikameinan, or (6) are inadvertently hit
by magic that has been prepared for non-malicious purposes. People are also said to fall ill if they
eat irregularly or if a terrible, deadly illness sweeps through a community.
The traditional healers in Mentawai are sikerei (shamans), a class of men set apart by their
ability to see spirits (Loeb, 1929c; Singh & Henrich, 2020). As healers, sikerei are experts in
herbal medicine and the special songs used for communicating with souls and spirits. A man
hoping to become a sikerei must host a series of ceremonies, find another sikerei to teach him
songs and herbal medicines, and have his eyes magically treated. In some cultural regions, sikerei
are marked by their continued use of the loincloth and their full-body tattoos.
Sikerei treat ailments in healing ceremonies called pabetei. The family hosting the
ceremony sacrifices pigs or chickens, which are shared with the sikerei (as a kind of payment)
and the patient’s kin. Healing ceremonies are like doctor’s appointments: All of them share
common elements, but each is also geared to target particular ailments. There are, for example,
special treatments for souls that have wandered too far, illness caused by Silakikio, and removing
Sikameinan for a patient’s house. The intervention regarded as most paradigmatic of the sikerei
institution is lajo simagre, an all-night treatment during which sikerei dance and summon
beneficent spirits. As these spirits approach the dancing shamans, they enter trance (gobok).
Governmental programs and missionaries have both affected the religious landscape of
Mentawai. Arat Sabulungan was officially forbidden in 1954 as a part of the West Sumatran
government’s policy of modernization (Bakker, 1999; Schefold, 1998). In the 1960s, the
Indonesian government identified the Mentawai as suku terasing, an “isolated ethnic group” that
had to be brought into the fold of the Indonesian nation. By the 1970s, the state began concerted
efforts at acculturation in the island’s interior. Shamanism, tattooing, teeth-sharpening, and
other indigenous practices were prohibited. Longhouses were demolished. Partly as a result of
these efforts, shamanism and other traditional elements declined in the Mentawai islands. Yet
they survived in interior areas of Siberut, likely due to the resistance of local clans, people fleeing
to ever-more-remote areas, and the difficulty of encouraging officials to stay in villages and
enforce the rules (Bakker, 1999; Hammons, 2010). After 1987, the government eased off the
acculturation efforts, allowing indigenous religious practice to continue largely unmolested
(Bakker, 1999).
The German Royal Missionary Society (Protestants) established the first mission station
in the Mentawai Archipelago in Sikakap, on the island of North Pagai (Hammons, 2010). The
first missionary took up residence in 1901, and the first conversion occurred in 1915. In 1932,
the first Protestant mission station was established on Siberut, in Maileppet (Sabirut region).
Following World War II, Catholic Italian missionaries established a station on the island, also in
p. 8
Sabirut, in Muara Siberut
2
. But Christian missionaries were slow to reach many parts of the
interior. The first outside religion to arrive in Buttui and Salappa was not Christianity (or Islam,
which has taken the longest to reach the interior), but Baha’i, which came to Siberut in the mid-
1950s (see Rudito, 2001). Catholicism eventually arrived in Both Buttui and Salappa (dates are
uncertain), but, at least in the case of Buttui, it was permissive of indigenous religion, likely
because these communities actively resisted acculturation (the people of Buttui avoided the
government settlement program until the early 2010s, when the barasi was built there). For
example, when the Catholic priest who worked in Buttui established a small jungle school for
children to learn to read and write, he named it Parurukat Toga Sikerei Butui, or “Meeting of the
Buttui Shamans’ Children”.
Islam has spread more slowly in Siberut. When, in 1954, government officials gave
groups of Mentawai people the option to choose between Protestantism and Islam, most
nominally chose Protestantism because it permitted them to eat pigs (Bakker, 1999). More
recently, Islam has targeted areas that have otherwise experienced less missionary activity. It
arrived in Buttui around 2012 with the development organization Yayasan Aksi Peduli Bangsa
(roughly, “Foundation for Nation Caring Action”) (Aksi Peduli Bangsa, 2020), coinciding with
the departure of the Catholic priest and the disappearance of his jungle school. The development
organization built a mosque, housing for teachers, a small shop, a cement football court, and a
single-room schoolhouse that, since the 2016 construction of a primary school (Sekolah Dasar) in
Buttui, has served as a kindergarten (Taman Kanak-Kanak). Several families in the clans
Salakkirat and Sabettiliake converted when the organization first arrived, likely to receive gifts
and allow their children to attend school. In 2017, while M.S. was conducting fieldwork, more
families in Salakkirat converted, as well as families in the clans Tasirikeru and Satairarak. Many
of the families that converted were those that continued to live in the surrounding forest. The
same development organization seems to also have spread into Salappa in the region Silaoinan.
This discussion of how missionaries and the state have affected the Mentawai religious
landscape demonstrates just how varied the four regions’ experiences have been. The differential
effects of these programs are clear when comparing villages in Sabirut (a coastal region where
enforcement and missionizing were easier) to those upriver in Sarereiket (a region in the
interior). In 2017, M.S. counted fewer than 5 shamans across two Sabirut villages (Maileppet,
Muntei), while he has found at least 90 shamans across four villages in Sareireket (Rogdog,
Madobag, Ugai, Buttui). Studying regions with such varied experiences is useful because it allows
us to determine to what extent our results may be influenced by contact with the state and
Abrahamic religions. If we identify similar beliefs across four regions that vary both in the kind
of outside exposure (e.g., Islam, Catholicism, Protestantism, Baha’i) and its degree, we can more
confidently conclude that those beliefs predated those forces. We return to this point in the
discussion.
3. Data collection
Our report draws on three sources of data: ethnographic interviews about Sikameinan conducted
with 96 participants across four cultural regions of southern Siberut; systematic data covering 66
2
Siberut is the name of the island, Sabirut is the name of a region and dialect in South Siberut, and Muara Siberut
is the administrative center of South Siberut.
p. 9
healing ceremonies, most of which occurred in the Buttui-Ugai region of Siberut in 2017; and
participant observation occurring over 11 months in the Buttui-Ugai region.
The Harvard University Committee on the Use of Human Subjects approved this
research project. All participants provided verbal informed consent before the study. All analyses
and cleaned data are available online on the OSF project website online at https://osf.io/bjq6f/.
3.1. Sikameinan interviews
M.S. interviewed participants about Sikameinan in 2017 as part of a larger cross-regional study
of cultural beliefs in southern Siberut. All interviews were conducted in the Mentawai language
by M.S. Interviews were conducted in the villages of Buttui and Ugai (Sarereiket) (n = 40),
Muntei and Maileppet (Sabirut) (n = 19), Salappa (Silaoinan) (n = 21), and Taileleu (Taileleu) (n
= 16) (Figure 1). We aimed to interview those participants living in the same village in as few
days as possible and visited different sections of each village on different days. We selected
participants opportunistically and made efforts to both (1) interview participants privately and
out of earshot (such as in the participant’s house or a small building to which participants were
invited) and (2) prevent participants who had completed the interview from discussing it with
other individuals (such as by having a research assistant sit with participants as they waited to be
interviewed).
Interviews consisted of demographic variables (name, age, patrilineal clan, village of
residence, village of origin, spouse’s clan, religion, and years of schooling), as well as surveys on
shaman taboos and Sikameinan. M.S. prepared the Mentawai translation of the interview script
with the assistance of a native speaker. With the exception of eight participants in Sarereiket, all
participants were asked the seven questions about Sikameinan in the same order:
1. Number –
English: According to you, how many spirits are there? One? Two? Many?
Mentawai: Menurut ekeu, piga simagrena e? Ale sabbek? Ale rua? Ale maiggi?
2. Sex –
English: According to you, is the spirit female? Male? Both?
Mentawai: Menurut ekeu, ale sinanalep ia simagrena e? Ale simateu? Ale ka duada?
3. Location –
English: Normally, where does the spirit live?
Mentawai: Biasania, kapa ilelek simagrena e?
4. Reason for attack –
English: Why does the spirit strike us?
Mentawai: Ponoili iorag ita simagrena e?
5. Method of attack –
English: When it strikes us, what does the spirit do?
Mentawai: Kana iorag ita, kipa ikut simagrena e?
6. Misfortune –
English: How is the injury/illness when it strikes us?
Mentawai: Kipa besitta kana iorag ita simagrena e?
7. Ritual objects –
English: When they do the meinan ritual, what objects do they put out?
Mentawai: Kana dakut meinan, ponia daudutnakke alaket e?
p. 10
Participants were afterwards invited to draw Sikameinan.
Responses to these questions were recorded during the interviews and later translated,
cleaned, and binned. M.S. later translated responses to English, receiving assistance from other
Mentawai people and, when necessary, other Mentawai experts (Maskota Delfi, Rob Henry,
Juniator Tulius). Supplementary Tables S1-S7 present participants’ responses for each of the
seven questions.
All analyses were conducted in R 3.4.3 (R Core Team, 2015). As a part of our analysis,
we calculated the agreement among respondents about a given question (represented here with a)
according to the following formula:
𝑎 =
∑
𝑝!𝑛!
"
!#$
∑
𝑛!
"
!#$
where ni is the number of respondents who gave the ith answer, pi is the proportion of
respondents who gave the ith answer, and s is the total number of types of answers. See the
supplemental materials for more details about his formula.
3.2. Healing ceremonies
M.S. interviewed the patients of healing ceremonies or their family members in 2017. M.S.
prepared the Mentawai interview script with the assistance of a native speaker. We either
interviewed patients after attending or having heard about a healing ceremony or asked people to
tell us about their or a family member’s most recent healing ceremony. All interviews were
conducted along the Rereiket River with most occurring in Ugai and Buttui. Interviews were
conducted in the Mentawai language.
Respondents were asked about the treatments in the healing ceremony, the number of
pigs and chickens sacrificed, the length of the ceremony, and which shamans provided care.
Participants were also asked about the illness, such as what the ailment was, how long it had
gone on, whether they had hosted a healing ceremony for it before, why they believed it to have
occurred, which other treatments they had pursued, and whether the ceremony had worked.
When possible, participants’ responses were verified in one or more of the following
ways: (1) M.S. attended the healing ceremony; (2) M.S. verified details with other community
members, preferably one of the sikerei said to administer the ceremony; or (3) M.S.
reinterviewed the participant or another family member. In total, 90 interviews were conducted
about 77 healing ceremonies and 75 bouts of illness (some healing ceremonies targeted several
bouts of illness, such as when siblings or spouses were healed, while some illness bouts were
healed in several successive ceremonies). Eleven healing ceremonies and 10 bouts of illness were
removed from analyses because no ceremony was actually conducted (n = 1), other observers
denied that a ceremony was conducted (n = 4), the participant admitted to not remembering
details clearly (n = 3), there were discrepancies when the participant was interviewed again (n =
2), or other parties disagreed about noteworthy details (n = 1).
Responses to the questions were recorded during the interviews. M.S. later translated
responses, receiving assistance from other Mentawai people or experts when necessary. All
analyses were conducted in R 3.4.3 (R Core Team, 2015).
p. 11
4. Sikameinan
Throughout southern Siberut, people describe a water-dwelling spirit that causes illness, is
appeased in shamanic healing ceremonies, and is often referred to as Sikameinan or meinan
(derived from kameinan, “aunt”). The spirit’s other names include Silapualai (derived from alai,
“hair”) and Sibeulepei (“that with big clothes”), although these names are more variable than
Sikameinan and some respondents regarded them as distinct spirits. In line with these names,
people often describe Sikameinan as having long hair and wearing big clothes.
Almost all participants drew a close connection between Sikameinan and crocodiles. In
fact, the spirit is often known by the word for crocodile, sikaoinan (“that which is in water”) (e.g.,
Schefold, 1982, 1988). Many participants considered Sikameinan to be the spirit of a crocodile
or to look like a crocodile (see Figure 3 for some illustrations). Those participants who
considered Sikameinan to be more human-like often considered the spirit to be a friend of
crocodiles, to order them to attack humans, or to use a crocodile as its “canoe” (abak) (see
Supplementary Table S8 for some participants’ descriptions).
Figure 3. Illustrations of Sikameinan by Mentawai people in the regions of Sabirut (SAB),
Sarereiket (SAR), Silaoinan (SIL), and Taileleu (TAI). Illustration TAI-2 distinguishes between
p. 12
the physical body of Sikameinan (a crocodile) and its soul (a long-haired figure with the torso
and limbs of a crocodile).
4.1. Sikameinan has moral concerns, but they are limited in scope and domain
Across four cultural regions of southern Siberut, participants near-unanimously reported that
Sikameinan attacks people who fail to share (Figure 4A; Supplementary Table S4). Moreover,
participants who specified which items must be shared almost exclusively mentioned meat,
although a tiny minority mentioned other items including money and kat (edible plant food)
(Figure 4B). We did not collect data on who food must be shared with, but food-sharing norms
in Mentawai dictate that people share with social intimates, especially clanmates, affines, and
neighbors who live nearby (Hammons, 2010; Schefold, 1982). Sikameinan has moralistic
concerns, but they are limited in domain (sharing meat) and scope (clanmates and other
relatives).
p. 13
Figure 4. The reasons Sikameinan attacks people, according to free responses (N = 96). Colors
correspond with cultural regions: red = Sabirut; yellow = Sareireket; green = Silaoinan; blue =
Taileleu. Panel A shows the proportion of respondents reporting different reasons that
Sikameinan attacks people. Panel B shows the proportion of respondents reporting different
domains of sharing that Sikameinan enforces.
Unprompted, respondents sometimes suggested that the fear of Sikameinan causes
people to share. One woman noted that the Mentawai have no need for foreign religions and
moralistic prescriptions because Arat Sabulungan (indigenous Mentawai religion) already deters
wrongdoing. The fear of sorcery prevents people from stealing, she noted, and Sikameinan
ensures that they share.
0.00
0.25
0.50
0.75
1.00
Chicken Cookies/
bread
Drink Kat Meat Money Necklace Pig Smoked
Meat
Tobacco
Domain of sharing that Sikameinan enforces
Proportion of respondents
B
0.00
0.25
0.50
0.75
1.00
Being asocial Cutting down trees Failing to share Lying Stealing
Behavior punished by Sikameinan
Proportion of respondents
A
p. 14
An older woman also observed the instrumental effect of Sikameinan: “That’s why we
give meat to other people; that’s why we give plant-food. We give them plant-food; we give
them meat. They see our food, we give it. [Otherwise, Sikameinan] carries us to the bottom [of
the river]. The body [of the crocodile] – not just a spirit. To the bottom it carries us.”
4.2. People suspect Sikameinan after many kinds of harms
People believe that Sikameinan watches them. When asked where the spirit lives, most
respondents said that Sikameinan lives in water, such as the river or the small puddles that collect
in broken bamboo (Figure 5; Supplementary Table S3). Several respondents reported that it
drifts through the air like wind; one person compared it to the regional government – “wherever
we go, it knows.” Other respondents said that it lives in people’s houses.
p. 15
Figure 5. Where Sikameinan lives, according to free responses (N = 96). Colors correspond with
cultural regions: red = Sabirut; yellow = Sareireket; green = Silaoinan; blue = Taileleu. Panel A
shows the proportion of respondents reporting different locations where Sikameinan lives,
aggregating responses about water. Panel B shows the breakdown of different responses about
water.
0.00
0.25
0.50
0.75
1.00
Everywhere Forest House Sick people Water
Where does Sikameinan live?
Proportion of respondents
A
0.00
0.25
0.50
0.75
1.00
Broken bamboo
(puddles)
Crocodile
body
Ocean Pond River Shore Water
(general)
Where in the water does Sikameinan live?
Proportion of respondents
B
p. 16
When Sikameinan sees someone refusing to share, it attacks. It enters its victim’s house
and climbs onto a cross beam. It radiates its illness-causing energy (bajou), scratches its victim, or
weighs down on top of them, making it hard to breathe. Alternatively, it might wait for the
stingy person or their family members in the river. When its target arrives, Sikameinan attacks
them as a crocodile, or it directs a crocodile to attack them, or it pulls them to the bottom and
drowns them (see Supplementary Figure S1 and Supplementary Table S5 for participants’
responses about how Sikameinan attacks; see Supplementary Figure S2 and Supplementary
Table S6 for participants’ responses about the kinds of illnesses and injuries Sikameinan causes).
People sometimes connect others’ misfortunes to their stinginess and Sikameinan’s
retribution. In 1988, Bakels was told about a crocodile that carried off a small boy (Bakels, 1994).
The animal’s violence was blamed on the boy’s mother’s stinginess. Similarly, when a boy
drowned in August 2017 in a community where M.S. was conducting research, people covertly
discussed possible reasons. The leading explanation that developed was that Sikameinan attacked
after the boy’s parents failed to share their growing wealth.
4.3. People pay costs for shamans to remove Sikameinan in a dramatic, public ritual
When it’s concluded that a patient’s illness is caused by Sikameinan, the sikerei conduct a special
ceremony, known most frequently as meinan, to remove it from the house and return it to the
river. The sikerei assemble in the patient’s house or the patient’s clan’s longhouse, usually with
the patient’s family. They put out objects that will attract Sikameinan, such as fabric and a
fishing net (Figure 6A; Supplementary Table S7). They publicly acknowledge that the patient or
their family did not share, but they promise Sikameinan that this was not intentional. The
patient and their family members apologize, assuring Sikameinan and any observers that they
will share in the future.
p. 17
Figure 6. Sikerei (Mentawai shamans) conduct ceremonies to remove Sikameinan from a
patient’s house. (A) A sikerei presents items to attract Sikameinan, including a necklace, fabric,
and a luat (shaman headband) (photo taken in 2019; credit/copyright Rob Henry, Indigenous
Education Foundation). (B) The sikerei invite Sikameinan into a small container of water (2019;
credit/copyright Rob Henry, Indigenous Educatoin Foundation). (C) Sikerei prepare to release
Sikameinan into the river (1978; credit/copyright Reimar Schefold).
The sikerei then use special songs to lure Sikameinan into a small basin of water (Figure
6B). These songs are considered powerful ways of inviting Sikameinan, and for that reason,
people are forbidden from singing them outside of the Sikameinan ceremony.
Once Sikameinan is captured, the sikerei carry it to the river. They hold the basin steady
to prevent Sikameinan from tumbling out. They bring the lavish gifts and a small food item,
such as smoked meat or piece of chicken. Finally, they release Sikameinan into the water (Figure
6C).
There are three noteworthy points about this ceremony:
1. The ceremony is costly. The costliness of healing ceremonies provides evidence that
people sincerely believe in Sikameinan. If they did not believe that Sikameinan had harmed
them, they would presumably be less motivated to incur costs for shamans to come and banish
the spirit.
p. 18
There are two ways in which ceremonies are costly. First, the family hosting the
ceremony must provide food for the shamans and other family members. This means preparing
taro, sago, sweetened beverages, and, most importantly, meat.
The consensus is that the “price” of a meinan ceremony is 1 or 2 chickens. Examining the
subset of healing ceremonies that included only the meinan ceremony (or the meinan ceremony
and a treatment for the spirit-illness kisei, which is said to require no additional sacrifice), we
found that the number of animals sacrificed varied from no chickens (when a shaman conducted
the ceremony privately for his sick child) to a large female pig and 2 chickens (a large female pig
is roughly equivalent in value to ten adult chickens) (see Table 1 for the list of ceremonies and
their accompanying sacrifices). Although the family also consumes the food, (1) they must share
it with more people than they would if they killed it in private, (2) they give up other
opportunities to use the animals (such as other household ceremonies or, in the case of pigs,
bride prices or penalties for crime), and (3) they give up the best portions of the meat to the
shamans.
Table 1. Prices of twelve healing ceremonies that either only included the ritual for removing
Sikameinan (S) or included that ritual and a treatment for kisei (K), which is said to require no
additional sacrifices. The PBID refers to the identifier in the dataset, available online. The
estimated value is listed in thousands of Indonesian Rupiah (IDR); on 1 July 2017, 1000 IDR
was equivalent to about 0.075 USD. Note that not all chickens are equivalent in value. For
PB03, the patient reported two sets of responses on two different occasions; both are reported in
the table.
PBID
Animals sacrificed
Estimated value
(thousands of IDR)
Interventions
PB30
1 large female pig and 2 chickens
1010
S
PB42
1 medium-sized pig and 2 chickens
660
S
PB66
1 medium-sized pig and 4 chickens
700
S
PB03
2 chickens // 1 rooster
160 // 100
S // S, K
PB39
2 chickens
160
S, K
PB40
2 small pigs
350
S, K
PB41
2 chickens
160
S, K
PB51
0 [shaman healed his sick child]
0
S, K
PB57
1 small pig, 1 chicken, and 1 duck
330
S, K
PB60
1 medium-sized pig
500
S, K
PB63
2 chickens and 1 duck
235
S, K
PB73
1 chicken
80
S, K
The second important way in which hosting the ceremony incurs costs is that it redirects
resources from other treatments. Of 62 bouts of illness for which we have information, 63%
(39/62) were also treated with herbal or magical remedies. For 50% of those illnesses (31/62), the
patient or their family also sought healthcare at a village health post (Poskesdes) or community
health center (Puskesmas). Obtaining herbal remedies and healthcare at a clinic incurs costs,
including the time involved in collecting herbs or the price of the medicine. Patients also visit
p. 19
alternative health providers, such as Muslim or Christian healers; these visits also consume time
and resources. By spending half a day organizing and paying for a Sikameinan ceremony, families
lose the opportunity to pursue other treatment options. People’s willingness to pay for
Sikameinan ceremonies, especially considering the opportunity costs, suggests that
representations of supernatural punishment motivate people to engage in behavior that would be
perceived as costly were they not to hold those beliefs.
2. Elements of the ceremony reinforce belief. Two aspects of the Sikameinan ceremony seem
to reinforce beliefs that Sikameinan is responsible for illness. First, hosting a healing ceremony
exhibits the features of a CRedibility-Enhancing Display (CRED) (Henrich, 2009). It is a
behavior that would only be sensible if a person held a given belief. In this case, the ceremony is
sufficiently costly that someone should host it only if they believe that Sikameinan may have
caused illness. The ceremony thus reliably indicates belief, increasing the likelihood that others
will adopt similar representations about Sikameinan.
The second element of the Sikameinan ceremony that should reinforce others’ beliefs is
that it features prestigious individuals speaking to the spirit. Sikerei are said to be uniquely
capable of seeing spirits, including Sikameinan. They are also trusted for their expertise,
especially in Mentawai religious practice (see Supplementary Study 1 and Supplementary Figure
S3), and are frequently called to intervene in conflicts, evidencing their status and people’s trust
in their decisions. The Sikameinan ceremony features these reputable men speaking to the spirit,
apologizing to it, coaxing it into a container, and then releasing it into the water. This
performance may further reinforce that Sikameinan is real and to be taken seriously.
3. The natural course of recovery reinforces belief. Eventually, most people recover from their
illness. The timing of their recovery influences their theories about which treatments were
effective and, in turn, what originally caused the illness. For example, a patient who recovers
soon after shamans remove Sikameinan from their house might attribute their improvement to
the ritual and thus their illness to Sikameinan’s attack. This seems to happen: Of families that
paid for a Sikameinan ritual, 11% later reported that Sikameinan was a reason the patient
became ill. In total, 7 of 65 illness bouts were thought to have been caused by Sikameinan after
the patient recovered.
4.4. People infer features of Sikameinan from ceremonies
Sikameinan ceremonies occur often. In our sample of 66 healing ceremonies, nearly all of which
occurred in a single year, 46 included a Sikameinan ceremony – more than any other special
intervention (Figure 7). Although it remains uncertain how often individuals attend the
Sikameinan ceremony, we counted 19 Sikameinan ceremonies in a community of 262
individuals. Assuming that each ceremony draws 20 observers, we estimate that an individual has
a 78% chance of attending a Sikameinan ceremony in a given year.
p. 20
Figure 7. The proportion of healing ceremonies featuring different treatments. Treatments used
infrequently (>0.05) have been removed.
Two lines of evidence suggest that these ceremonies transmit beliefs about Sikameinan:
First, in explaining their answers, respondents explicitly drew inferences from the
Sikameinan ceremony, such as that Sikameinan lives in water because shamans place it in water,
that it lives near the ceiling because shamans look there during ceremonies, and that it is female
because shamans refer to it (her) as “aunt”.
Second, of those features of Sikameinan we systematically studied, all of the most widely
shared beliefs are represented in the ritual or can be readily inferred. Figure 8 shows the extent to
which participants gave similar responses about the different features of Sikameinan, following
the formula described in section 3.1. We found that participants agreed most on why
Sikameinan attacks people (not sharing), followed by the number of Sikameinan (one) and the
objects used to appease it in ritual (fabric, fishing nets, and necklaces), all of which are
represented in the ritual. By contrast, elements of Sikameinan that are not present in the ritual
(how it attacks people and the kinds of misfortune it causes) showed much less agreement, both
within regions (Fig. 7A) and across them (Fig. 7B). To appreciate the variation in agreement,
contrast Figure 1A, which shows participants’ responses about why Sikameinan attacks (reason),
to Supplementary Figure S2, which shows participants’ responses for what happens when it
attacks (misfortune). Although we did not quantify this information, participants also varied
considerably in how they imagined Sikameinan’s appearance (Supplementary Table S8), another
feature not represented in the meinan ritual.
Pasiala ka ratei
for an illness caused when one’s
soul wanders in a graveyard
Kerei
for an illness caused when
sikerei violate taboos
Pasibele
for an illness caused by
disturbed or disrespected spirits
Lajo simagre
a trance-inducing dance for
calling spirit and good spirits
Pamopok
for an illness caused by
the forest spirit Silakikio
Kisei
for the spirit illness kisei
Meinan
for an illness caused by
Sikameinan
0.0 0.2 0.4 0.6 0.8 1.0
Proportion of healing ceremonies
Ceremonial treatment
p. 21
0.00
0.25
0.50
0.75
1.00
Number
Sex
Location
Reason
Method
Misfortune
Objects
Feature of Sikameinan
Agreement
B
0.00
0.25
0.50
0.75
1.00
Number
Sex
Location
Reason
Method
Misfortune
Objects
Feature of Sikameinan
Agreement
A
p. 22
Figure 8. Agreement among participants about different features of Sikameinan (see 3.1 for the
full questions). Error bars are bootstrapped 95% confidence intervals (10,000 samples). Colors
correspond with cultural regions: red = Sabirut; yellow = Sareireket; green = Silaoinan; blue =
Taileleu. Panel A shows agreement among respondents living in the same cultural region; Panel
B shows agreement among respondents pooling across cultural regions.
By this logic, we should expect higher agreement about Sikameinan’s sex (female) and
location (water). What explains the lower values? For responses about Sikameinan’s location, the
lower agreement is a result of people’s diverse responses. Still, the majority of responses referred
to the water (e.g., river, puddles inside broken bamboo), a belief again reinforced in ritual (see
Figure 5). For sex, the lower agreement partly reflects the high frequency with which participants
responded that they didn’t know (Supplementary Figure S4). Moreover, the participants who did
respond usually replied that Sikameinan is either female (29%) or both sexes (23%) with only 4%
of respondents claiming that Sikameinan is male (Supplementary Table S2). The bias towards
Sikameinan being female is what we expect on the basis of Sikameinan’s name (aunt) and the
objects used to appease it (e.g., the fishing net, which women typically use). This evidence
suggests that the Sikameinan ceremony is a potent vehicle for transmitting cultural information
and creating shared beliefs among community members.
5. Discussion
At the outset of this article, we asked three questions:
1. Are supernatural agents believed to enforce cooperation in small-scale societies and, if so,
how does their jurisdiction differ from that of bigger gods in more complex societies?
2. Do beliefs in these supernatural agents motivate people to engage in behavior that would
perceived as costly were they not to sincerely hold those beliefs?
3. Does ritual reinforce beliefs in these agents?
For the first question, our study found strong evidence that a supernatural agent in a
small-scale society is said to punish non-cooperative behavior. But in comparison with the
moralizing forces of world religions, Sikameinan’s scope and domain of interest are limited.
Whereas moralistic supernatural forces such as karma or the Christian god are believed to care
about large-scale morality, including whether one is greedy, harmful, unkind, or dishonest
towards socially distant individuals such as strangers (White & Norenzayan, 2019), Sikameinan
attacks people who fail to share meat with fellow clan members. Unlike karma or the Christian
god (White & Norenzayan, 2019), Sikameinan cannot read minds, which is implied in the
shamans’ assurance to the spirit that the patient’s failure to share was inadvertent. Sikameinan is
morally concerned but provincial in its scope and limited in its knowledge compared to the
powerful gods of the largest-scale societies (see also Tyvan spirit-masters: Purzycki, 2016;
Purzycki, 2013).
For the second question, our findings suggest that beliefs in supernatural punishment
motivate people to incur costs. Patients and their families pay for shamans to come and remove
Sikameinan from their homes, and they do so at the expense of other treatment options, such as
visiting a health clinic. Moreover, shamans call other shamans to perform the ceremony, and
shamans even perform the ceremony in private.
p. 23
An alternative explanation for these observations is that people’s representations of
Sikameinan do not motivate them to incur costs and instead families host the ceremony for
reputational benefits, perhaps because it broadcasts to observers that they fear Sikameinan and
can therefore be trusted. Aside from the fact that such a hypothesis still invokes erroneous beliefs
– observers would erroneously infer that the family holds Sikameinan beliefs that make them
trustworthy – readers should also be aware that conducting a Sikameinan ceremony is an
admission that one violated an important social norm. Future research should evaluate the
inferences observers make about families that host Sikameinan ceremonies and test whether
Sikameinan beliefs have other effects on behavior, such as motivating prosocial decision-making.
For the final question, our study showed that ritual and belief can mutually reinforce each
other. People’s belief that Sikameinan harmed them or their family motivates them to call
shamans to conduct the appropriate ceremony. The costly and public nature of the ceremony in
turn strengthens observers’ beliefs about Sikameinan, especially the idea that Sikameinan
punishes for not sharing. The patient’s recovery, such as through the spontaneous remission of
the illness or, possibly, placebo effects stimulated by ritual healing (Kaptchuk, 2011), further
convinces patients and observers that the ceremony worked and that Sikameinan was the
originating cause of the illness. The belief-producing effects of ritual manifest in population-level
variation in beliefs. Those beliefs that vary the least across individuals are represented in ritual,
while those beliefs not represented vary the most. Belief motivates ritual; ritual reproduces belief
(Figure 9; see also Henrich, 2009).
Figure 9. Belief in Sikameinan and rituals (healing ceremonies in which shamans remove it)
seem to mutually reinforce each other.
A single ethnographic case study need not necessarily generalize. That Sikameinan is
moralistic tells us little about the frequency of supernatural punishers in animist traditions more
broadly. Still, our findings are consistent with observations of scholars such as Boehm (2008) and
Watts, Greenhill et al. (2015). In their ethnographic surveys of small-scale societies, these
researchers found evidence of beliefs in spirits that care about cooperation. Crucially, they noted
BELIEF
Sikameinan punishes
stinginess with illness.
RITUAL
Shamans publicly apologize to
Sikameinan and remove it from the
house; patient later recovers.
p. 24
that those spirits seem to be less powerful and to have smaller moral jurisdictions. Our detailed
investigation of Sikameinan, combined with ethnographic surveys, substantially qualifies the
classical anthropological view that moralistic supernatural forces are restricted to modern, large-
scale societies.
To what extent did interactions with world religions in the twentieth century shape these
beliefs and practices? Three lines of evidence suggest that answer is “not very much”. First, the
four regions studied have experienced distinct histories of religious contact and acculturation, yet
they reported strikingly similar beliefs about Sikameinan. In the region of Sabirut, Protestant
missionaries arrived in the 1930s, followed soon by Islam and Catholicism. This region was also
dramatically affected by governmental programs bent on eradicating traditional Mentawai
culture. The people of Buttui, in contrast, were better able to escape early missionizing and
government acculturation. They first encountered Baha’i in the 1950s, followed later by
Catholicism and much more recently by Islam. All the while, they actively maintained traditional
practices. Salappa seems to have had a similar history to Buttui insofar as Baha’i arrived first
followed by Christianity, although Islam may have come there earlier. We know less about the
history of Taileleu. It seems unlikely that, despite variation in which world religions these
communities encountered and in the intensity of those interactions, the peoples in these
communities convergently constructed Sikameinan so that it specifically attacks people who do
not share meat.
The second reason why these beliefs seem to predate recent missionizing activity is that
they are consistent with those described by Schefold (1988). Starting in 1967, Schefold
conducted fieldwork with Sakuddei, a remote Mentawai clan that had evaded outside contact. If
missionaries appreciably affected these beliefs, we would expect Sakuddei to have been much less
affected in 1967 and for the contemporary people living in Sabirut (Maileppet and Muntei) to be
much more affected. For instance, the Sakuddei conception of Sikameinan might have been less
moralistic while the modern Sabirut conception might be more so, with a broader moral
jurisdiction. Instead, Sakuddei’s conception of Sikameinan seems to have been strikingly similar
to the one documented throughout southern Siberut: A punitive water spirit associated with
crocodiles attacks non-sharers, and shamans are called to remove that spirit in a public ceremony.
The third line of evidence suggesting minimal recent influence consists of the
observations of early ethnographers. In 1921, Kruyt, a missionary and ethnographer, spent two
months in the archipelago. He reported that people referred to water spirits as “kameinan” and
that these water spirits directed crocodiles to attack individuals who broke taboos or who, in
anger, threw food, damaged property, or split wood that they did not mean to use (Kruyt, 1923,
pp. 103-104). In the same ethnography, he reported that malevolent kameinan had loose hair
and wide shirts and were also known in some areas as Sibeulepei (“sibeoelepei”). They became
angry either when people disturbed their habitat, such as by washing entrails there, and, crucially,
when people did not distribute equal portions of fish (Kruyt, 1923, p. 153). (Confusingly, Kruyt
then asserted that kameinan spirits were not associated with crocodiles, contradicting his earlier
observation.) Kruyt’s observations differ slightly from ours, which may be due to his short time in
the archipelago, the diverse locations (he visited North Pagai and southern Siberut), or changes
that have occurred since then. But he documented beliefs in a punitive water spirits that were
known by the word “aunt” (kameinan), punished non-sharing, could be appeased by shamans,
and, depending on the particular passage you cite, were associated with crocodiles (see also
p. 25
Loeb’s (1929a, 1929b) descriptions of punitive crocodiles from his research in the Pagai Islands
in 1926, although he does not mention food-sharing).
If interactions with modern missionizing religions did not have a significant effect on the
reported beliefs and rituals, what about earlier contact with outsiders? The Portuguese were
aware of the islands as early as 1606, although they seem not to have landed (Tulius, 2012).
According to the Englishman John Crisp, who visited the Pagai (“Poggy”) islands in August
1792
3
, the English had attempted to cultivate pepper there 40 or 50 years earlier (Crisp, 1799).
Even before Europeans knew of the Mentawai Islands, the inhabitants likely traded with Malay
and Chinese merchants from Sumatra. Crisp reported that the Mentawai he met wore beads and
used machetes, and he saw Malays in the Pagai Islands during his visit. Moreover, items like
gongs and iron cooking pots are central features of Mentawai social life and likely streamed in
from trade with outsiders before European contact. We cannot reject the possibility that these
longstanding interactions somehow shaped the beliefs we have reported, even among those
groups that never met Europeans or merchants from Sumatra.
Even if these beliefs were shaped by outside pressures, the belief in a punitive spirit
connected to crocodiles seems to have deep roots. The people of Nias—who are the Mentawai’s
closest linguistic relatives and the most similar of 70 Indonesian communities, to the Mentawai
in terms of mtDNA (Hammarström, Forkel, & Haspelmath, 2019; Tumonggor et al., 2013)—
regarded crocodiles as both gatekeepers to the underworld (Marschall, 2010) and punitive agents
(Bakels, 1994). Nias society was much more socially stratified than in Mentawai, however, and
the punitive power of crocodiles was closely connected to the chief. There, crocodiles were said
to attack condemned prisoners and to devour escapees as they crossed rivers.
In closing, we hope our study demonstrates the capacity for primary ethnographic
research to provide rich descriptions of behavior in diverse societies and address targeted research
questions. This point is important, because, as with all sciences, the naturalistic study of cultural
behavior is undergoing a methodological revolution. Researchers increasingly draw on massive
online experiments, cross-cultural studies, and analyses of large-scale databases to test the
behavioral and historical predictions of existing theories (Mehr et al., 2019; Purzycki et al., 2016;
Slingerland & Sullivan, 2017; Watts, Sheehan, et al., 2015). These methods open up novel
research programs, but they have limitations. Analyses of large-scale databases must leverage
existing descriptions, and despite the richness of the ethnographic and historical record, many
observers failed to document behaviors and beliefs that are most relevant for key questions. A
comprehensive understanding of human behavior requires a diverse methodological toolkit,
including rich and detailed ethnographic investigations.
3
Crisp described Mentawai culture, but did not mention Sikameinan. In the passage most relevant to this paper, he
wrote, “Sometimes a fowl and sometimes a hog is sacrificed to avert sickness; to appease the wrath of the offender
power, or to render it propitious to some projected enterprise… But they have no form of religious worship, nor do
they appear to have the most distant idea of a future state of rewards and punishment (Crisp, 1799, p. 86).”
p. 26
References
Aksi Peduli Bangsa. (2020). Tentang kami — Sejarah. Retrieved April 10, 2020, from
http://aksipedulibangsa.org/tentang-kami/
Bakels, J. (1994). But his stripes remain: On the symbolism of tiger in the oral tradition of
Kerinci, Sumatra. In Text and Tales: Studies in Oral Tradition (pp. 33–51).
Bakker, L. (1999). Tiele! Turis! The social and ethnic impact of tourism in Siberut (Mentawai).
Leiden University.
Barrett, J. L. (2000). Exploring the natural foundations of religion. Trends in Cognitive Sciences,
4(1), 29–34.
Barro, R. J., & McCleary, R. M. (2006). Religion and economic growth across countries.
American Sociological Review, 68(5), 760. https://doi.org/10.2307/1519761
Barth, F. (1975). Ritual and knowledge among the Baktaman of New Guinea. New Haven: Yale
University Press.
Baumard, N., & Boyer, P. (2013). Explaining moral religions. Trends in Cognitive Sciences, 17(6),
272–280. https://doi.org/10.1016/j.tics.2013.04.003
Boehm, C. (2008). A biocultural evolutionary exploration of supernatural sanctioning. In J.
Bulbulia, R. Sosis, E. Harris, R. Genet, C. Genet, & K. Wyman (Eds.), Evolution of
religion: Studies, theories, and critiques. Santa Margarita, CA: Collins Foundation Press.
Boyer, P. (2001). Religion explained: The evolutionary origins of religious thought. New York: Basic
Books.
Boyer, P. (2018). Minds make societies: How cognition explains the world humans create. Yale
University Press.
Boyer, P. (2019). Informal religious activity outside hegemonic religions: wild traditions and
their relevance to evolutionary models. Religion, Brain and Behavior.
https://doi.org/10.1080/2153599X.2019.1678518
Chwe, M. S.-Y. (2001). Rational ritual. Princeton: Princeton University Press.
Clingingsmith, D., Khwaja, A. I., & Kremer, M. (2009). Estimating the impact of the Hajj:
Religion and tolerance in Islam’s global gathering. Quarterly Journal of Economics, 124(3),
1133–1170.
Crisp, J. (1799). An account of the inhabitants of the Poggy, or Nassau Islands, lying off
Sumatra. Asiatick Researches, or, Transactions of the Society Instituted in Bengal, for Inquiring
into the History and Antiquities, the Arts, Sciences, and Literature of Asia, 6, 77–92.
Delfi, M. (2013). Islam and Arat Sabulungan in Mentawai. Al-Jami’ah: Journal of Islamic Studies,
51(2). https://doi.org/10.14421/ajis.2013.512.475-499
Edelman, B. (2009). Red light states: Why buys online adult entertainment? Journal of Economic
Perspectives, 23(1), 209–220. https://doi.org/10.1257/jep.23.1.209
Gervais, W. M., & Henrich, J. (2010). The Zeus Problem: Why Representational Content
Biases Cannot Explain Faith in Gods. Journal of Cognition and Culture, 10(3), 383–389.
https://doi.org/10.1163/156853710X531249
Hammarström, H., Forkel, R., & Haspelmath, M. (2019). Glottolog 4.1. Retrieved February 6,
2020, from http://glottolog.org
Hammons, C. S. (2010). Sakaliou: Reciprocity, mimesis, and the cultural economy of tradition in
Siberut, Mentawai Islands, Indonesia. University of Southern California.
Henrich, J. (2009). The evolution of costly displays, cooperation and religion: credibility
p. 27
enhancing displays and their implications for cultural evolution. Evolution and Human
Behavior, 30(4), 244–260. https://doi.org/10.1016/j.evolhumbehav.2009.03.005
Hill, K. R., Walker, R. S., Bozicević, M., Eder, J., Headland, T., Hewlett, B., … Wood, B.
(2011). Co-residence patterns in hunter-gatherer societies show unique human social
structure. Science (New York, N.Y.), 331(6022), 1286–1289.
https://doi.org/10.1126/science.1199071
Indigenous Education Foundation. (2012). 2011-2012 community research report.
Kapitány, R., Nelson, N., Burdett, E. R., & Goldstein, T. R. (2019). The child’s pantheon:
Children’s hierarchical belief structure in real and non-real figures.
Kaptchuk, T. J. (2011). Placebo studies and ritual theory: a comparative analysis of Navajo,
acupuncture and biomedical healing. Philosophical Transactions of the Royal Society of London.
Series B, Biological Sciences, 366(1572), 1849–1858. https://doi.org/10.1098/rstb.2010.0385
Kruyt, A. C. (1923). De Mentawaiers. Indische Taal, Land-En Volkenkunde, 62, 1–188.
Lang, M., Purzycki, B. G., Apicella, C. L., Atkinson, Q. D., Bolyanatz, A., Cohen, E., …
Henrich1, J. (2019). Moralizing gods, impartiality and religious parochialism across 15
societies. Proceedings of the Royal Society B: Biological Sciences, 286(1898), 20190202.
https://doi.org/10.1098/rspb.2019.0202
Lanman, J. A., & Buhrmester, M. D. (2017). Religious actions speak louder than words:
exposure to credibility-enhancing displays predicts theism. Religion, Brain and Behavior,
7(1), 3–16. https://doi.org/10.1080/2153599X.2015.1117011
Loeb, E. M. (1929a). Mentawei myths. Bijdragen Tot de Taal-, Land-En Volkenkunde van
Nederlandsch-Indië, 85(1), 66–244.
Loeb, E. M. (1929b). Mentawei religious cult. University of California Publications in American
Archaeology and Ethnology, 25(2), 185–247.
Loeb, E. M. (1929c). Shaman and seer. American Anthropologist, 31(1), 60–84.
Marschall, W. (2010). Nias and Simeulue. In A. Graf, S. Schröter, & E. Wieringa (Eds.), Aceh,
history, politics, and culture (pp. 227–242). Singapore: Institute of Southeast Asian Studies.
Mehr, S. A., Singh, M., Knox, D., Ketter, D. M., Pickens-Jones, D., Atwood, S., … Glowacki,
L. (2019). Universality and diversity in human song. Science, 366, eaax0868.
https://doi.org/10.1126/science.aax0868
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, e1. https://doi.org/10.1017/S0140525X14001356
Peoples, H. C., Duda, P., & Marlowe, F. W. (2016). Hunter-gatherers and the origins of
religion. Human Nature. https://doi.org/10.1007/s12110-016-9260-0
Peoples, H. C., & Marlowe, F. W. (2012). Subsistence and the evolution of religion. Human
Nature, 23(3), 253–269. https://doi.org/10.1007/s12110-012-9148-6
Pew Research Center. (2017). The changing global religious landscape.
Purzycki, B. G. (2013). The minds of gods: A comparative study of supernatural agency.
Cognition, 129(1), 163–179. https://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(June), S000–S000. https://doi.org/10.1086/685729
Purzycki, B. G., Apicella, C., Atkinson, Q. D., Cohen, E., McNamara, R. A., Willard, A. K.,
p. 28
… Henrich, J. (2016). Moralistic gods, supernatural punishment and the expansion of
human sociality. Nature, 1–10. https://doi.org/10.1038/nature16980
Purzycki, B. G., & Sosis, R. (2019). Resistance, subversion, and the absence of religion in traditional
societies.
Purzycki, B. G., Willard, A. K., Kundtov, E., Apicella, C., Atkinson, Q., Bolyanatz, A., …
Ross, C. T. (2020). The Moralization Bias of Gods ’ Minds : A Cross-Cultural Test.
R Core Team. (2015). R: A language and environment for statistical computing. Retrieved from
http://www.r-project.org/
Reeves, G. (2000). Entities of life; entities of death. In The anthropology of the Mentawai Islands.
Retrieved from https://www.mentawai.org/anthropology-92-93/7-entities-of-life-entities-
of-death/
Roes, F., & Raymond, M. (2003). Belief in moralizing gods. Evolution and Human Behavior,
24(2), 126–135. https://doi.org/10.1007/s12110-014-9214-3
Rossano, M. J. (2007). Supernaturalizing social life: Religion and the evolution of human
cooperation. Human Nature, 18(3), 272–294. https://doi.org/10.1007/s12110-007-9002-4
Rossano, M. J. (2012). The essential role of ritual in the transmission and reinforcement of social
norms. Psychological Bulletin, 138(3), 529–549. https://doi.org/10.1037/a0027038
Rudito, B. (2001). Magi sebagai acuan identitas diri orang Mentawai dalam hubungan antar
suku bangsa. Makalah Dalam 2nd International Symposium of The Journal Antropologi
Indonesia Globalization and Local Culture: A Dialectic towards the New Indonesia, 18-21 July
2001. Padang: Universitas Andalas.
Schefold, R. (1982). The culinary code in the puliaijat ritual of the Mentawaians. Bijdragen Tot
de Taal-, Land-En Volkenkunde, 138, 64–97. https://doi.org/10.1163/22134379-90003482
Schefold, R. (1988). Lia: Das grosse Ritual auf den Mentawai-Inseln (Indonesien). Berlin: Dietrich
Reimer Verlag.
Schefold, R. (1998). The domestication of culture: Nation-building and ethnic diversity in
Indonesia. Bijdragen Tot de Taal-, Land- En Volkenkunde, 154(2), 259–280.
https://doi.org/10.1163/22134379-90003898
Schefold, R. (2007). Ambivalent blessings: Head-hunting on Siberut (Mentawai) in a
comparative Southeast Asian perspective. Anthropos, 102, 479–494.
Shariff, A. F., & Rhemtulla, M. (2012). Divergent Effects of Beliefs in Heaven and Hell on
National Crime Rates. PLoS ONE, 7(6), e39048.
https://doi.org/10.1371/journal.pone.0039048
Shariff, A. F., Willard, A. K., Andersen, T., & Norenzayan, A. (2015). Religious priming: A
meta-analysis with a focus on prosociality. Personality and Social Psychology Review,
(November). https://doi.org/10.1177/1088868314568811
Singh, M. (2018). The cultural evolution of shamanism. Behavioral and Brain Sciences, 41, e66.
https://doi.org/10.1017/S0140525X17001893
Singh, M., & Henrich, J. (2020). Why do religious leaders observe costly prohibitions?
Examining taboos on Mentawai shamans. Evolutionary Human Sciences, 2, e32.
https://doi.org/10.1017/ehs.2020.32
Slingerland, E., & Sullivan, B. (2017). Durkheim with data: The Database of Religious History.
Journal of the American Academy of Religion, 85(2), 312–347.
https://doi.org/10.1093/jaarel/lfw012
Sosis, R. (2006). Religious behaviors, badges, and bans: Signaling theory and the evolution of
p. 29
religion. In P. McNamara (Ed.), Where god and science meet: How brain and evolutionary
studies alter our understanding of religion (Vol. 1).
Tulius, J. (2012). Family stories: oral tradition, memories of the past, and contemporary conflicts over
land in Mentawai - Indonesia. Leiden University.
Tumonggor, M. K., Karafet, T. M., Hallmark, B., Lansing, J. S., Sudoyo, H., Hammer, M. F.,
& Cox, M. P. (2013). The Indonesian archipelago: an ancient genetic highway linking Asia
and the Pacific. Journal of Human Genetics, 58(3), 165–173.
https://doi.org/10.1038/jhg.2012.154
Tylor, E. B. (1920). Primitive culture: Researches into the development of mythology, philosophy,
religion, language, art, and custom, vol. 2 (6th Ed.). Retrieved from
https://archive.org/details/primitiveculture02tylouoft
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 B, 282, 20142556.
Watts, J., Sheehan, O., Atkinson, Q. D., Bulbulia, J., & Gray, R. D. (2016). Ritual human
sacrifice promoted and sustained the evolution of stratified societies. Nature, 1–7.
https://doi.org/10.1038/nature17159
Watts, J., Sheehan, O., Greenhill, S. J., Gomes-Ng, S., Atkinson, Q. D., Bulbulia, J., & Gray,
R. D. (2015). Pulotu: Database of Austronesian supernatural beliefs and practices. PLoS
ONE, 10(9), 1–17. https://doi.org/10.1371/journal.pone.0136783
White, C. J. M., & Norenzayan, A. (2019). Mental models: Associations between beliefs about
karma and God, moral values, and prosocial behavior. PsyArxiv. Retrieved from
https://psyarxiv.com/a7398/
Whitehouse, H. (1995). Inside the cult: Religious innovation and transmission in Papua New
Guinea. Clarendon Press.
Whitehouse, H. (2002). Modes of religiosity: Towards a cognitive explanation of the
sociopolitical dynamics of religion. Method & Theory in the Study of Religion, (14), 293–315.
Whitehouse, H. (2004). Modes of religiosity: A cognitive theory of religious transmission. Walnut
Creek, CA: AltaMira Press.
Whitehouse, H., & Lanman, J. A. (2014). The Ties That Bind Us. Current Anthropology, 55(6),
674–695. https://doi.org/10.1086/678698
Wiener-Bronner, D. (2020, March 13). Online sales of hand sanitizer and other “virus
protectors” spiked over 800% this year in the US. Retrieved April 11, 2020, from CNN
Business website: https://www.cnn.com/2020/03/13/business/hand-sanitizer-
sales/index.html
Woolley, J. D., Boerger, E. A., & Markman, A. B. (2004). A visit from the Candy Witch:
Factors influencing young children’s belief in a novel fantastical being. Developmental
Science, 7(4), 456–468. https://doi.org/10.1111/j.1467-7687.2004.00366.x