In this investigation, Balinese Hindus were interviewed to explore the impact of ritual practice on the flexibility and pattern of afterlife beliefs. Adults from communities where ancestral ritual practices are widespread were asked whether bodily and mental processes continue after death. Prior research with the ancestor-worshiping Malagasy Vezo revealed that their responses to such questions varied depending on narrative context (tomb vs. corpse scenario) and which conception of death they subsequently deployed: A religious conception, wherein death marks the beginning of a new form of spiritual existence, or a biological conception, wherein death terminates all living processes (Astuti & Harris, 2008). No studies to date have looked at the narrative effect in a culture having close proximity to altars dedicated to ancestors and frequent rituals to honor them. To explore the cross-cultural replicability of the narrative effect, an adaptation of Astuti and Harris’ experiment (Study 1, 2008) was conducted with Balinese Hindu adults. Participants heard one of two death scenarios and were asked about a deceased person’s capacities. Results revealed that Balinese adults were not influenced by narrative context. While they ascribed more mental than bodily capacities to the dead, they attributed comparatively more capacities overall than the Vezo. A distinctive Balinese pattern of capacity attribution was found, notably high attributions of an enduring spirit and real-time perceptual capacities. Findings suggest that the proximity and high frequency of rituals directed toward ancestors serve to shape, strengthen, and stabilize religious conceptions of death, while weakening the salience of solely biological conceptions.
Balinese Hindus’ Afterlife Beliefs as Stable Constructs: An Effect of High Frequency Domestic Rituals
Afterlife beliefs are one of the oldest and most enduring elements of human culture, dating back at least 100,000 years (McBrearty, 2007; Mellars, 1990; Piette, 2013). While adults and even young children have a basic understanding of the biology of death as the total cessation of living processes (Bering, 2002; Bering & Bjorklund, 2004; Slaughter, Jaakkola, & Carey, 1999), the majority of people in the world today report believing in some kind of continuing existence after death (Ipsos/Reuters, 2011). This phenomenon is in line with recent research showing that conflicting conceptions of death can coexist in people’s mind and may be selectively deployed depending on how death is presented (Astuti & Harris, 2008; Harris & Giménez, 2005; Lane, Zhu, Evans, & Wellman, 2016; Legare, Evans, Rosengren, & Harris, 2012; Rosengren et al., 2014; see also Nyhof & Clark, 2015). In the present research, we sought to further explore this interplay and the potential coexistence of different conceptions of death by conducting research with Balinese Hindus. They differ from other cultures that have been the focus of experimental research on death reasoning in that they practice elaborate daily domestic rituals dedicated to honoring the dead (e.g., Ottino, 2000; Sebestény, 2013, 2017a, 2017b).
Although the focus of the present study was to explore how cultural and contextual factors impact afterlife beliefs, much of the extant work on afterlife reasoning has sought to better understand the potentially universal cognitive basis for why afterlife beliefs are pervasive across cultures. As Astuti (2007) and Hutchins (1995) have pointed out, the study of cognition often has relegated culture to a peripheral role, while the study of culture is often done without reference to the cognitive organization and development of individual minds. Yet, there has been a growing recognition on both sides (e.g., Bloch, 1998; Sperber, 1996) that the human mind cannot be studied independently of the cultural context in which it develops, and that cultural knowledge cannot be studied without reference to the cognitive machinery underpinning it (Astuti, 2007).
Research focused on the cognitive foundations of afterlife beliefs has found that an early emerging and enduring tendency to treat the mind and body as distinct kinds of entities — a phenomenon known as intuitive mind-body dualism — likely facilitates the acquisition and maintenance of formal, religious afterlife beliefs (Astuti & Harris, 2008; Bek & Lock, 2011; Bering, 2002, 2006; Bering & Bjorklund, 2004; Bering, Hernández Blasi, & Bjorklund, 2005; Harris & Giménez, 2005; Huang, Cheng, & Zhu, 2013; Rosengren et al., 2014). More specifically, studies across a range of cultures find that children and adults are more likely to say that a deceased character retains mental capacities (e.g., capacities to know, desire, and feel) than bodily capacities (e.g., have a beating heart or legs that can move). These patterns are consistent with other findings on reasoning about one’s disembodied existence (Cohen, Burdett, Knight, & Barrett, 2011) and the time before biological conception (i.e., “prelife”; Emmons & Kelemen, 2014). Such findings illustrate that there is a recurrent bias to conceptualize people primarily in terms of their mental capacities rather than their bodily states.
While intuitive mind-body dualism is arguably a cross-culturally robust conceptual bias that likely underpins the acquisition of formal religious afterlife beliefs, death can also be construed in other terms. Indeed, two conceptions of death have been found to develop within cultures and individuals. Researchers have labeled these competing conceptions a “religious” and “biological” conception of death: While the religious conception of death focuses on the enduring, immaterial, and social aspects of a person, in which death marks the beginning of a new form of existence, the biological conception focuses on the corpse, in which death marks the end of agency and of all processes associated with life (Astuti & Harris, 2008; Barrett & Behne, 2005; Harris, 2011; Harris & Giménez, 2004; see also Hodge, 2011, 2012). As evidence of the latter, even preschoolers recognize that a dead animal is no longer able to move or know that other people are around (Barrett & Behne, 2005). Given that these two conceptions of death have contradictory assumptions about what is possible postmortem, the current study sought to explore how cultural and contextual factors influence which conception becomes most elaborated and strongest in the individual mind.
Prior work has shown that in different cultures, children and adults who are regularly exposed to both biological and religious conceptions of death selectively deploy these competing conceptions with a measurable flexibility depending on context (Astuti & Harris, 2008; Harris & Giménez, 2005; Lane et al., 2015). For example, Astuti and Harris (2008) studied beliefs about death among the ancestor-worshipping non-Christianized Vezo of rural Madagascar, who live in small traditional villages, practice fishing, and allocate a large amount of time and resources to appease and honor the ancestors. Astuti and Harris proposed that while religious afterlife beliefs predominate in certain contexts, for instance when looking for the cause of a misfortune or performing a ritual dedicated to ancestors, the biological conception of death predominates in other contexts, such as when preparing the corpse for the funeral ritual and washing it with cold water. This proposal was supported by findings from a series of experiments wherein Vezo participants heard one of two narratives about death before responding to questions about the postmortem continuity of mental and bodily capacities. The two scenarios were designed to mimic real-life situations that were either connected with ritual and ancestors (i.e., the “tomb” scenario) or with medical treatment, disease, and the physical corpse (i.e., the “corpse” scenario). Capacities attributed to a dead person were taken as mirroring the capacities attributed to ancestors. Results showed that when death was presented in the corpse scenario designed to trigger a biological conception of death, Vezo participants 8 years of age and older judged that fewer capacities would continue to function after death compared to when death was presented in the tomb scenario designed to trigger a religious conception of death. Astuti and Harris (2008) concluded that two conceptions of death arise over the course of development and are selectively deployed according to the situation in a non-overlapping way (p. 733).
In light of these findings, we sought to explore whether a repeated cultural emphasis on one conception of death might moderate the degree to which beliefs about death are flexible and thus affected by narrative contextual primes. For example, cultures that regularly engage in ancestor worship and frequently emphasize their ancestors’ postmortem capacities might be less affected by a biological presentation of death when making judgments about the deceased. Thus far, the Vezo are the only ancestor-worshipping culture whose beliefs about death have been experimentally examined. Although ancestor worship is central in their culture and religion, the ancestors’ acknowledged dwelling place is far away from their homes, deep in the forest, and is seldom visited. They have a low frequency of ritual communication with the dead, and it is primarily aimed at keeping a safe distance between the dead and living (Astuti, 1994; Astuti & Harris, 2008). We suspected that this distant and infrequent ritual engagement with the ancestors may help to account for why the narrative context in which death is presented impacted Vezo children’s and adults’ inferences about the postmortem capacities of a dead person.
How then might death be perceived in an ancestor-worshipping culture where altars dedicated to one’s ancestors are close to the dwelling place and where rituals addressed to them are frequent? Would the continuous observation of and participation in these kinds of rituals lead to the strengthening of afterlife beliefs to a point where they become stabilized and therefore impervious to a biological presentation of death? To answer this, the current study replicated Astuti and Harris’ Vezo experiment (Study 1, 2008) with Balinese Hindu adults who perform elaborate and frequent ancestor worship at altars in every home. As in the Vezo study, Balinese participants were read a story about the death of an adult — either the “corpse” story with a medical narrative designed to trigger a biological representation of death or the “tomb” story with a ritual narrative designed to trigger a religious representation of death. Participants were then asked questions about the bodily and mental capacities of the deceased person in the story. This research built on the first author’s long-term anthropological field knowledge of Bali, the second author’s expertise in cross-cultural experimental psychology, and both authors’ regular communication with researchers who have carried out cross-cultural experimental studies in the field.
Given that Balinese Hindus participate in daily domestic offering rituals addressed to the ancestors’ altars located in every home’s courtyard (Sebestény, 2013, 2017b), we thought it possible that they would not show flexible beliefs about death but instead would ascribe numerous capacities to a deceased person regardless of the narrative context in which death was presented. We further thought it possible that based on their frequent ritual engagement with the ancestors and close proximity to altars where this ritual engagement occurs, Balinese Hindus’ afterlife beliefs would be collectively stronger than those of the Malagasy Vezo. Evidence of elevated afterlife beliefs regardless of narrative context would highlight the importance of frequent ritual practice in strengthening and stabilizing afterlife beliefs. Based on the differences in the rituals dedicated to the ancestors, we also suspected that we would find differences in how the Vezo and Balinese conceptualize the ancestors. Namely, we expected to see differences in the types of capacities they attribute to the dead. Evidence of cultural differences in how ancestors are represented would further attest to how cultural practices associated with the treatment of the dead impact afterlife beliefs.
Furthermore, based on prior research (e.g., Astuti & Harris, 2008; Bering & Bjorklund, 2004), we predicted that Balinese Hindus would attribute more mental than bodily capacities to the dead. Notably, this prediction was in line with formalized Balinese religious afterlife beliefs, which support that the soul (atma, roh) continues after death and can be reincarnated (e.g., Hobart, 1978; Ottino, 2000; Sebestény, 2017a). Evidence of Balinese Hindus privileging enduring mentality over physicality following death would provide new evidence that mentality is more central to eternal conceptions of personhood and that this is a cross-culturally robust pattern.
The Ethnographic Context of the Study
To provide a richer picture of the cultural context in which this study took place, the first author conducted substantial fieldwork in Bali between 2000 and 2011. The first author’s fieldwork focused on ritual practices among common people, mostly involved in tourism and services, who lived in urban settings in the south of Bali, mainly in Sanur, Denpasar, Ubud, and the surrounding areas (Sebestény, 2012, 2013, 2014, 2015, 2017a, 2017b). The study interviews were conducted in these same areas.
Ancestor Worship, Reincarnation, and Social Divisions
Balinese Hinduism, also referred to as Balinese religion (Guermonprez, 2001), is Bali’s primary religion and has been influenced by Indian Hinduism, Buddhism, and local traditions (Picard, 2011). It is important to point out that it differs from Indian Hinduism in crucial ways. One of these is the link between ancestors and reincarnation: Unlike Indian Hindus, Balinese Hindus have a pervasive ancestor worship practice linked to the belief that reincarnation happens within family groups, such that ancestor spirits are reincarnated each time a new baby is born (Ottino, 2000, Guermonprez, 2001). Another important difference is the role of caste. Among Indian Hindus, social status and profession are defined through the caste system, drawing sharp boundaries between social and professional groups. In Bali, caste is meaningful in some but not all social contexts, and there are no socially excluded “untouchables” as in India. The most important social border in Bali is between the descendant groups corresponding to the three higher castes (Brahmana, Satria, and Wesia) forming the triwangsa nobility and the “commoners” corresponding to the lower Indian Sudra caste (Guermonprez, 1987). Professional and economic status does not predict the person’s caste, except in cases of the high priests who have to be Brahmana and politicians who are often triwangsa. Some Balinese Hindu groups outright reject the caste system (Barth, 1993; Guermonprez, 1990; Ottino, 1994; Méric, 2016). In the present investigation, the majority of Balinese Hindus interviewed were commoners, yet we did not exclude triwangsa nobility. As the belief in reincarnation and the rituals connected with ancestors are shared by all Balinese Hindus, we did not expect there to be differences in afterlife beliefs as a result of an individual’s caste or nobility. Nevertheless, we examined it to rule out this possibility.
With regard to other social divisions, Balinese villages and towns are divided into local communities (banjar) that have a common protecting divinity and celebrate ceremonies together (Sebestény, 2017b). Real or acknowledged kin groups (dadia) with a common mythical ancestor are also important social units (Geertz & Geertz, 1975; Ottino, 1994). Although religious traditions differ slightly between each city, village, banjar, and dadia, the major features of social and domestic rituals are similar across these groups (Sebestény, 2013, 2015). In the present investigation, people from various neighborhoods across the south of Bali were interviewed to ensure that a broad representation of individuals who identify as Balinese Hindu were included in the study.
Rituals for the Ancestors
Balinese Hindus have specific funeral rituals. When death occurs, the body is either cremated shortly following death or first buried and later unburied and cremated, sometimes years later. The local banjar community and the extended family, including children, are present at the ceremonies. Following the funeral ceremonies that lead to the disappearance of all physical remains, the spirit (roh) is said to go into the invisible world (neskala), and after a series of purifying ceremonies, thought to become an ancestor who dwells in a domestic altar. (Sebestény, 2017a)
The dead remain highly present in everyday life, both topographically and emotionally. Many rituals emphasize the ancestors’ presence and relation to the living. The ancestors’ altars, which are located in the family sanctuary in every courtyard, receive small daily offerings, mostly with incense, flowers, and tiny bits of food (Basset 1990; Ottino, 2000; Sebestény, 2013, 2015; Fox, 2015). Although the offerings are physical in nature, it is their invisible “essence” (sari) that is offered (i.e., the essence of the food, the beauty of the flowers, and the scent of the incense; e.g., Ottino, 2000, Sebestény, 2015).
Larger and less frequent rituals addressed to the ancestors are also celebrated in the homes. For example, when a family member has a ceremony linked to his birthday (otonan, every 210 days), offerings are presented both to the person having the birthday and to the ancestors, thereby ritually connecting them (Sebestény, 2015). On these occasions, offerings of food are also made, and after their essence (sari) is offered, the food is eaten and is considered blessed too. The extended family gathers every 210 days and presents offerings to their common ancestors. In these ceremonies, all deceased members of the lineage are represented as returning to the ancestors’ altar. In exchange for the offering, the descendants receive tirta water that gets blessed over the course of the ritual and thus carries a vitalizing essence for them. Another noteworthy ceremonial moment is connected with the concept of reincarnation. When a baby is born, parents usually go to a local specialist (balian) to ask which ancestor was reincarnated into the newly born child. In sum, the presence of the ancestors is construed in multiple ritual ways, through both small and very frequent but also bigger and less frequent rituals.
Additionally, the ancestors’ domestic altars are linked not only to the deceased that are known but also to the mythical ancestors that are referred to in social life for defining descent groups. This grants even more importance to the domestic altars because reference to common ancestors is an important part of how Balinese society is generally structured through dadia groups (Guermonprez, 1987, Geertz & Geertz, 1975), both at the family level and on larger social levels.
Maintaining Emotional Bonds in the Afterlife
Balinese funerals mark the end of an individual’s earthly life, and at the same time, they symbolically and ritually stage the continuity and gradual transformation of the dead into an immaterial entity. As described elsewhere (Sebestény, 2015), the funeral procedure begins from the first moment after death, staging the continuity of existence of the deceased person and the continuity of his relationships with the living. Right after death, the corpse is laid down to a dedicated place in the house and offerings of drink and food are placed beside the body for the enduring spirit. As the first author concluded from her field research, the relationship of the living with the deceased is transferred from the deceased’s physical body — which is gradually dematerialised — to new, symbolic bodies, then later to the altars. An example of the ritual process of reconstructing the relation to the deceased is observable by the end of the cremation ceremony: Close relatives carefully carry an effigy containing the cremated remains where the spirit of the dead relative is thought to dwell (Sebestény, 2014, 2015, 2017a). They place the effigy on their laps when sitting, reminiscent of the way small children are handled at ceremonies. In the Tabanan region, the first author observed people carrying the effigy with a cloth arranged the same way around the shoulders as if they were carrying a child (Sebesteny, 2015), ritually staging a close emotional bond. At the end of the ceremony, the effigy is thrown into the sea or a river. Through further ceremonies, the relation with the deceased is later transferred to the altars in the family courtyard, assuring its continuity (Sebesteny, 2013).
Balinese Hindus’ concern for the postmortem well-being of the dead person is strong, as evidenced not only in their funerary and daily rituals addressed to the dead, but also in their communication through traditional specialists (balian) who act as a medium between the living and dead (e.g., Barth, 1993; Sebestény, 2015). The typically very close connection and attention to the ancestors represents a cultural pattern that was repeatedly observed during fieldwork in many spontaneous expressions of relational and emotional closeness as well as strong beliefs that deceased relatives care about their descendants and continue to exert an influence on their lives. When someone has success in life (e.g., in business or studies), a thanksgiving ceremony is organized for the ancestors. When the relationship was good and loving during the deceased’s lifetime, the structure of Balinese Hindu rituals and habits appear to help to maintain it that way (Sebestény, 2013, 2015, 2017a).
The first author observed spontaneous expressions of relational and emotional closeness in Balinese Hindus’ communication with and in reference to the ancestors and altars. For instance, when introducing her family to the field anthropologist, a mother presented her husband, sister, two daughters, and also photographs of her younger daughter who had died four months earlier. She then pointed to the ancestors’ altar in the domestic sanctuary (sanggah), saying: “She now stays there. I bring her offerings every day. I cannot talk to her any more with common language, only in high Balinese.” High Balinese is used to address divinities, ancestors, and priests. It appeared that for this lady, the presence of her child in the ancestral altar was perceived as quite real, and she could regularly communicate with her through ritual procedures, using specific offerings and language. This matter-of-fact way of talking about communication with ancestors was noted on several occasions.
Differences Between Balinese Hindu’s and Malagasy Vezo’s Afterlife Beliefs and Practices
The Malagasy Vezo, based on Astuti and colleague’s description (Astuti, 1994; Astuti & Harris, 2008), also invest a considerable amount of time, money, and energy into rituals for their ancestors. However, their emotional and geographic relation to their ancestors bears striking differences with the Balinese. An important objective of Vezo funerals appears to be cutting the emotional bond with the dead. For example, when children view their dead parents for the last time, they are told never to call their name again (Astuti & Harris, 2008). Although this act does not equate with forgetting the person, it serves to emphasize the severing of the emotional relationship. Another difference is that the culturally determined way in which Vezo communicate with ancestors is through dreams wherein ancestors appear to their descendants complaining about being hungry or cold. These encounters are judged fearsome and are interpreted as requests for a new tomb or food offerings, which, if not met, may lead the ancestors to become angry and bring misfortune. The Vezo appear to view their ancestors as frightening entities that are best kept happy and as far as possible from the living so that they do not cause harm (Astuti, 1994).
There are also important differences in the treatment of the corpse and its physical distance from the living. During the funeral rituals, Vezo corpses are not continuously treated as sentient bodies. Astuti reports that when preparing the corpse before the funeral, for instance combing its hair or washing it with cold water, the Vezo do not view it necessary to proceed with as much care as for a living person, as it “can’t feel anything” (Astuti & Harris, 2008, p. 734). Such statements are not made for Balinese corpses, which are treated with continuous respect and attention (Sebestény, 2015, 2017a). Offerings are constantly placed beside the body from the moment of death, suggesting the continuous presence of the spirit. As already noted, Balinese ancestor spirits are believed to return into the family home and dwell in altars within the family courtyard after the corpse is entirely dematerialized through being cremated and the remains being thrown into the sea (Sebestény, 2014). Among the Vezo, corpses are buried, then subsequently unburied, re-buried, and given a new tomb, such that their earthly remains stay physically present in the tomb where the ancestor spirits are believed to dwell too. This is at a place far away from the village: Initial and subsequent tombs are all located in a cemetery deep inside the forest and are visited only for a few ceremonial occasions (Astuti, 1994). Vezo corpses and the ancestor spirits linked to them thus appear geographically much more distant from the living than Balinese ancestor spirits. In sum, while the Vezo also engage in ancestor worship, the nature of their rituals devoted to the dead are vastly different than what is observed among Balinese Hindus. For these reasons, we expected there to be differences in the pattern in which Balinese would respond to questions about the dead, as well as to the two presentations of death — religious versus biological.
Forty-five adults ranging from 17 to 83 years of age participated (23 men and 22 women; Mage = 33;6,
Participants completed interviews in a private setting, either at their home or during a quiet time at their workplace. Interviews were conducted by the first author — a Westerner — and a local assistant — an Indonesian anthropology student from Sulawesi island. Preliminary analyses revealed that there was no effect of interviewer on participants’ responses. Participants were not paid for taking part in the study. Given that Bali is a tourist destination, offering payment risked triggering a service reaction and eliciting more schematic answers. The researchers introduced themselves as university students, which they were, and invited individuals to spend ten minutes answering questions to help them better understand Balinese culture.
Procedure and Materials
Interviews were conducted in the national Indonesian language. Indonesian was chosen because it was known to all participants and is the language spoken in universities and workplaces in Bali, being socially much more neutral than the local Balinese language. The interviewer began by telling participants that there were no right or wrong answers and that she was interested in their personal opinions. Participants then heard one of the two narratives, either the tomb or the corpse scenario. Afterwards, they were asked a series of questions about the bodily and mental capacities of the dead person in the story. Half of the participants received the mental state questions first, half received the bodily state questions first. Preliminary analyses confirmed that the order in which questions were asked had no impact on responding. Responses were taped during the interview and verified later via listening to the audio recordings of the sessions.
The two narratives presented were translations of the ones used in Astuti and Harris (2008), with some minor adaptations to be appropriate for Balinese culture. A common Balinese name, Wayan, was chosen for the main character. The corpse narrative, designed to elicit a biological conception of death, described Mr. Wayan’s sickness, hospital treatment, and death. Modifications to this narrative included changing malaria to dengue fever, which, unlike malaria, is common in Bali. The tomb narrative, designed to elicit a religious conception of death, focused on Mr. Wayan’s funerals and presented him as an old man with many descendants. The main modification to this narrative was that, following traditional Balinese practice, the family and local banjar community participated in the funeral ceremony, and later, a small plangkiran altar was made for the dead in the domestic home where he received daily food offerings and coffee. The description of the burial remained as part of the narrative, given that in Bali, many corpses are first buried and only later cremated. Pilot testing indicated that both narratives sounded natural for the events described (see Appendix for the full narratives).
The original set of questions used by Astuti and Harris (2008) consisted of seven bodily state questions and seven mental state questions. All but one question was used in the present study (see Table 1). One of the bodily state questions (i.e., “Does he get older?”) was omitted because pilot testing revealed that it was difficult to understand and caused interruption in the flow of the interview. Additionally, for the mental state question connected to the wife’s name, the term “knowing” was changed to “remembering” because it was judged to be very similar in content yet much more natural during pilot testing.
Following Astuti and Harris (2008), three additional “entity” questions were asked about the global functioning of the body, the mind, and the spirit. Two questions had to be adapted. The original “Does his mind work?” question was modified to “Does his thinking still work?” as “thinking” (pikiran) was the closest Indonesian term for the concept of “mind.” The question “Does his spirit work?” was rephrased, as pilot testing indicated that the action naturally associated with the term “spirit” appeared to be about being there or not rather than being functional or not. Thus, this question was modified to mean “Does his spirit still exist?” and “Is his spirit still there?” given that the verb “to be” (ada) in Indonesian means both.
Responses were scored based on the frequency with which participants gave “yes” answers endorsing the continuity of a capacity’s functioning following death. Responses were assigned a score of either 0 or 1 for each capacity and were averaged for the mental and bodily state questions. To facilitate comparisons, reporting of analyses largely followed Astuti and Harris’ (2008) presentation. One key difference, however, is that analyses and percentages of “yes” rather than “no” responses are reported here. We also included some additional theoretically relevant analyses. For example, we compared Balinese and Vezo adults’ responses to explore whether the Balinese had stronger afterlife beliefs.
Effects of Narrative and Question Type
Preliminary analyses revealed no effects of gender or caste, and these variables were collapsed in subsequent analyses. To explore the extent to which narrative context and question type impacted Balinese adults’ reasoning, a mixed 2 (question type: bodily, mental) x 2 (narrative context: tomb, corpse) analysis of variance (
Effect of excluding two mental state items (feeling hungry and cold). In light of prior work showing that the capacities to feel hungry and cold are often associated with representations of the body (e.g., Bering & Bjorklund, 2004; Cohen et al., 2011; Emmons & Kelemen, 2014) and given that these capacities are not relevant for ancestor worship in Bali where ancestors are conceptualized as immaterial entities who consume the invisible essence (sari) of offerings rather than their visible form, we explored whether Balinese responses would show any effect of narrative context with these two items removed. A mixed 2 (question type) × 2 (context)
Individual subjects analysis. To better understand individual patterns of response and to further ensure that narrative had no effect on responding, following Astuti and Harris (2008), Balinese adults were allocated to one of four categories based on their overall levels of endorsing mental and bodily state continuity: those who gave 1) no continuity responses; 2) more continuity judgments for mental than bodily states; 3) more continuity judgments for bodily than mental states; 4) the same amount of continuity judgments for mental and bodily states. A total of four participants were classified into the final category, which took into account the slight disparity in total questions for the mental and bodily state categories (7 vs. 6 questions): Two participants gave all continuity responses; one participant gave 4 continuity responses for each of the mental and bodily state categories, while another gave 6 continuity responses for each category. A one-sample chi-square test revealed that participants were not evenly distributed across the four categories, χ2(3, N = 45) = 39.18, p < .001: This effect held for both the corpse and tomb narrative conditions, ps < .01. Across both narratives, most Balinese adults (64%) ascribed more mental than bodily states to the deceased: 20% gave no continuity judgments, 7% ascribed more bodily than mental states, and 9% gave equal continuity judgments for mental and bodily states. A chi-square test indicated that the context in which death was presented had no impact on the distribution, p = .33, providing further evidence that Balinese religious practices serve to stabilize and strengthen afterlife beliefs.
Effect of Culture (Balinese vs. Vezo)
To directly examine whether Balinese Hindus ascribed more capacities to the dead than the Vezo, we compared adults’ performance across the two cultures (excluding the “getting older” item for both cultures).1 A mixed 2 (question type: bodily, mental) × 2 (culture: Balinese, Vezo)
Differences in reasoning about the continuity of cognitive and perceptual-psychobiological capacities. Following Astuti and Harris (2008), we examined whether Balinese reasoned differently about cognitive capacities (i.e., the abilities to remember the location of one’s house, the name of one’s wife, and to miss one’s children) and perceptual-psychobiological capacities (i.e., the abilities to see, hear, and feel hungry).2 A 2 (question type) x 2 (context) mixed
To further explore this cultural difference in reasoning, we removed the culturally irrelevant hunger item from the Balinese data, which may have brought the average down for the perceptual-psychobiological category. This was found to be the case: The Balinese’s average endorsement of seeing and hearing was 70% (
Responses to Body, Mind, and Spirit Entity Questions
Following Astuti and Harris (2008), we explored reasoning about the body, mind, and soul as whole entities. To first evaluate whether narrative context impacted reasoning about these items, we ran a one-way
To further evaluate whether Balinese adults reasoned differently about the continuity of the body, mind, and spirit after death, we ran a series of three McNemar’s binomial tests: This revealed that the body was more likely to be rejected as continuing after death than either the mind or spirit (exact ps < .001). The functioning of the mind was also more likely to be rejected than the functioning of the spirit (exact p < .001).
Effect of Culture (Balinese vs. Vezo) on Body, Mind, and Spirit Attribution
Finally, to determine the potential impact of culture on adults’ attributions of a mind, body, and spirit to the dead, we ran three chi-square tests to evaluate the extent to which Vezo and Balinese adults differed. Analyses revealed that the two cultures did not significantly differ in their overall attributions of mind and body functions to the dead, ps > .38. However, they did differ in their attributions of a spirit, χ2(1, N = 91) = 12.17, p < .001: Balinese adults endorsed the functioning of the spirit significantly more often than Vezo adults (93% vs. 63%), which provides more evidence of stronger afterlife beliefs among the Balinese.
The aim of this interdisciplinary investigation was to explore the impact of ritual practice on the flexibility and pattern of afterlife beliefs observed among Balinese Hindu adults, while also gathering detailed data on Balinese Hindus’ afterlife beliefs. Building on existing research examining the cognitive underpinnings of afterlife beliefs and the first author’s extensive field knowledge of Balinese religion and culture, we replicated the adult part of Astuti and Harris’ (Study 1, 2008) Malagasy Vezo experiment with Balinese Hindus. Although both of these cultures practice ancestor worship, ethnographic data indicated that their ritual practices and beliefs about the ancestors were vastly different. For example, the Vezo believe the ancestors to reside far away from their dwellings, and they have low frequency ritual communication with them; whereas, Balinese Hindus engage in daily rituals honoring their ancestors at domestic altars and communicate with them on a routine basis. Consistent with these differences, results from the present investigation revealed that Balinese Hindus’ afterlife beliefs were more stable and less likely to be influenced by narrative context compared to the Vezo. Results also revealed meaningful cultural differences in how the dead were represented.
Regarding the stability of afterlife beliefs, our first research goal was to determine whether a narrative contextual prime designed to trigger either a religious or a biological conception of death would impact mental and bodily state attributions of the dead among Balinese Hindu adults. Evidence of a narrative effect was observed among Vezo aged 8 years and older, and finding a similar pattern among Balinese adults would have supported the claim that there are two different conceptions of death that develop and are selectively deployed according to context (e.g., Astuti & Harris, 2008; Harris & Giménez, 2005; Lane et al., 2015). Results indicated that while Balinese Hindu adults privileged enduring mentality over bodily state functioning, the narrative context in which death was presented did not affect their capacity attributions to the deceased person in the story. This observed stability in their conceptions of the dead is likely the result of the importance and proximity of ancestors’ altars and the high frequency rituals dedicated to them in Balinese Hindu’s everyday life. Notably, an effect of narrative context was absent even after factoring out culturally irrelevant items, examining individual patterns of reasoning, and when looking at attributions of the mind, body, and spirit as whole entities. Importantly, we do not interpret these findings as evidence that a biological conception of death is entirely absent among Balinese Hindus; however, we believe the findings suggest that this conception of death — characterized as the belief that all living processes terminate at the moment of death — is suppressed in favor of the religious conception of death that is repeatedly emphasized in Balinese Hindu religious practices, wherein a continued form of postmortem existence is understood.
In addition to examining the relative flexibility of afterlife beliefs, we sought to directly look at how Vezo and Balinese beliefs about the dead differ. In doing so, we aimed to gain a nuanced picture of the Balinese concept of ancestors. Based on important differences in the way Balinese and Vezo ritually communicate with their ancestors, which has already been described in detail, we suspected that the pattern and magnitude of beliefs would differ between the two cultures. Results revealed that Balinese afterlife beliefs were not only more stable than those of the Vezo, but that they were also stronger: The Balinese ascribed significantly more capacities to the deceased than the Vezo. Their attribution of an enduring spirit was also significantly stronger: Ninety-three percent of Balinese ascribed a spirit to the dead while only 63% of the Vezo did so. These differences are noteworthy given that both the Vezo and Balinese invest considerable amounts of money, time, and energy in the performance of grand-scale ceremonies to honor their ancestors. One might therefore predict that their level of adherence to the idea of an afterlife should be equally strong and not easily altered. However, the present findings help to reveal the importance of the way rituals are performed and suggest that certain aspects of rituals, namely frequency and domestic proximity, appear to impact the overall level of adherence to the idea of spiritual and mental state continuity of a deceased person.
Furthermore, the differences observed between how the Vezo and Balinese conceptualize the dead allow us to gain some insight into the links between ritual practice and afterlife beliefs. Results of the comparison suggest that the capacities attributed most frequently are those that are viewed as necessary for the ancestors to fulfill their part in the rituals addressing them. For example, while the Vezo tended to ascribe more cognitive capacities for remembering and missing, they ascribed fewer perceptual capacities for seeing and hearing. This is consistent with ethnographic data indicating that Vezo ancestors are neither viewed as being part of the families’ everyday life nor are wished to be part of it, not even in dreams (Astuti, 1994; Astuti and Harris, 2008). The very low frequency ritual practice connected with them means that they may not be regularly conceptualized as having real-time perceptual capacities. However, capacities for remembering their past (i.e., the location of their house and their wife’s name) and missing their children may be seen as a minimum requirement for ancestors to be conceptualized as social agents to whom one can address rituals (see Hodge, 2011, 2012 for discussion on the social aspects of death concepts; Emmons & Kelemen, 2014 for discussion on the social aspects of prelife concepts).
In contrast to the Vezo, Balinese endow their ancestors with high levels of mental state capacities allowing them to relate to both the present and the past. They frequently ascribe the dead with the ability to remember their home and wife and miss their children, yet they ascribe real-time perceptual capacities for seeing and hearing even more. This appears to be consistent with the daily domestic rituals addressing them, wherein the ancestors are thought to be present and participating. These daily rituals involve aesthetic offerings with flowers and incense addressed to the ancestors through their altars, possibly eliciting the inference in those who perform and witness the ritual that the ancestors are able to see the beauty of the offerings. Although inferring that the ancestors can hear does not appear to be as obviously triggered by daily domestic rituals, it is consistent with a representation of the ancestors as being present at the altars, caring about their descendants, and participating in a ritual communication. Daily domestic rituals are indeed staged as ritual communications, wherein the person presenting the offering stands tall, faces the altar, and executes ritual hand movements towards it (Sebestény, 2013, 2015). This conception of ancestors as able to hear is further supported by Balinese reports of talking to their deceased relatives at the altar or through mediumnic rituals through ritual specialists (balinan).
The present findings therefore suggest that ritual practice not only appears to impact the strength and stability of afterlife beliefs but also yields culturally specific conceptions of ancestors. Indeed, they illustrate that the types of capacities attributed to the ancestors are shaped by the kinds of rituals used to address them and may be guided by assumptions of the capacities the ancestors must have in order to participate in these rituals (see also Hodge, 2011, 2012). Further research should seek to explore the development of death concepts and afterlife beliefs in a wider array of cultural settings to further examine how various cultural factors interact with intuitive, early developing cognitive ideas (e.g., intuitive mind-body dualism). For instance, factors such as urbanization may play a role in the relative strength of biological versus religious conceptions of death as evidenced in other related reasoning domains (e.g., Emmons & Kelemen, 2015). Other cultural factors such as the emotional and social connection to one’s ancestors could also be examined. Lastly, contextual primes beyond simple narratives (e.g., engagement in or observation of activities that differ in context) could be examined to explore whether they might be strong enough for participants to mobilize one of the two competing conceptions of death.
Astuti, R. (2007). Weaving together culture and cognition: an illustration from Madagascar. Intellectica: revue de l’Association pour la Recherche Cognitive, 46/47, 173–189.
Barrett, H. C. & Behne, T. (2005) Children’s understanding of death as the cessation of agency: A test using sleep versus death. Cognition, 96, 93–108.
Bek, J., & Lock, S. (2011). Afterlife beliefs: category specificity and sensitivity to biological priming. Religion, Brain & Behavior, 1, 5–17.
Bering, J. M. (2002). Intuitive conceptions of dead agents’ minds: The natural foundations of afterlife beliefs as phenomenological boundary. Journal of Cognition and Culture, 2, 263–308.
Bering, J. M., & Bjorklund, D. F. (2004). The natural emergence of reasoning about the afterlife as a developmental regularity. Developmental Psychology, 40, 217–233.
Bering, J. M., Hernández Blasi, C., & Bjorklund, D. F. (2005). The development of ‘afterlife’ beliefs in secularly and religiously schooled children. British Journal of Developmental Psychology, 23, 587–607.
Bloch, M. E. F. (1998). How we think they think: Anthropological approaches to cognition, memory and literacy. Boulder, CO: Westview Press.
Cohen, E., Burdett, E., Knight, N., & Barrett, J. (2011). Cross-cultural similarities and differences in person-body reasoning: Experimental evidence from the United Kingdom and Brazilian Amazon. Cognitive Science, 35, 1282–1304.
Emmons, N. A., & Kelemen, D. (2014). The development of children’s prelife reasoning: Evidence from two cultures. Child Development, 85, 1617–1633.
Emmons, N. A., & Kelemen, D. A. (2015). I’ve got a feeling: Urban and rural indigenous children’s beliefs about early life mentality. Journal of Experimental Child Psychology, 138, 106–125.
Fox, R. (2015). Why do Balinese make offerings? On religion, teleology and complexity. Journal of the Humanities and Social Sciences of Southeast Asia, 171, 29–55.
Guermonprez, J.-F. (1990). On the elusive Balinese billage: Hierarchy and values versus political models. Review of Indonesian and Malaysian Affairs, 24, 55–89.
Harris, P. L., & Giménez, M. (2005). Children’s acceptance of conflicting testimony: The case of death. Journal of Cognition and Culture, 5, 143–164.
Hobart, M. (1978). The path of the soul: The legitimacy of nature in Balinese conception of space. In G. B. Milner (Ed.), Natural symbols in south east Asia (pp. 5–28). London: School of Oriental and African Studies.
Hodge, K. M. (2012). Context sensitivity and the folk psychology of souls: Why Bering et al. got the findings they did. In D. Evers, M. Fuller, A. Jackelen, & T. Smedes (Eds.), Is religion natural (pp. 49–63). New York: T&T Clark International.
Huang, J., Cheng, L., & Zhu, J. (2013). Intuitive conceptions of dead persons’ mentality: A cross-cultural replication and more. International Journal for the Psychology of Religion, 23, 29–41.
Ipsos/Reuters (2011). Belief in a supreme being strong worldwide: Reuters/Ipsos poll. Retrieved from http://www.reuters.com/article/us-beliefs-poll-idUSTRE73O24K20110425.
Lane, J. D., Zhu, L., Evans, E. M., & Wellman, H. M. (2016). Developing concepts of the mind, body, and afterlife: Exploring the roles of narrative context and culture. Journal of Cognition and Culture, 16, 50–82.
Legare, C. H., Evans, E. M., Rosengren, K. S., & Harris, P. L. (2012). The coexistence of natural and supernatural explanations across cultures and development. Child Development, 83, 779–793.
Méric, A. (2016). La fabrication des offrandes à Tenganan Pegeringsingan (Bali) et la mise en mouvement du monde. Moussons. Recherche en sciences humaines sur l’Asie du Sud-Est, 28. Retrieved from http://moussons.revues.org/3674
McBrearty, S. (2007). Down with the revolution. In P. Mellars, K. Boyle, O. Bar-Yosef, & C. B. Stringer (Eds.), Rethinking the human revolution (pp. 133–151). Cambridge, UK: MacDonald Institute.
Nyhof, M., & Clark, K. J. (2015). Afterlife beliefs in Chinese populations: The influence of culture and cognitive development. Poster presented at the biennial meeting of the Society for Research in Child Development, Philadelphia, PA.
Ottino, A. (1994). Origin myths, hierarchical order, and the negotiation of status in the Balinese village of Trunyan. Bijdragen tot de Taal-, Land- en Volkenkunde, 150, 481–517.
Picard, M. (2011). From Agama Hindu Bali to Agama Hindu and back – Toward a relocalization of the Balinese religion? In M. Picard & R. Madinier (Eds.), The politics of religion in Indonesia: Syncretism, orthodoxy, and religious contention in Java and Bali (pp. 117–141). New York: Routledge.
Rosengren, K. S., Miller, P. J., Gutiérrez, I. T., Chow, P. I., Schein, S., & Anderson, K. N. (2014). Children’s understanding of death: Toward a contextual perspective. Monographs of the Society for Research in Child Development, 79, 1–141.
Sebestény, A. (2013). L’offrande domestique à Bali – Un ancrage quotidien dans le monde. English title: The domestic offering in Bali – A daily anchoring in the world. In F. Hatchuel (Ed.), Transmettre? Entre anthropologie et psychanalyse, regards croisés sur des pratiques familiales (pp. 97–128). Paris: L’Harmattan.
Sebestény, A. (2014). Création collective d’une entité immatérielle: la crémation à Bali. English title: Creating an immaterial entity: The cremation in Bali. In Valentin F., Rivoal I., Thévenet C. & Sellier P. (Eds.), La chaîne opératoire funéraire: ethnologie et archéologie de la mort (pp. 40–41). Paris: De Boccard.
Sebestény, A. (2015). Les offrandes domestiques à Bali (Indonésie) comme point d’ancrage de la cohérence cérémonielle balinaise. English title: Domestic offerings in Bali (Indonesia) as the anchoring point of Balinese ceremonial coherence. (Unpublished doctoral dissertation). Nanterre University, Paris, France.
Sebestény, A. (2017a). Les funérailles à Bali – un calme émotionnel construit. English title: Balinese funerals: A constructed emotional calmness. Internationale de l’Imaginaire, 30, 169–190.
Sebestény A. (2017b). Bali – The island of the thousand temples, the thousand rice-fields and the million tourists: A successful encounter between international tourism and local culture. In T. Novák (Ed.), Go Hungary—Go Indonesia: Understanding culture and society (pp. 33–60). Budapest: University of Applied Sciences.
- Search Google Scholar
- Export Citation
( Sebestény A. ). 2017b Bali – The island of the thousand temples, the thousand rice-fields and the million tourists: A successful encounter between international tourism and local culture. In (Ed.), (pp. T. Novák 33– 60). Budapest: University of Applied Sciences.
Slaughter, V., Jaakkola, R., & Carey, S. (1999). Constructing a coherent theory: Children’s biological understanding of life and death. In M. Siegal & C. C. Petersen (Eds.), Children’s understanding of biology and health (pp. 71–96). New York: Cambridge University Press.
Two Scenarios of Death Presented to Participants
This is a story about a man called Mr. Wayan.3 Mr. Wayan worked hard all through his life. One day that was very hot, he worked hard, and caught dengue fever.4 His head and all his body were aching. His wife and children brought him to hospital,5 where he got four injections. His children and wife stayed with him in the hospital, but three days later he died.
Now Mr. Wayan is dead, his corpse was brought back to his home. Here are a few questions about him.
This is a story about a man called Mr. Wayan. He already has a lot of children and grandchildren. The day he died, many of his children and grandchildren came to his home. They made a ceremony at his home. All the people from the banjar6 came to the cemetery for his burial ceremony. At his home, they made a small plangkiran7 for him, and every day, his daughter-in-law brings coffee and rice to this plangkiran.
Now Mr. Wayan is buried in the cemetery. Here are a few questions about him.
The Vezo sample included 46 adults ranging from 19 to 71 years of age (23 men and 22 women; Mage = 34;9,
Astuti and Harris (2008) originally referred to the abilities to see, hear, and feel hungry as psychobiological capacities; however, we have adopted the phrase “perceptual-psychobiological” to be consistent with other prior work that has categorized capacities such as seeing and hearing under perceptual states (e.g., Bering & Bjorklund, 2004; Emmons & Kelemen, 2014).
Banjar: local territorial and religious community, one of the markers of Balinese culture and religion.