Taking people seriously

This work is licensed under the Creative Commons | © Rita Astuti. ISSN 2049-1115 (Online). DOI: http://dx.doi.org/10.14318/hau7.1.012


Taking people seriously

Rita ASTUTI, London School of Economics and Political Science

Taking the people we study seriously has resurfaced in recent years as a core aim of the ethnographic and anthropological endeavor. In this lecture, I present my way of taking people (i.e., my Vezo friends in a fishing village in Madagascar) seriously. For me, this involves understanding the multiple sources of their knowledge and the different ways of knowing that they mobilize in particular contexts and for particular purposes, at different ages, and fueled by different kinds of experience and cognitive resources. The argument is developed on the basis of empirical material that draws on an ongoing interdisciplinary collaboration between anthropologists and cognitive psychologists.

Keywords: death, ancestors, knowledge, learning, children, ontological turn, Madagascar

Interdisciplinarity is widely valued—by funders, academic leaders, students, and reformers—but it is hard work. For it to succeed—that is, for it to generate more knowledge and understanding than the sum of the knowledge and understanding generated by each constituent discipline working on its own—it requires a great deal of patience and a certain amount of tolerance. I have experienced this first hand: as I started collaborating with developmental and cognitive psychologists some twenty years ago, I had to patiently learn a new language, get used to different evidential standards, embrace new methodological tools, while at the same time politely insisting that my language, evidential standards, and methodologies be understood and respected. Inevitably, along the way, there were instances of miscommunication and moments of frustration. The effort, however, was well worth it: on the one hand, I have deepened my knowledge and understanding of the Vezo people of Madagascar who, over the course of nearly thirty years, have allowed me [106]to be part their lives; on the other, as I shall illustrate in this article, I have found a way of “taking people seriously.”

Anyone who has been following the onward trajectory of the so-called ontological turn will have noticed that “taking seriously” the concepts, the analogies, the discourse of the people one studies ethnographically is a recurrent theme in the writings of those who have taken this turn (e.g., Henare, Holbraad, and Wastell 2007; Holbraad 2012; Pedersen 2011). Their claim is that the only way to do so is to take people’s alterity at face value, that is, to be open to the possibility that the various people we encounter in the field inhabit different realities from our own (see Graeber 2015 for a critical appraisal of the philosophical underpinnings of this position).

To illustrate: when, famously, a Cuban diviner says that powder is power, he is not articulating a representation that happens to be different from that of the anthropologist, one that represents powder as the kind of thing that can have power; instead, he is bringing forth a different “world” altogether, one in which powder is power (Holbraad 2012). To unlock such a world, according to the “ontologists,” anthropologists have to invent new concepts that will enable them to unproblematically describe powder as power. And to do this, they have to shed their Cartesian dualistic ontology and realize that things, if taken seriously, are concepts (hence the injunction that we “think through things” as the non-dualist-other allegedly does; Henare, Holbraad, and Wastell 2007).

In this lecture, I shall advocate a different approach. Like those who have opted for the ontological turn, I also start from ethnographic puzzles. However, my puzzles are of a very different nature from those that exercise the “ontologists.” This is because mine are generated by paying resolute attention to the multiplicity of ways in which people create and deploy their knowledge in different contexts, at different ages, fueled by different kinds of experience. For example, were I to interact with Cuban diviners, the puzzle that I would choose to engage with would not so much be that they take powder to be power but that—as we are told in passing (Holbraad 2012: 157)—they are also certainly not confused about the difference between the two. How shall we account for this? What does this tell us about the radical alterity of the non-Cartesian reality inhabited by the diviners?

But this, of course, is not my puzzle. My puzzle, which I shall explore in this lecture, is generated by the complex and apparently contradictory nature of Vezo ancestors, the word I shall use as a short hand to refer to what my Vezo friends describe as “the people of the past who are now dead.”

The ancestors

When Jira acquired a small TV set—a gift from the owners of a South African yacht he had worked on as an assistant sailor—he placed it in the middle of the table in his father’s house, under the cover of a lace tablecloth. There it stayed for a week or so, in acknowledgment of the fact that juniors should never surpass their seniors, whether in height (in which case a gift is given by the junior to the senior as a kind of apology), the size of one’s house (in which case the senior is invited to be the first occupant, at least for a few days), or the acquisition of the latest novelty (which is temporarily [107]handed over to the senior, as with Jira’s TV). On the day chosen in consultation with his father, who happened to be the hazomanga, the mediator between the living and their dead forebears, Jira approached the house with a small bottle of rum and a small sum of money. His father accepted the rum and money and, sitting in the house facing east, called upon the ancestors—the most senior of all—to announce that Jira had returned from his trip (which they had previously been informed about) and that he had brought back this new thing. The ancestors were asked to take notice so as not to be surprised about this new acquisition; they were asked to rejoice in the fact that Jira had been successful in gaining such wealth, and to give their blessing. After the conversation was over and the rum consumed—a few drops by the ancestors, the rest by those family members who happened to be present—Jira picked up the TV and transported it the short distance between his father’s house and his own house next door. The arduous and largely unsuccessful process of making it work thus began, although at no point did I hear Jira surmise that the difficulties he encountered were due to the jealousy or the malice of the ancestors or to that of his father—the ones and the other being equally worthy of respect and a potential cause of trouble when slighted. It was more to do with the poor reception of the homemade antenna and the hopeless quality of the car battery he had been sold as nearly new.

By contrast, when Lola woke up with a swollen and extremely painful ear, everyone in the family was quick to conclude that the illness was caused by Lola’s disgruntled grandmother, who had died ten years earlier. A few months before, Lola had traveled north with her husband to explore new fishing grounds. Despite various setbacks due to bad weather and lack of food and clean water, the trip had been financially successful. But during her stay up north, Lola had a dream in which her grandmother had asked her to cook a meal for her because she was hungry. In the dream, Lola cooked rice and asked her grandmother what she wanted as a side dish. Fish, was the answer. As the grandmother ate the food, she commented on the fact that she was forced to ask her grandchildren for food because her own son (Gramera, Lola’s mother’s brother) did not give her any; it was as if she had never given birth, as if she had no children, seeing that her children let her sleep outdoors, and left her without a proper house. This, as everyone who heard about the dream clearly understood, was a reference to the fact that after so many years since her death, the family had yet to complete her “work,” namely the construction of a cement fence and cross to substitute the temporary wooden ones used at the time of her burial. As Lola’s swelling gave no sign of subsiding, the family gathered for an offering of cooked rice (people were relieved that in the dream the grandmother had asked for fish rather than meat, for this meant that a rice offering, rather than the provision of a zebu, would suffice). Interestingly, when Gramera, in his role as hazomanga, spoke to the ancestors, including his mother, he ignored the accusation that he had failed to take care of his mother’s house. Instead, he asked the ancestors to excuse Lola who, being young and unwise, had failed to properly announce her journey northward. I was surprised that Lola should be the one to get the blame, but no one else seemed to be since no one else paid much attention to Gramera’s words, spoken in the hushed sort of way that hazomanga use to invoke the ancestors. My guess is that Gramera was too scared to admit his family’s serious failing in the presence of the ancestors and that he hoped to divert their attention to a far less serious breach of etiquette.[108]

A few years earlier, I had my own problems dealing with Gramera’s mother. At that time, she was still alive but very old and frail. Despite this, temperamentally she continued to be extremely forceful. I had never felt that she liked me all that much, but our relationship got openly strained when I returned for a visit a few months before she died. She had become very demanding, as she apparently thought that she was entitled to ask for, and obtain, all my possessions. Her eldest daughter and Gramera had intervened to deflect her requests, but the most dramatic incident happened after she got me to hand over a new blanket I had bought for myself at the start of my visit. When Gramera found out about it, he summoned her and me to his house, and forced her to return the blanket. He was furious, and scolded her for always begging—couldn’t she see that I needed a blanket? Why was she always making him feel so ashamed? She didn’t speak a word. She just sat there, fiddling with the corner of her sarong. But she did send the blanket back.

At the time of my departure, I went to say goodbye and gave her the blanket, nicely washed and, by local standards, still in excellent conditions. I have a photo that captures this moment: she is sitting on a mat, leaning against the wall of her house, the folded blanket on her lap. She is looking sideways, lost in thought, brooding. When I look at this picture, I cannot help reading into it the events that unfolded four years later, when I returned to the village with my son.

Sean was four-and-a-half years old and, as predicted, he took village life in his stride and seemed very happy. But after about a month, he got severely ill. Late at night, after securing the medicines to treat him against typhoid and malaria, I was back in the house, trying to keep myself calm and him hydrated. Quietly, Gramera entered the house and sat on the floor in the near darkness, facing east and holding a little cup in his hand. He started speaking, in his usual muffled way, calling upon the creator Ndranahary, the four cardinal points and then the names of the ancestors. But after this initial invocation, Gramera’s tone suddenly changed for he was now talking to one person in particular: he was talking to his mother (addressing her as such). And he was angry: he raised his voice, accusing her of making Sean ill. He asked her what she thought she was doing. Had she not been told (a month earlier) that her granddaughter had come back from “the other side of the ocean” and that she had brought her son? Had she not been asked to protect them both? And look instead what she was doing! By now he was nearly shouting at her, as he told her that she made him feel ashamed. I was squatting in the middle, between a furious Gramera and a delirious Sean, not quite knowing what to make of it all. Things got worse when Gramera concluded his invective by sprinkling around the room the cold water he had been holding in the cup; a few drops fell on Sean’s burning back, and this made him scream in terror. To Gramera, Sean’s reaction to the blessing water confirmed his assessment of the situation: it was his mother’s vengeful nature that was responsible for Sean’s illness. What was striking about this conversation is that there was remarkably little difference in the way Gramera spoke to his mother about Sean’s illness—when she was dead—and about the blanket—when she was still alive.

I could fill many more pages, telling stories of how the Vezo people I know interact with the ancestors, ranging from almost routine blessings when the launch of a new canoe or a journey are announced, to more dramatic and costly offerings of rum, rice, or meat to seal a marriage or to claim the right to bury one’s children’s [109]bones into one’s tomb (see Astuti 1995). But the few examples I have presented are enough to convey some key aspects of these interactions and to reveal some key characteristics of Vezo ancestors.

First, the ancestors are expected to retain a keen interest in the affairs of the living, which is why they need to be kept informed of their descendants’ key life events. It is not that the ancestors are unable to find out for themselves, for example, that Jira has traveled far and has returned with a TV set. Since people imagine the ancestors to be like “wind/air,” invisible but ever present, they also assume that they know what happens to the living. But the point is that they want to be told, as a sign of respect. If affronted, they will retaliate.

A second key characteristic of the ancestors is that they reveal themselves in dreams. They use dreams to communicate with the living, normally as a way of making demands on them, but also to let them know that they are displeased or offended. Consequently, dreams that feature dead people should be reported to one’s elders at once (even though, in practice, they are often only remembered after an illness or some other misfortunes have occurred).

I was told that the reason the ancestors use dreams as their communication channel is that they no longer have a body, which rotted away after being buried under the sand. The ancestors are thus immaterial like “wind/air,” and invisible—except, of course, in people’s dreams. What people see in their dreams are the deceased, with body and all, just as they were at the time of their death: as young or old, tall or short, generous or stingy as they used to be.

And finally, the way one interacts with the ancestors can be strikingly similar to the way one interacts with living people (as noted long ago by Kopytoff 1971): most of the time honoring them, as they would do their (living) elders, but at times scolding them for acting unreasonably. To be sure, none of my Vezo friends ever expected to conduct a two-way conversation with the ancestors, at least not in the manner that conversations unfold among living people. This is because the ancestors take their time to disclose what they make of their descendants’ attempts to please or placate them: if they are satisfied, no harm will come; if they are not, they will send illness, misfortune, and even death.

To summarize, this is how Vezo ancestors can be characterized: they are people who were once alive and are now dead. When they died, their spirit permanently detached from their body, which was left to decay under the sand. As a result, the ancestors are immaterial but willful entities. Like wind, they move around and witness the successes and tribulations of their descendants; they enter people’s dreams to demand, complain, or punish. The way to appease them is to show them respect by keeping them informed, by respecting them, by remembering them, and by keeping their houses tidy.

The glitch

Although I regard this summary as largely correct, and I believe that most Vezo adults would understand it and endorse it, I shall now destabilize its apparent coherence. To do this, I will start by explaining one other important characteristic of Vezo ancestors, namely that they are universally expected to be masiake: aggressive, [110]unreasonable, unpredictable. Since I am not basing myself on Vezo exegesis as such, the reader will have to follow my attempt at imagining what it might feel like to be a Vezo ancestor.

So, why should most Vezo ancestors, especially those recently dead, whose names and personal stories are still remembered, be masiake? The answer, I think, is that the ancestors are not happy about being dead because they retain a strong longing for the living (several people did say this much: olo maty manino any olom-belo). It is their frustration and desire to be back among the living that make them angry and short-tempered.

As I have discussed at length elsewhere (Astuti 1994, 1995, 2000), one of the reasons the ancestors long for the living is that, inside their tombs, they are in the company of only “one kind of people” (raza raiky, what we might label a unilineal descent group), whereas in life they used to enjoy the benefits of a virtually boundless network of relations (filongoa, what we might label as cognatic kinship). Therefore, when the living work for the ancestors, by building tombs or crosses for them, they invite the largest possible number of people to attend—to make up for the difference, as it were, between what the ancestors have inside the tomb and what they have left behind. And so, although only a few of the participants can claim to be “the masters of the work” (those who will themselves be buried in the tomb that is being worked on), the point of the work is to let the ancestors enjoy, if only momentarily, the spectacle of all their descendants, who have gathered to remember them, to drink plentifully and to dance provocatively with them.

And this brings out the other reason why the ancestors long for the living: quite simply, because they miss the food, the drinks, the tobacco, the sex, the dancing that they enjoyed in life. Their descendants, through their actions, make this abundantly clear. Take the building work that happens at the cemetery, when the temporary wooden fence is replaced with a new cement one. Once the old enclosure is removed and the work on the new one gets under way, those who are not directly involved in the brickwork (most of the crowd) will dance straight over the bodies of the ancestors, stumping their feet as hard as they can so that the dead underneath can hear and feel the dance; the screeching tape-recorder is placed right in the middle of the tomb, so that the ancestors can hear the music. And every time a new jerrican of rum is brought out to keep things moving along, some of the contents are sprinkled over the ancestors so that they too can enjoy the general state of inebriation.

The same longing for the sensuousness of life is catered for during the “work of the crosses.” Cement crosses that bear the name of the recently dead are molded in the village; in a moment marked by intense emotions, the crosses are first raised to a vertical position and are then transported to the cemetery by large crowds of inebriated people. Depending on the status of the deceased whose cross is being carried—for example, if it happens to be that of an elderly grandmother, who is remembered for having had lots of children, grandchildren, and great grandchildren—the dancing that ensues can turn very exciting indeed, with people thrusting their pelvis against the cross, enticing the deceased to dance harder and harder. Some of the same actions can be seen during the funerals of very old people when those who carry the coffin rush back and forth with it, pushing and pulling, while at the same time engaging the surrounding crowd in very explicit dance moves. In both cases, whether the dancing and the thrusting is directed at a cement cross or [111]at a coffin, the people involved will impute their actions to the character of the deceased, which takes them over and guides their actions. The idea is that, for one last time before entering the tomb, the deceased wants to experience the full excitement and the intense sensuality of life.

And here is the glitch, the inescapable muddle (for sure, not a muddle that I claim to be unique to the Vezo). According to the account I have given above—which, I want to reiterate, would sound true to most of my Vezo friends’ ears and hearts—the ancestors are spirits without a body: their body rots, under the sand, while their spirit survives as an immaterial entity. And yet, as we have just seen, the body sticks. It sticks because the ancestors (via their spirit, one would assume) want to enjoy life and because their descendants can appease them in the only way they know: by stamping their feet, turning up the music, thrusting their pelvises, and pouring the rum. When the living are engaged in such activities, the ancestors are not only very much sentient—they are also very much embodied.


During the course of fieldwork, if one happens to be at the right place at the right time, one can witness fleeting but highly charged moments in which death is represented as a process of complete extinction and sensory annihilation. Typically, this happens when people are dealing with death at close range: when they handle the lifeless body of a relative, when they witness the onset of decomposition during a prolonged funeral, when they fend off the mourners’ distressed pleas to extend the funeral even longer. On these occasions, one hears statements such as “when one’s dead, one’s dead” or “she’s dead, she doesn’t feel anything” (see Astuti 2007 for one full account). And because of the muddle over the body—a body that rots but that nonetheless affords the feelings and sensations that keep the ancestors connected to the living—to state that dead people are just dead and that they cannot feel anything anymore is to deny them their ancestral existence.

The ethnographic puzzle

It thus seems that Vezo ancestors are represented and experienced in two contradictory ways. On the one hand, “the people of the past who are now dead” carry on existing as spirits, with no body but with bodily sensations, desires and a masiake disposition. On the other hand, “the people of the past who are now dead” are just dead, with no sensations or willfulness left in them. Admittedly, the weight of the ethnography leans heavily on the first way. One can observe people turning to the ancestors on countless occasions; expending time and money to build and maintain their houses; respectfully offering food and drinks to them; or teasing them with sexual banter. By contrast, the view that “when one’s dead, one’s dead” gets expressed much more rarely and fleetingly. Even so, its existence should not be ignored or diminished for it reveals that, in their thoughts and actions, Vezo people can live with an apparent contradiction. This, in turn, raises the analytical challenge of taking them—and their contradiction—seriously.[112]

Turning to children

My way of tackling this challenge has been to turn to children. This is because by studying how Vezo children come to know about death and the ancestors, one can begin to grasp the full complexity of adult knowledge—including its contradictions.

To plan my work with children, I enlisted the help of developmental psychologist Paul Harris. I turned to him in particular for two main reasons. First, because his distinctive theoretical interest has been in the crucial role played by testimony in children’s developing understanding of the world (Harris 2012; Harris and Koenig 2006). While in the case of adults, it is readily recognized that every aspect of their conception of the world is dependent on what they are told by other people, the theoretical trend in developmental psychology has typically been to ignore this dependence in the case of children. Following Piaget, the trend has been to concentrate instead on what children learn through their direct experiences and observations. As a corrective, Harris has engaged in a sustained program of research into what and how children learn from the testimony they receive, and often actively seek, from others.

It is worth noting that Harris’ criticism of the conventional wisdom in developmental psychology offers an interesting counterpoint to the by now well-established criticism of the conventional wisdom in anthropology. While Harris complains that children in his discipline are largely approached as “stubborn autodidacts,” a common complaint among anthropologists has been that in their discipline children have largely been approached as, so to speak, “passive assimilators,” who absorb whatever testimony the cultural environment throws at them. In truth, what is highlighted by these discipline-specific complaints is that children resort to a range of strategies, sources, and resources to make sense of the world around them, which gives complexity to their knowledge—and that of adults. The fact that Harris’ empirical and theoretical focus on testimony places this complexity center stage made him the ideal partner for my investigation into the nature of Vezo ancestors.

The second reason I turned to Paul Harris is that, as it happened, one of his case studies about the role of testimony looked specifically at how children learn about death and the afterlife. To start off, Harris notes that the extensive literature on the development of the understanding of death has a distinctively “secular” flavor: the emphasis is squarely on the development of children’s understanding of death as a biological phenomenon, while the fact that children are exposed to a variety of ideas about the afterlife is hardly registered. And so, through a study of Spanish children, Harris and his collaborator Marta Giménez wanted to explore how children’s biological knowledge of the life cycle relates to ideas about the afterlife, as conceived in the Catholic community in which they grow up (Harris and Giménez 2005).

To do this, they designed a simple experimental tool—what I shall call the death interview. The design is as follows: children were told a short story about the death of a person and they were then asked which of the deceased’s properties, if any, remain viable after death. Some of the questions concerned the body: Does the mouth work? Does the brain work?; and some of the questions concerned the mind or spirit: Does the deceased think? Does she feel emotion?

In addition, the design included a simple contextual manipulation, such that the death of the character was either presented in a “secular” or “religious” setting: [113]in one story, a doctor announced it; in the other story, a priest announced it. The results of the study showed that children modulated their responses accordingly, giving more biologically informed answers in the secular context and more religiously informed answers in the religious context, providing justifications to match (see Harris and Giménez 2005 for full details).

The death interview in Madagascar

As I have discussed in more detail elsewhere (Astuti and Harris 2008), I adapted Harris and Giménez’s death interview to make it suitable to the Malagasy context. I chose slightly different properties, which would map onto the local distinction between the body that rots (do his eyes work? does his heart beat?) and the spirit that survives (does the deceased remember where his house is? does he miss his children?). In addition, I transposed the two narratives to situations that would resonate with the local experience of death and the ancestors, with one story (the corpse narrative) evoking the presence of a dead body waiting to be buried and the other (the tomb narrative) evoking the presence of a ritually satisfied ancestor housed in his well-kept tomb. I then recruited forty-six adults (between the age of 19 and 71) and forty children (between the age of 9 and 13) and asked half of them to participate in the corpse interview and the other half in the tomb interview.[114]

There are two main headline findings (see Astuti and Harris 2008 for full details). First, both adults and children were more likely to judge that the bodily properties of the deceased would stop working after death than his mental properties; in other words, their answers reflected the view that when a person dies the body rots and all its properties stop functioning, while the spirit and its cognitive and emotional functions continue to be viable. Second, both adults and children were affected by the narrative context, as they were on average more likely to give extinctionist answers (i.e., it doesn’t work) when they heard the corpse narrative than when they heard the tomb narrative; at an individual level, participants who heard the corpse narrative were more likely than those who heard the tomb narrative to say that all the fourteen functions they were quizzed about would cease at death.

Beyond these broad similarities, children differed from adults in two ways: across the two versions of the interview, they differentiated less sharply between bodily and mental properties and they were more strongly affected by the narrative context. These two trends are related to each other. On the one hand, children overall differentiated less sharply than adults between the fate of the body and the fate of the spirit because they were more likely to say that mental properties will cease to function after death when they heard the corpse story and they were less likely to say that bodily properties will do so when they heard the tomb story. On the other hand, while the effect of the tomb narrative on adults was restricted to their judgments about mental properties (that is, adults were as likely to say that bodily properties cease to function after death in the two versions of the task), in the case of children, the effect of the tomb narrative extended to their judgments about the body of the deceased (that is, when children heard the tomb narrative they were more likely to say that the deceased’s bodily properties remain viable after death than when they heard the corpse narrative).

There are thus two interesting and distinctive aspects of children’s responses to the death interview. When presented with the evocation of a recently dead person (in the corpse narrative) they were more extinctionist than adults; when presented with the evocation of a satisfied ancestor (in the tomb narrative) they were more survivalist than adults, but in a very particular (nonadult) fashion.

I shall comment on these two points in turn.

First, I want to speculate as to the source of children’s extinctionist stance, well captured by a nine-year-old boy who told me, “the body goes bad, the skin is all decomposing and inside the tummy is full of worms, and in the head there are all sorts of animals that go inside it, and there is nothing that dead people need and they can’t hear anything.”

Vezo children do not learn about life or death in the classroom. I know this because most of the children I worked with hardly ever go to school, and when they do they seem to spend all of their time copying out letters and numbers on their small blackboards. But in the course of their everyday lives, they have plenty of opportunities to experience first-hand a range of biological processes, including death.

The death of animals is perhaps the most accessible as they themselves often inflict it on their animal “toys,” whether butterflies they use as kites, crabs they use as dinky cars, or birds they keep as temporary pets before turning them into an eagerly shared snack. They are also keen observers of big kills: the rare slaughtering of a zebu or the slow and painful death of a sea turtle are sure to attract a thick crowd of children, clearly interested in scraps and snatches, but also seemingly fascinated by the process of death itself—the noise, the twitching, the smell. And, more casually, they handle dead fish on a daily basis, as the canoes return with their catch and the beach becomes dotted with blood and entrails, which the children collect, squeeze, and squash and throw at each other.

And then, there is human death. Adults often reported to me that in the past children were not allowed to be present at funerals but that nowadays, seeing that people die “isanandro isanandro” (literally, “everyday”), it is just impractical to keep them away. Be that as it may, at funerals one can now observe large crowds of children of all ages. They sometimes help fetching firewood or water for the communal cooking, but mostly they gather to play. As they do so, new visitors will arrive and will enter the house where the deceased is laid out under a mosquito net. Their entry triggers the sudden and dramatic eruption of bouts of loud wailing, which can be heard all around. At these moments, children will abandon their games, rush to the house and hang from windows and doors and any other available opening to witness the commotion. As they watch, aside from the outpouring of grief, they must also take in the stillness of the deceased and the smell of its fast-decomposing body.

With all of this going on—the deadly games with animals, the big kills, the routine handling of the daily catch, and the many funerals they participate in—it seems plausible to assume that it is as a result of these experiences that children come to the conclusion that death is a process that extinguishes all of life’s properties, such as movement, sensory experiences (including suffering), emotions (including pain), and the preservation of the body. It is this conclusion, I suggest, that manifested itself in the extinctionist answers that children gave in response to the corpse story.[115]

Given what I said at the outset about Harris’ theoretical interest in testimony, the reader might be surprised to see so little of it in the account I have given so far. But the fact is that Vezo children receive very little explicit and explanatory testimony from adults, who claim that children lack any wisdom that would justify talking to them about such “difficult” things as death and a host of other topics. This does not mean that children live in a world devoid of testimony: they extract it as they avidly listen in to adults’ conversations and, probably more importantly, they receive it (as well as create it) as they interact, play, and investigate the world around them in the company of other children. What they learn through these interactions is hard to document ethnographically: children are always on the move, prodding and poking, chatting and laughing, fighting and crying. But I fully recognize that Vezo children learn through and with others as much as they learn through their first hand experiences—as we are about to see when we move on to the second point that emerged from the results of the death interview.

The second point is that, when primed with the tomb story, children were swayed to attribute to the deceased not only mental properties but also a range of bodily functions. What is the source of this very specific pattern of responses?

One possibility is that children get this idea from observing adults’ interactions with the ancestors. As discussed earlier, adults perform a range of actions—they talk to the ancestors, they offer them food, they pour rum over their tombs, they play loud music for them, they provoke them with sexual banter—which can give the impression that the deceased retain a sentient and responsive body. I have two reasons to believe, however, that this is not why children judged that the character of the tomb narrative would preserve bodily functions.

The first one is that children are eager but far from attentive participants in the rituals. They are loud and joyful—which is what the ancestors are imagined to expect and to enjoy—and, as their parents and elders often remarked, they are only ever interested in the food that gets cooked and distributed; as a result, they hardly notice what else is going on. Although adults are not entirely correct in their assessment (children actually know more about the proceedings than adults seem to think, see Astuti 2011) they do have a point: children’s focus seems indeed to be largely on the food that they expectantly wait for and boisterously consume. The intriguing fact that it is cooked for people whose bodies have rotten away—and that words are spoken to, and music is played for, people whose ears are no more—may well pass them by.

The second reason is that there is a much more obvious and—if you are a Vezo child—more compelling source that can account for children’s distinctive pattern of judgments. The source are the stories, in circulation among adults and children alike, about people who have died, have been buried as normal but, for reasons that remain quite vague, reemerge, fully alive, from under the sand. When they make their way back to the village, their families send them away, frightened and ashamed. As a result, they are forced to live in hiding, entering the village at night, stealing people’s leftovers and, more typically, the food left out for the pigs. If sighted, they are easily recognizable because they smell bad and they are draped in white cloth.

When I approached children for an informal chat or to observe their games, I found that they loved talking to me about such people, assuring me that they [116]had encountered one or two when they were out in the forest to collect firewood, and competing with each other to demonstrate their bravery. Over time, I began to notice that whenever I asked them open-ended questions about the ancestors and about their surviving spirit (which I would do only after they had participated in the death interview) the children would often end up talking about wandering people that enter the village at night to steal food from the pigs’ troughs. In other words, they seemed to conflate two very different entities: the disembodied spirits of the ancestors and the resurrected bodies of a few unfortunate and scary people. And if this conflated entity is what the tomb story evoked for the children, then it makes sense that they would attribute to the deceased a host of bodily as well as mental functions.

Eventually, this way of imagining the ancestors will give way to that of an immaterial entity—albeit with embodied sensations—which is common among adults. I was able to gain some insights into this transition thanks to a handful of interviews (sixteen in total) with sixteen- and seventeen-year-olds. This is an interesting age group because, as they begin to have children and families of their own, they are drawn more and more into having personal responsibilities about following Vezo customs, obeying ancestral taboos, contributing to the financing of ancestral rituals, and so on. No longer children, they are expected to know and to understand “difficult” stuff. And so they do, at least some of it. In particular, and notwithstanding the limitations of my small sample, it was quite clear that the adolescents (as I shall call them as a short hand) were quite different from the children, as they were not swayed to attribute bodily functions when presented with the tomb narrative. Unlike children, adolescents had understood the difference between existing as a disembodied spirit and existing as a resurrected body. In my conversations with these youths, it became apparent that the differentiation between the two entities was driven by the realization that the ancestors are invisible. Thus, while a ten-year-old would maintain that one cannot see the spirits of the ancestors because they hide in the forest (which is what the resurrected people are typically forced to do), a sixteen- or seventeen-year-old would maintain that one cannot see them because they are like air—they move about but they are invisible to the eye.

When I asked a particularly articulate seventeen-year-old—who seemed to know everything that I knew about the disembodied existence of the ancestors—how he had come to know what he knew, he told me that “there was no teaching me this stuff, no there wasn’t.” Instead, when he was little, he would listen in when adults talked about such things so that, little by little, he knew because—he added modestly—being a “little person” rather than “a person from the past,” he actually had to learn the stuff by listening to the stories of his elders (see Astuti 2011 for more details on this particular youth).


Let me now bring the various threads together and tackle the puzzle I started off with: that Vezo people represent and experience the ancestors in apparently contradictory ways.

If I had to weave just one thread and condense the whole story in just one line, I would say the following: that the contradiction exists because of the multiple [117]sources that feed into children’s, adolescents’, and adults’ knowledge of death and of the ancestors.

Here are some of the sources I have identified: the more or less violent interactions that children have with animals; the experience of the stench and of the stillness of dead human bodies; the emotional commotion that surrounds human death; the exciting stories about the people who die but come back to life with a full functioning body; the conversations one overhears; the words of the elders that, by adolescence, are listened to attentively and respectfully; the dreams featuring dead people, which get recounted and discussed to decide what actions need to be taken; the announcements and the offerings of food and drinks directed at the ancestors that one witnesses and participates in.

This is only a partial list, since there is much more going on in people’s lives that I did not have the space to mention: the visits to the diviner who tells you that a persistent illness is caused by the anger of a displeased ancestor; the highly charged acts of separation that mark the end of funerals, such as children being forced to come face-to-face with their dead mother, just before the coffin is nailed shut, while being told, forcefully and loudly, that she is dead and that they shall never again utter her name; the palpable anxiety that pervades the performance of any ritual as it is vital to get every detail absolutely right: the position of the pot, the time of day and the day of the week, and even the length of the tail of the cow to be slaughtered.

I could go on, but I shall not since it is obviously impossible to capture all the contexts and interactions—ranging from fleeting and negligible to consequential and transformative—that contribute to this process of learning and knowing. All I wish to underscore is precisely the great number of sources that feed into it. And it is not just a matter of how many they are. It is also a matter of their qualitative difference. For the children who learn about the process of death by decapitating a bird or poking at the heart of a dying turtle are learning in a different way from the adolescents who realize that what they need to know can only come from the words and the actions of their elders. And the adults, who fret about the position of the pot or the length of the cow’s tail, know what they know about the conduct of the ritual in a different way from what they know about the coldness, the stillness, and the stench of a dead body.

The difference is captured by two expressions often used by Vezo adults. One expression refers to knowing something because you have seen it with your own eyes (hita maso): this, for example, is how children might claim to know about the rotting of a dead body. The other expression invokes the ways of doing of the big people of the past (fomban’ny olo be taloha) as the reason for doing, saying, knowing something. To use Maurice Bloch’s conceptual vocabulary, the difference captured by these two expressions is the difference between, on the one hand, one’s critical and inquisitive engagement with the world—which of course includes other people, their words, and actions—and, on the other, deference, one’s willingness to trust and follow others (Bloch 2004).

Deference is most clearly in evidence in ritual contexts when adults will explicitly say that they themselves do not know, for example, why the pot has to be put in a certain position or why it matters that the tail of the slaughtered cow is of a certain length. They do not know but they trust that somebody else does: if not the elders, then the big people of the past who have handed down such customs and taboos.[118]

These two different ways of knowing create a kind of fault line—to borrow a term employed by Harris (2000: 175)—between what can be usefully subjected to critical and inquisitive scrutiny and what cannot. In the case of the middle-class Euro-American children studied by Harris, the fault line gets established as they fire off questions after questions and gradually realize that some of these questions (for example, about why angels don’t fall through the sky) will not get a satisfactory answer. For a while, children will continue to contemplate in their imagination exactly how angels can stay up in a floor-less sky, searching for a way to map this fact onto what else they know about the world (for example, their knowledge of gravity). But eventually, they will give up questioning and just take it on trust that “somehow” that’s where angels are—in the sky and on the other side of the fault line (Harris 2000: 174–75).

Vezo children do not ask questions—at least not in their interactions with adults. But recall the conflation they make between the body of resurrected people and the disembodied spirit of the ancestors. This conflation could well be driven by their attempt to bridge the gap between what they hear (that the ancestors are cold and that they want to be fed) and what they know (that living people like themselves experience cold and hunger through their body). And although they might have never seen “with their own eyes” a dead body come back to life, they might find this eventuality more coherent and easier to contemplate than that of a disembodied but sentient spirit. Still, by the time they are adolescents, they will have accepted on trust that “somehow” the ancestors are like air but nonetheless need to be fed and kept warm.

The fault line between qualitatively different ways of knowing explains why human beings—not just the Vezo—can comfortably hold contradictory views, for example, about the nature of the ancestors. This is because the “contradiction” is not generated by incompatible evidence of the same kind, but by different kinds of knowledge, which are used in radically different ways and for different purposes. As a result, what people know with their own eyes, on one side of the fault line, and what they know on trust, on the other side, need not become integrated, all contradictions ironed out. So, when people say that “when one’s dead, one’s dead”—which they know to be the case because, from the time they were children, they discovered the devastating consequences of death—they are not actually contradicting what they know when they worry about a dream that features a dead relative, when they talk to the ancestors or cook food for them—which they know to be the case because, “somehow,” this is what needs to be done, trustingly following in the footsteps of the big people of the past. There is thus no “cognitive dissonance” (Festinger 1957; Cooper 2007) between their actions and their beliefs, since one set of actions (for example, treating the corpse as a nonsentient entity) is motivated by knowledge that is constructed on one side of the fault line, and another set of actions (for example, relating to the ancestors as sentient beings) is motivated by knowledge that stems from the other side.

In the flow of everyday life, people move from one side of the fault line to the other, and when they do so, they signal the transition in subtle ways—a change in body posture, the clearing of one’s voice, a little laughter—or they do so more starkly, as when an elder marks the beginning and the end of a ritual, bracketing off a time when deference prevails.[119]

This movement back and forth brings me back to the death interview, for in the light of what I have just said I need to comment on the kind of knowledge that I elicited with my hypothetical stories and odd questions. One striking finding was that across individual participants there was little consensus about which and how many of the various properties they were asked about would remain viable after death. Such lack of consensus is a tell tale sign of how participants engaged with the task: they critically considered my questions, they deliberated over them, and they produced their idiosyncratic answers. This is exactly what I asked them to do, as I reassured them that my questions did not have right or wrong answers since people have different opinions about such things, and that I was interested in “your own way of thinking.”

With this, however, I do not want to imply that the death interview presented people with an altogether novel experience. For sure, the task created an extremely artificial context that forced participants to think things through: they paused, they pondered, they resolved. But the fact that they were willing to participate in the exercise and were often eager to take it further—in the lengthy open-ended conversations that followed the more formal part of the interview—revealed that at least some people (and in this respect too there is variation) are quite keen to make incursions across the fault line, as they ask themselves how exactly things work and spend time and effort trying to find some answers for themselves. Regardless, there is no doubting that the death interview most definitively engaged participants’ nondeferential stance. And yet, the fact that those participants who heard the tomb story were more likely to judge that the deceased’s mental functions remain viable after death suggests that the task, despite its odd questions, managed to push them, ever so slightly, toward deference. They too had to think—they paused, pondered, and resolved—but the tomb narrative was seemingly just enough to make them suspend judgment and resolve that, “somehow,” the hypothetical character of the story would retain at least some of the mental properties they were quizzed about.


At the start I stated that I would present a different way of “taking people seriously”—different from what has been promoted by colleagues who have taken the ontological turn. I hope to have delivered on this by showing in some detail how I work. To conclude, I shall highlight what I regard as my key moves.

My first move is to take seriously the fact that my Vezo interlocutors say things that do not quite add up. For me to be able to hear this, I had to listen to them carefully, at different moments of their lives, when they are engaged in radically different activities, such as dealing with the corpse of a dead relative, recounting a dream, or fretting about a ritual. This is why ethnographic fieldwork remains an essential part of my way of taking people seriously.

My second move is to take seriously “the people,” by actually focusing on them, for it is people that construct knowledge and ideas, not the other way around. When people—living human beings—are firmly center stage, one is bound to see temporal and contextual transformations (from child to adult, from grief to anxiety, from deference to inquisitive experimentation) that make it impossible to give a one-dimensional account of the supposedly radically different way in which they conceive the world.[120]

My third move is to take seriously the fact that people are critical and inquisitive as they interact with the world around them and with other people. I therefore find it helpful to incorporate in my ethnographic endeavor methodological tools that invite people—children and adults alike—to think about novel riddles and see where this takes them and my understanding of them.

My final move is to take seriously the possibility that, as an anthropologist who has devoted her career working with a small group of Vezo folks in Madagascar, I can discover something general about human beings: for example, about the different ways in which they learn, about the way they use their imagination to work things out, about the fact that they are capable and willing to defer to others they trust. Working alongside cognitive and developmental psychologists, who are also in the business of discovering what we all share as members of the human species, has helped me immensely to hold on to this aspiration. So for me, the interdisciplinary effort has definitively been worthwhile.


Astuti, Rita. 1994. “Invisible objects: Funerary rituals among the Vezo of western Madagascar.” RES: Anthropology and Aesthetics 25: 111–22.

———. 1995. People of the sea: Identity and descent among the Vezo of Madagascar. Cambridge: Cambridge University Press.

———. 2000. “Kindreds, cognatic and unilineal descent groups: New perspectives from Madagascar.” In Cultures of relatedness: New approaches to the study of kinship, edited by Janet Carsten, 90–103. Cambridge: Cambridge University Press.

———. 2007. “What happens after death?” In Questions of anthropology, edited by Rita Astuti, Jonathan Parry, and Charles Stafford, 227–47. Oxford: Berg.

———. 2011. “Death, ancestors and the living dead: Learning without teaching in Madagascar.” In Children’s understanding of death: From biological to supernatural conceptions, edited by Victoria Talwar, Paul. L. Harris, and Michael Schleifer, 1–18. New York: Cambridge University Press.

Astuti, Rita, and Paul L. Harris. 2008. “Understanding mortality and the life of the ancestors in rural Madagascar.” Cognitive Science 32: 713–40.

Bloch, Maurice. 2004. “Ritual and deference.” In Ritual and memory: Towards a comparative anthropology of religion, edited by Harvey Whitehouse and James Laidlaw, 65–78. Walnut Creek, CA: Altamira Press.

Cooper, Joel. 2007. Cognitive dissonance: Fifty years of a classic theory. London: Sage Publications.

Festinger, Leon. 1957. A theory of cognitive dissonance. Evanston, IL: Row Peterson.

Graeber, David. 2015. “Radical alterity is just another way of saying ‘reality’: A reply to Eduardo Viveiros de Castro.” HAU: Journal of Ethnographic Theory 5 (2): 1–41.[121]

Harris, Paul L. 2000. “On not falling down to earth: Children’s metaphysical questions.” In Imagining the impossible: Magical, scientific and religious thinking in children, edited by Karl S. Rosengren, Carl N. Johnson, and Paul L. Harris, 157–78. Cambridge: Cambridge University Press.

———. 2012. Trusting what you’re told: How children learn from others. Cambridge, MA: Belknap Press/Harvard University Press.

Harris, Paul, and Melissa Koenig. 2006. “Trust in testimony: How children learn about science and religion.” Child Development 77: 505–24.

Harris, Paul L., and Marta Giménez. 2005. “Children’s acceptance of conflicting testimony: The case of death.” Journal of Cognition and Culture 5 (1–2): 143–64.

Henare, Amiria, Martin Holbraad, and Sari Wastell, 2007. Thinking through things: Theorising artefacts ethnographically. London: Routledge.

Holbraad, Martin. 2012. Truth in motion: the recursive anthropology of Cuban divination. Chicago: University of Chicago Press.

Kopytoff, Igor. 1971. “Ancestors as elders in Africa.” Africa 41: 129–42.

Pedersen, Morten Axel. 2011. Not quite Shamans: Spirit worlds and political lives in Northern Mongolia. Ithaca: Cornell University Press.

Prendre les gens au sérieux

Résumé : Prendre au sérieux les personnes que nous étudions a refait surface récemment comme une des ambitions clés de l’entreprise ethnographique et anthropologique. Dans cette lecture, je présente ma façon de prendre les gens au sérieux (en particulier, mes amis Vezo d’un village de pêcheurs à Madagascar). A mon avis, il est nécessaire pour cela de comprendre les origines de leur savoir, et les différentes façons de connaître le monde qu’ils mobilisent en fonction des contextes et à diverses fins, selon leur âge, et animés par différentes experiences et resources cognitives. Je développe cet argument en m’appuyant sur des matériaux empiriques issus d’une collaboration interdisciplinaire incluant des anthropologues et des psychologues cognitivistes.

Rita ASTUTI is an anthropologist with interests in kinship, personhood, gender, group identity and belief, and is Professor of Anthropology at London School of Economics. She has conducted extensive periods of fieldwork in a fishing village in Madagascar. She has published an ethnographic monograph on kinship, death, and identity, People of the sea: Identity and descent among the Vezo of Madagascar (Cambridge University Press, 1995) and, with developmental psychologists Susan Carey and Gregg Solomon, she reanalyzed her ethnographic findings through the use of experimental techniques, in Constraints on conceptual development (Monographs of the Society for Research in Child Development, 2004). In her publications [122]and in her teaching, she has advocated a closer integration between anthropology and psychology.

Rita Astuti
Department of Anthropology
London School of Economics
Houghton Street
London WC2A 2AE


This article is a revised version of the Robert H. Layton Lecture, delivered at Durham on October 7, 2015. I would like to thank Maurice Bloch, Sean Epstein and Evan Killick for comments and discussions.