The last fifteen years or so have seen the emergence of a self-conscious anthropology of Christianity.Footnote 1 This scholarly development has from the outset been framed as a comparative endeavor. Part of its promise has been that it will allow people working in different parts of the world to ask comparative questions about how processes of Christianization have unfolded in the places that they work and about how the resulting Christian configurations are similar to and different from one another. As the anthropology of Christianity moves into its second decade, one can say that in practice some of this comparative work has already begun to appear. There are now broad cross-regional dialogues on such issues as language, media, discontinuity, conversion, and individualism. Yet it is equally true that there has been very little said in theoretical terms concerning how to think about comparing Christian groups, and few if any works have appeared formulated wholly around comparative projects. This paper, co-written by two Melanesianists and an Amazonianist, aims to help address this gap. Its goal is to exemplify one new kind of comparative effort that can be undertaken when commitment to one or other strand of the Protestant tradition is shared by all of the groups under study.
One can argue that the appearance of the anthropology of Christianity as a comparative project around the turn of the millennium renders it part of a broader upsurge in anthropological interest in comparison. At the same time that the anthropology of Christianity was gathering steam, several important publications appeared that called for a reconsideration of the value of comparative anthropological research (Gregor and Tuzin Reference Gregor, Tuzin, Gregor and Tuzin2001b; Fox and Gingrich Reference Fox, Gingrich, Gingrich and Fox2002; Keane Reference Keane2003). The authors of these pieces argued that anthropology had for several decades largely ignored comparison both because of the lingering legacy of Boasian particularism and, of special importance during the last quarter of the twentieth century, because of the success of postmodern arguments in casting doubt on both the cross-cultural applicability of comparative categories and the grand theoretical programs that had grounded earlier comparative efforts (Gregor and Tuzin Reference Gregor, Tuzin, Gregor and Tuzin2001b: 3, 5; Fox and Gingrich Reference Fox, Gingrich, Gingrich and Fox2002: 4; Keane Reference Keane2003: 234, 241). They then go on to suggest that after having left comparison aside for so long, the time has come for anthropologists to revisit its intellectual potential.
The anthropology of Christianity is well positioned to contribute to such a resurgence of anthropological comparison for at least two reasons. First, Fox and Gingrich (Reference Fox, Gingrich, Gingrich and Fox2002: 7) have argued that it is possible to put comparison on a new footing because globalization means “people around the globe are increasingly reacting to comparable conditions,” challenging scholars “to compare how people react and what results culturally from their reactions.” Anthropologists of Christianity are highly aware of this, focusing as they most often do on forms of a global religion that have spread quite widely and that in doing so have confronted members of many societies with such “comparable conditions” to which to react. Having acknowledged this fact, anthropologists of Christianity have developed a range of theories of conversion and cultural change designed to explore how processes of reaction have unfolded in different places, and the anthropology of Pentecostalism has at least implicitly been relying on something like this framework for some time (Robbins Reference Robbins2004b). Second, in a recent discussion of comparison, Keane (Reference Keane2013: 7) urges anthropologists to consider the “affordances” provided by various aspects of the material world with which humans interact. While not all people who come into contact with similar material forms will take up all the affordances they provide, the possibility of exploring comparatively cases in which they pick out similar ones, and those in which they do not (ibid.: 13), provides a new grounding for comparison. Keane's own interest is in studying the affordances provided by “the materiality of language and people's encounters with it,” particularly with formulaic and written language (ibid.: 13).
Anthropologists of Christianity have only sometimes focused on materiality as such, but they have generally adopted an understanding of their object of study that fits well with Keane's program. The understanding in question is one that sees forms of Christianity as cultural in themselves. This is not to say that all forms of Christianity share a single culture or cultural substrate, but that like any humanly intelligible phenomena capable of shaping social action, all forms of Christianity possess systems of linked cultural categories and promote patterns of institutionalized interaction that converts can take up, and that in situations meaningfully defined as Christian they do take up to one or other extent. This is to say, the globalizing forms of Christianity that Fox and Gingrich encourage us to follow as they move into various social formations around the world possess precisely the kind of structured, durable, or “material” quality that Keane suggests enables us to examine comparatively how the affordances they provide are taken up (or disregarded) by those who encounter them. Our own efforts in this paper are in line with these recent attempts to develop new frameworks for comparison, and we hope to demonstrate that anthropological studies of Christianity can exemplify the value of putting them to use.
We begin our analysis from the observation that various traditions of evangelical Christianity tend in most of the places they travel to afford people the opportunity to engage a linked set of problems having to do with cultural categories and practices related to what we will call the “self.”Footnote 2 This should at least on some level be an uncontroversial claim. As Mauss (Reference Mauss, Carrithers, Collins and Lukes1985) observed, the modern Western notion of self is certainly not universal. Instead, like all cultural representations, it has a history. In the case of the modern self, the long genealogy Mauss provides culminates in certain forms of Protestantism that finally emancipate the self as consciousness from the social roles with which the understanding of the person had previously been entangled:
We cannot exaggerate the importance of sectarian movements throughout the seventeenth and eighteenth centuries for the formation of [modern] political and philosophical thought. There it was that were posed questions regarding individual liberty, regarding the individual conscience and the right to communicate directly with God, to be one's own priest, to have an inner God. The ideas of the Moravian Brothers, the Puritans, the Wesleyans and the Pietists are those which form the basis on which is established the notion: the ‘person’ (personne) equals the ‘self’ (moi); the self (moi) equals consciousness, and is its primordial category (ibid.: 21).
The Christian groups Mauss is writing about here are part of the lineage that has issued in modern evangelicalism. Our claim is that wherever forms of such evangelicalism have alighted and found local people ready to engage them in culturally meaningful terms, one of the things those people have had to reckon with is how to understand and live with some version of this modern self.
Uncontroversial as we hope this basic point is on the surface of things, we also recognize that in placing the self at the center of our discussion, we raise challenging questions concerning how one is to define this self in specific terms and how one is to make comparable the sometimes different versions carried by different evangelical traditions, not to mention the myriad forms these take once put in play in local contexts. In the anti-comparative moment of the last decades of the last century, questions such as this about the diversity of models of selfhood in Christianity would have been enough to sink our whole project under the weight of what Robbins (Reference Robbins2003) has elsewhere called “object-dissolving critique.” But we want to argue that there are ways to complicate our object without losing it altogether; and indeed, the promise of such complication is, we hope to show, part of what should motivate the project of comparison.
To see how this is so, we want to introduce at this point one further important recent statement on comparison. This one issues from the pen of the anthropologically sophisticated classicist Marcel Detienne. In his 2008 book entitled Comparing the Incomparable, one of his primary concerns is determining, as he puts it in the title of his second chapter, the best way of “constructing comparables.” The task, he suggests, is to develop comparative categories that are “generic enough to allow the beginnings of a comparison but neither too general nor too specific to any given culture” (2008: 25). Moreover, what the analyst aims to study is not any single category itself as it may be reproduced in its entirety in several cultures, but rather the different “choices” people have made in adopting elements of such a category and the ways those elements chosen are constrained in their own relations to each other by their need to work together in coherent ways (ibid.: 32). As Detienne puts it: “The comparables that we were setting up … could thus be said to be orientations, interlinked choices: choices made in preference to other possibilities. When a society … adopts a particular element of thought, it makes a particular choice that might have been different. The job of the … analyst is to discern the constraints that affect the configurations that he or she is studying…. what are ‘comparable’ are … interconnecting plates determined by some initial choice” (ibid.).
As we read Detienne, his point is that when presented with new cultural configurations and, as Keane would put it, the affordances they provide, people make choices about which elements to take on. Having made a choice, they are then “constrained” to work with these elements in coherent ways—that is to say, they will find themselves engaged with some or all of the cultural logic such configurations carry with them in the linkages between their elements. Thus, for example, if people take up a Protestant notion of an inner self that can communicate directly with God, it will be hard for them to claim that the person's relationship to God has no bearing on their salvation. The reason such constraints are important is that they help to ensure comparability. While we expect very broad diversity in the choices people (both members of different missionizing Christian groups and the people they convert) have made about what parts of the Protestant self to engage, the diversity their choices produce is not wholly open-ended or boundless; the way people are constrained to work with the logics linking the elements they have chosen gives us confidence that even quite diverse versions of the Evangelical Christian self will be comparable to one another. Studying the range of possible diversity, and how the diverse outcomes are formed, is the goal of the comparative project.Footnote 3
One further aspect of Detienne's approach to comparison is important to us. Because he is interested not in comparing general categories but rather the choices people make among the elements that make them up, he expects that comparison will frame issues in ways that do not accord with disciplinary “common sense” or rely on “immediate ‘givens’” (ibid.: xi). Often the familiar categories one starts with, such as “self” in our case, will “fracture and disintegrate” as a comparative project develops (ibid.: 26). This has proven true with the category of self in our case. For the purposes of our project, we have found it necessary to break self down into several elements. These are a notion of mind (which includes the idea of an inner locus of thought, feeling, and motivation); a notion of personal identity or boundedness (which includes ideas about the relationship of the mind to the body); and a notion that the person is set within social and cosmic worlds that require it to relate to a number of kinds of others (human, divine, animal) through various forms of communication. In missionizing Western forms of Protestantism, all three of these elements are present, and are generally tied together by an overarching individualism that we can define, following Dumont (Reference Dumont1986), as a stress on the fact that a person is saved as an individual on the basis of actions they are understood to initiate themselves that bear on their relation to God. Work already published in the anthropology of Christianity indicates that each of these elements of the notion of self have been important in different ways to Christian converts in various parts of the world, and this suggests the comparative value of taking this set of ideas as making up a Christian model of the self with which converts often find themselves engaging (e.g., van Dijk Reference van Dijk and Werbner1998; Keane Reference Keane2007; Luhrmann Reference Luhrmann2012; Robbins Reference Robbins2004a).
In this paper, we examine how the Amazonian Wari', and the Melanesian Bosavi and Urapmin, have made different choices about adopting Protestant notions of mind, personal boundedness, and communication, as well as the individualism that in Western Protestantism generally ties them together, and have worked the elements they have adopted into a number of distinctive but comparable ways of thinking about the Christian self. We recognize that by fracturing the “self” into its more basic elements we approach this topic in a somewhat unfamiliar way (the stress on communication in particular might be surprising to some readers, though we aim to show that it is warranted by the ethnographic materials). We hope that the results of our comparative effort will prove the value of taking this de-familiarizing tack.
As a final note before commencing to present the three micro-ethnographies that make up our cases for comparison, we should note that just as our approach fractures some commonsense notions of “self,” it also runs against the grain of some previous approaches to comparing Amazonian and Melanesian societies. We do not, for example, foreground gender as a meeting point between Amazonia and Melanesia in the way most of the contributors to Gregor and Tuzin's (Reference Gregor and Tuzin2001a) comparative volume on the two regions have done (see Vilaça Reference Vilaça2005: 460, n. 12; Reference Vilaça2011: 244). This is not to say that gender is uninteresting in this regard (though see Descola Reference Descola, Gregor and Tuzin2001). Rather, we do not focus on it because our cases do not highlight it—and indeed, Protestant Christianity in some places at least appears to put gender differences more in the background of converts' lives than it had been previously by virtue of the way it emphasizes the potential Christian equivalence of all devout selves (see Robbins Reference Robbins2012).
Perhaps of more moment for present purposes, our comparison does not ultimately turn on comparing or contrasting with one another the currently important perspectivist approach to Amazonian cultures and the equally influential relationalist approach to Melanesian ones. While some work has already been done on this topic (see Strathern Reference Strathern1999: 252–53; Robbins Reference Robbins, Vilaça and Wright2009: 236–37; Vilaça Reference Vilaça, Vilaça and Wright2009; Reference Vilaça2011), our elements of the self crosscut these two theoretical-cum-ethnographic approaches at a more fine-grained level of resolution. Our case study of Amazonia thus draws on themes relevant to perspectivism, just as our Melanesian cases touch on important relationalist themes. But when it comes to carrying out our comparisons between our three cases, we combine these themes in new ways. We hope that a secondary achievement of our paper may thus be suggesting new ways to bring these regional research programs into dialogue.
CHRISTIAN LANGUAGE SOCIALIZATION AND SELF (RE)FASHIONING IN BOSAVI, PAPUA NEW GUINEA
The Bosavi people live north of Mt. Bosavi in the Southern Highlands of Papua New Guinea.Footnote 4 In this rainforest environment, two thousand or so Bosavi people inhabit scattered longhouse communities ranging from sixty to one hundred people.Footnote 5 Traditionally egalitarian, they practice swidden horticulture, hunt, and fish. During the research period reported here (1975–1995), most people were monolingual speakers of the Bosavi language, though this was changing.
Although Bosavi people first encountered non-Indigenous people in 1935, there was very little interaction between that date and 1964. At that time two Australian members of the Unevangelized Fields Mission, a fundamentalist, nondenominational, Protestant “faith” mission made contact and began constructing a small airstrip and mission station. But it was not until the early 1970s, when an Australian couple, members of the Asia Pacific Christian Mission (formerly the Unevangelized Fields Mission) began intensive proselytizing activities around the mission station, that things began to quickly change.
Convinced that the Second Coming was imminent, the missionaries' goal was rapid conversion. They emphasized a doctrine of “last things”—death, judgment, heaven, and hell, and elaborated the dire consequences for nonbelievers. They opposed what they called dramatic, emotional, and visible signs of Christian conversion, such as speaking in tongues and other charismatic gifts. They also rejected the idea that knowledge of local cultural practices might be helpful to their own agenda, and viewed such practices as not only irrelevant to conversion, but as obstacles to its success.
The missionaries considered the Bible the center of all preaching, and as fundamentalist Christians they took a literalist stance toward Scriptural translation and interpretation. Lacking linguistic training but fluent in Tok Pisin, they mediated Christianity through that language and the Nupela Testamen (New Testament, published in Tok Pisin in 1969). This limited the people they could interact with to a small group of younger men who had learned some Tok Pisin working outside of the area. These men lacked formal schooling but were interested in what the missionaries promised both spiritually and materially. At the mission station, they acquired basic reading skills in Tok Pisin and learned about missionary life. By 1975 several were baptized, and sent back to the villages as pastors to preach and convert others.
While not all Bosavi people were interested in Christianity, from 1975 onward Christian activities reorganized village life. Services held several times a week centered on pastors reading biblical passages in Tok Pisin, which they spontaneously and literally translated back into the Bosavi language. These recently missionized Bosavi men became active missionizers and over the next two decades produced the language and conceptual framework through which people understood Christianity. During this period, some Bosavis were baptized while others chose not to participate.
Frustrated by the lack of total conversion and disappointed by the back-sliding of many early converts, the missionaries left Bosavi in 1990, leaving Christianization to local pastors. Without mission support, however, the school, clinic, and airstrip rapidly deteriorated and their operation became erratic. While pastors held services through the 1990s, government and resource extraction projects increasingly pulled local men away, further fracturing the small Bosavi villages. By the end of the twentieth century, Christianity had an unstable future as many Bosavi people questioned what missionization had done for them beyond disabling traditional cultural practices.
Bambi Schieffelin's research on Christianization is based on ethnographic and sociolinguistic fieldwork in Bosavi (1975–1995). Bosavi people tended not to talk about Christianity in an abstract or speculative manner; similarly they did not engage in ethnopsychological discussions about the changes they experienced or observed. Transcribing audio-recorded church services with pastors and others, however, generated extensive metacommentaries, providing insights into the lives of first-generation Christians and the role of language in changing how they think and talk about themselves and others.
Highlights of pre-Christian Bosavi language socialization practices through which children acquire appropriate ways of speaking, feeling, and acting, as well as patterns of Bosavi language structure and use pertinent to local notions of bodies and selves, contextualize these changes. Christian missionization introduced new languages and put language ideologies into contact, changing the relationships between speaking, acting, and thinking in the construction of the self. Local pastors applied the mission's literalist orientation to Scripture as they translated it into the vernacular, extending Scriptural messages beyond their textual boundaries. This inverted several key cultural concepts about the nature of persons, and transformed ideas about bodies, selves, and the nature of communication itself.
Traditional Bosavi Bodies and Selves: From Soft to Hard
Before Christian contact, Bosavi people, like many other Melanesians, when asked by anthropologists explained how the body (do:mo: Footnote 6) was “made” in utero from different parental contributions. While little was made of this in everyday settings, it represents intertwined strands of social relatedness and sociality primarily shaped by kinship that are significant throughout a person's life in this agnatic society.Footnote 7 For a child's development, active nurturing by parents and other close kin, shared experiences, and participation in social activities were of utmost importance. Given these conditions, a person developed life-long relationships based on reciprocity and exchange mediated by material objects, the basis of affective connections and relatedness, which were central to establishing ontological security.
While Bosavi people discussed some of their notions about bodies and selves, their most significant concepts became evident through analyzing their talk to their children and other people and their child raising practices and taboos. As previously detailed (Schieffelin Reference Schieffelin1990), Bosavis say babies are taiyo: ‘soft,’ have no understanding, and naturally beg. People feel sorry for them, and give them what they want based on their appeals. This changes at the onset of language, marked by the child's use of two words ‘mother’ and ‘breast,’ key words signifying social relatedness and its earliest mediation. To enter the world of social reciprocity children must be able to be held accountable for their own desires, and this requires that they verbally express what they want through assertive language. Through language socialization activities, (socialization to use language and socialization through the use of language [Ochs and Schieffelin Reference Ochs, Schieffelin, Shweder and Levine1984]), which occur throughout the life cycle, Bosavis acquire the discursive and cultural knowledge necessary for participation in community life.
To accomplish this, mothers initiate extensive verbal routines, “showing” language by modeling what they take as appropriate utterances and directing children to repeat them to another addressee. Addressees respond appropriately to the child, and through guided participation in these conversational sequences small children become adept at repeating exactly what someone else has told them to say, with appropriate affect. They perform pragmatically complex utterances before understanding their full meaning, which they acquire through the contextually contingent responses of others. In this way they come to understand the efficacy of socially appropriate performances. While primarily, but not exclusively directed to children, how Bosavis “show” each other new forms of knowledge through repetition of a model conveys the importance of paying close attention to how something is performed. The ability to say or do something in a particular or formulaic way is positively valued, tied to social and pragmatic convention as well as ritual efficacy.
Such routines are central to becoming communicatively competent, what Bosavi people called halaidan ‘hardening,’ which includes strength, maturity, and assertiveness—all positively valued and essential for participation in social life, even for small children. While similar in form to early language socialization routines, cultural interventions only begin once a child independently demonstrates interest or ability since Bosavi people claim that one person cannot make another do something that he or she is not already ‘thinking of himself/herself’ (ina:li asula:ga:), or has not initiated. Even from an early age, within a social world of interdependence, the self is recognized as bounded, having autonomy and agency.
Cultural and linguistic routines shape what children learn about internal states and how to talk about them, central to socially constructing and maintaining boundaries of the self. Children speak about their own internal states, but through language socialization activities are taught that visible and audible evidence play a critical role in assessing and talking about others' intentions and feelings. Children are discouraged from verbally speculating about others' unexpressed feelings or thoughts, or even verbally guessing at others' unclear utterances. These activities further display how members grant each other psychic privacy through specific verbal routines, for example, rhetorical questions, which register stance toward or assessment of another's actions, without expecting an answer or account (example, “is it yours?!” = it isn't). Rhetorical questions, used to tease and shame, are a powerful form of social control aimed at inclusion and conformity, while also protecting the privacy of one's internal states or motives. Children also learn that these and other types of utterances have both literal and metaphorical meanings, and when and how to determine them. These early practices establish Bosavi communicative and interpretive frameworks in which one does not expect that an addressee will search for a speaker's intentions in utterances, but rather, will privately co-construct meaning according to his or her own desires. Furthermore, these routines also lay the groundwork for interpreting and producing utterances with non-literal meanings that are central to how meaning was and was not shared in public discourse. Thus language socialization routines shaped by Bosavi language ideology enabled participation, providing the content, forms, and interpretive procedures for arriving at meaning.
The Bosavi language was also a guide to apprehending the domain of internal states. Names of internal organs (kuf ‘stomach’; yo:g ‘liver’; himu ‘heart’; misa: dubus ‘brain’) denoted anatomical objects, but were not associated with particular functions, verbs of cognition, or internal states.Footnote 8 The forehead (wo:lokan) displayed worry, anger, relaxation, and happiness, among other states.Footnote 9 When Bosavis talked about others' feelings, they looked on the body, and only described visible signs. Thus, reporting that someone was smiling (not happy), or crying (not sad) avoided verbal speculation about unexpressed cognitive or affective states, though undoubtedly speakers inferred them.
Depending on the context, the verb asulab can mean ‘think,’ ‘know,’ ‘understand,’ ‘want/desire,’ or ‘miss someone or something,’ but does not locate the source. Terms denoting other internal affective and cognitive states, such as anger, sadness, or happiness similarly lack source, location, or association with an internal organ.
In terms of self-presentation, Bosavi people valued a confident, energetic demeanor, joking, teasing, arguing, and giving opinions; loud, exuberant, expressive multiparty conversations characterized village life, and dramatic exchanges and ceremonies marked weddings and other ritual events. Bodies were not only decorated with shells, feathers, paint, and other colorful ornaments on these occasions, but in ordinary circumstances as well people enjoyed looking good. While halaido: (‘hard’) characterized children's maturation and competence, it also described adults. Those characterized as halaido: (‘hard’) were admired for their strength, confidence, and health, evidenced by an assertive demeanor. While the term could be applied to women, it typically indexed a male vitality and dramatic style.
Bosavi Christian Bodies and Selves: From “Hard” to “Soft”
While missionization deeply affected many traditional cultural and communicative preferences, one's choice to become Christian, like other identity markers, was displayed publicly through particular embodied practices, which included language. The development and semiotics of these practices are relevant to an anthropology of Christianity and inform comparisons of the choice and uptake of Christian-identified demeanors. In Bosavi, the demarcation of Christian and non-Christian bodies can be linked to local pastors' literal translations and interpretation of Mark, the first Gospel translated into the vernacular. Guided by language ideologies, pastors selectively took portions as foundational for how Bosavi Christians should act and feel. Their translation practices illustrate how they understood Christianity and transmitted it in Bosavi.
Like other Protestant missions, the Asia Pacific Christian Mission privileged the mother tongue or vernacular, “the shrine of a people's soul” (Rule Reference Rule and Wurm1977: 1341Footnote 10), as the most effective means for converting native people because of its affective potential to reach the “heart.” The missionaries charged Bosavi pastors with translating Christian ideas and texts, which were not only new, but also presented in a language (Tok Pisin) in which local pastors were minimally literate. Consistent with the mission's literalist orientation to Scripture, pastors were instructed to stay very close to the written Tok Pisin when orally translating it from the Nupela Testamen into the vernacular during services. In Bosavi metalinguistic terms, that meant staying on the surface (wa:la) and not looking for non-literal or hidden meanings, which were underneath (ha:ga). While pastors were familiar with closely following a model (as in “showing” language), at another level, these instructions were in direct opposition to Bosavi language ideology that views language as having multiple meanings, both surface and underneath; what Bosavi called “turned over words” (Feld Reference Feld1982: 138–39). The mission's literalist, fundamentalist language ideology urged transparency and sincerity in communication. This challenged Bosavi pastors since the mission's views also contradicted local communicative preferences, which used indirection as well as other stylistic forms to avoid the attribution of responsibility, and took the performance of an utterance as its public meaning. Bosavi pastors sought lexical and semantic equivalencies in their vernacular, creating Christian idioms used in Christian language socialization practices to convey what Christianity was about.
Mission language ideology assumed that translating Christian concepts from Tok Pisin would be simple, transparent, and straightforward. It was anything but that, and the concept of sin illustrates cultural and translation challenges. The Tok Pisin word sin (from English) was meaningless. Looking to Gospel stories as contexts from which to extract the meaning, pastors turned to Mark 2: “Jesus heals a paralyzed man,” one of the first Gospel stories translated into Bosavi. A favored text for preaching for over twenty-five years, Bosavis identified Christ as a healer, and themselves as sick persons needing to be healed. Pastors expounding this view translated sin as walaf ‘sickness,’ and elaborated the performative dimensions of being sick, for example, speaking quietly, moving slowly, displaying low affect, and wearing no body ornaments or decoration. Additional social displays of being Christian included refraining from smoking, cursing, showing anger, lamenting for the dead, and participating in other “traditional” activities including receiving bridewealth or compensation.
New identity terms also entered the language. Baptized Christians, keriso: kalu (‘Christian people’), also referred to themselves as ‘people who go slow and easy’ (ha:sa ha:fa:no:lo: kalu), which took on a positive meaning. These linguistic designations of how the body looked further constructed selves needing to be healed. It complemented how Bosavi Christians were framed as children, who, in a relationship with Papa God needed to be looked after, always spoke softly, and appealed to him in prayer. In contrast, Christians labeled those who choose not to convert as ‘hard,’ negatively re-signifying the word to mean vain and obstinate. They inverted a valorized, semiotically rich binary (‘soft’/‘hard’; incompetent/competent) to also demarcate Christian and non-Christian bodies and selves.
Bosavis displayed their interest in becoming Christian, or not, most visibly on their bodies, but those desiring baptism also needed to refashion other facets of the self. One crucial aspect of this was the notion of belief (Tok Pisin bilip; Bosavi asugo: imilise, tilidabu), crystallized in the Western concept of sincerity, a prerequisite to prayer, confession, and conversion. For Bosavi people having this elusive quality defines a Christian. The concept of interiority, however, had to be established first. The source/location of internal states, including belief, could not be left unspecified; rather, they had to be conceptualized as originating inside of the body. Only by linguistically marking the source/location could the concept of interiority, and everything associated with it, be communicated and adopted. Drawing on Tok Pisin phrases, which lexically mark internal states as inside the body, specifically the stomach (Tok Pisin bel; Bosavi kuf),Footnote 11 pastors literally translated Tok Pisin Bible verses into the vernacular (loan translations or calques), accomplishing this transformation. They generated extensive linguistic innovations articulating the interiorization of affective and cognitive states, which were adopted into the vernacular. Some phrases even evoked a doubled interiority. († new Christian meaning)
asulo: I thought
† kuf-a: us-a asulo: I thought < Tok Pisin tingting long bel
stomach POSS inside LOC thought thought in stomach
More than simply shifting forms of linguistic expression, this Christian register enabled the constitution of a self according to a Western, and Christian moral geography that mapped emotion, cognition, and belief having an explicit interiority. To participate in this emerging Christian community Bosavis had to acquire new ways of speaking their vernacular, which made explicit new ways of thinking about themselves and others. While many Bosavis adopted these expressions as public acts of identity, they also expressed difficulty in grasping the meaning of scripture, and performed new genres such as prayer by rote and formulaically. Confession, however, was resisted by almost everyone, and was one of the major obstacles to conversion to Christianity in Bosavi (Schieffelin Reference Schieffelin2008).
Confession challenged Bosavi concepts of the self that privileged personal choice in when and how to talk about one's internal states. Confession and other demands for Christian accountings went directly against a Bosavi sense of autonomy—not being made to do something that one did not want to do. Confession also called for explanations of motives, and asked speakers to reveal what they “really thought,” challenging what they said as potentially untrue or insincere. Pastors accused those unwilling to confess of harboring secret, angry, or sinful thoughts such as stealing or coveting what was not theirs, but this did little to encourage confession, either to the pastor or publicly. While many who wanted to become Christians could enact the sick body that needed to be healed, they could not perform the required speech act—confession—that revealed their private internal states, and thus they were not accepted into the church.
Before missionization, reciprocity and exchange were the trusted enactments of social relationships, central to establishing and maintaining Bosavi community, past and present. Speaking and acting as embodied practices were the visible and audible evidence for grounding everyday sociality, while linguistic and social etiquette protected the psychic privacy of the self. Christianity disrupted these relationships: It shifted the primacy of connections to kin and community to one's relationship to God, who was said to know and judge not only what persons said and did, but what they thought. Through linguistic innovation, intention and thought were placed in a privileged role, potentially undermining the meaning of what people did and said. This intervention targeted the psychic privacy of the self, creating a new epistemological framework. While Bosavi Christians performed the requisite speech and other embodied practices, if assessed by the rarity of confession and the maintenance of psychic privacy through everyday verbal practices, a bounded self was nonetheless preserved during this first phase of Christianization despite the creation and use of a rich vocabulary of interiority.
TRANSFORMATIONS IN THE NOTION OF THE SELF AMONG THE WARI’ OF BRAZIL
Encountering Christianity
The Wari' are an indigenous people of southwestern Amazonia living in the Brazilian state of Rondônia. Speakers of a Txapakura language, their population today numbers around three thousand individuals. Their first relations with white people occurred at the start of the twentieth century and were based exclusively around warfare until so-called “pacification” in the late 1950s, when the Wari’ were contacted by the Indian Protection Service (SPI) and American fundamentalist Protestants from the New Tribes Mission (see Vilaça Reference Vilaça2010). These missionaries settled in the Wari’ territory and began to devote themselves to studying the native language and translating the Bible.
At the start of the 1970s, the Wari’, according to their own accounts, experienced a mass conversion to Christianity and remained Christian until the first years of the 1980s when they de-converted, again collectively. When Vilaça started her field research in 1986 few people called themselves Christian. The festivals and shamanism had resumed.
Another moment of conversion took place in September 2001, linked, the Wari’ say, to the attack on the World Trade Center, which they were able to see on the community television, and which was taken as a sign that the end of the world was looming. This marked the beginning of a phase of Christian revival that has led to about 70 percent of the adult population now declaring themselves “believers.” These worshippers attend services, mostly conducted by native pastors entirely in the Wari’ language, and involve themselves in village church activities.
This section provides a brief account of Wari’ conceptions of humanity and the self, and proposes for discussion a few hypotheses concerning how the Christian notion of the self—focused on the individual's inner being—has altered these conceptions.
The Body and the Heart
As among other native groups of Amazonia, the Wari’ have a complex definition of the body (kwerexi’ ‘our [inclusive] body’) covering not only the idea of flesh, or matter, but also the personality or way of being. The Wari' say that a person has a particular way of behaving or being because their body is like that. This body not only differentiates individuals through their particularities, it also differentiates the Wari’ as a whole from other Indians, whites, and other kinds of beings. Everything that exists has a body, which is the seat of its capacities and affects. Wari’ say, for example, that peccaries wander in bands because “the peccaries’ body is like that.”
As part of the body, the heart (ximixi’ ‘our (inclusive) heart’) is a central organ, responsible for the most vital physiological functions, in turn associated with cognitive and emotional capacities and dispositions. At a more general level, the term heart simply designates the inside or core of something, like the “hearts” of some fruits. At a narrower level, the heart refers to intellectual capacity and understanding and is associated with vitality and agency. Hence when people say that active living beings have a heart, they mean that these beings know how to act, what to do, and what to eat. As with the body, the heart is not restricted to those beings conceived to be human. Animals, for instance, know how to search for food, find a shelter, and so on.
When it comes to the Wari’ themselves, the notion of heart acquires a complexity absent from their discourse on other beings: as well as understanding, it also refers to thinking, emotional and moral attitudes, attributed here to specific physiological processes controlled by the heart (see Conklin Reference Conklin1989; Reference Conklin2001). The heart is responsible for producing blood and distributing it to the rest of the body. Vitality is associated with the quantity of blood, which increases in proportion to its speed of circulation. A slow, shrunken heart leads to physical and emotional weaknesses. At the same time, sadness, especially yearning for dead kin, can cause the heart to malfunction, leading to physical debility, illness, or even death. When worried or anxious, the Wari’ often say: “my heart groans,” or “our heart is not breathing well; it is panting.” By contrast, being well means that one's heart is “well-settled.”
Emotional state and moral behavior are likewise not separable from cognitive thought processes. For the Wari’, being well emotionally means thinking well and understanding things properly. An intelligent person who quickly understands what is explained to him or her, who learns a skill quickly, “has a heart,” just as a sensible person does things correctly, such as not becoming overly sad with a death, avoiding sex outside of marriage, and not stealing. Like the body, the heart has a collective aspect typical to the “species” and an individual aspect, singularizing a person through their appearance and their way of acting and feeling. In the Christian context, “having a heart” is one way of saying that the person behaves like a good Christian.
Returning to the more general Wari’ idea of the body, we can observe that it evokes a notion very similar to the “mindful body” or “embodied mind” used to reformulate descriptions of the Euro-American notion of the body previously based on the Cartesian paradigm and its strict differentiation of mind and body, a separation alien to people's everyday experience (Lock and Scheper-Hughes Reference Lock and Scheper-Hughes1987; Vilaça Reference Vilaça2005; Reference Vilaça, Vilaça and Wright2009). However, the Wari’ notion of the body involves an additional level of complexity since it is based on a distinct conception of humanity predicated on its instability.
In the non-Christian Wari’ world, although everything had a body and a heart, only those entities possessing a double, what the Wari’ call jamixi’ (‘our [inclusive] double’), and translate to Portuguese as “spirit,” were considered human. These included not only the Wari’, other indigenous groups, and white people, but also diverse species of animals. All humans see themselves as people, wari’, and share the same way of life or the same culture: they live in houses with their families, prepare their food, and hold festivals fuelled by fermented drinks. What differentiates them is precisely their perspective, a conception consistent with the way of being described above, associated with their bodies. While both the Wari’ and jaguars live in houses, what the Wari’ see as houses are constructions made from timber and straw, while jaguars see them as mountain caves. For the Wari’ fermented drink is made from maize, while for the jaguar it is blood.
As Viveiros de Castro argues (Reference Viveiros de Castro1996; Reference Viveiros de Castro1998), this is a conception of alterity very different from the otherness implied in our own cultural relativism, since in the Amerindian cases he has analyzed we are dealing with the same “culture” and multiple “natures” or bodies. This “perspectivism” implies a very specific way of conceiving relations and knowledge. According to Viveiros de Castro (Reference Viveiros de Castro2004: 6–7), those Amerindians relate between themselves and with others through the notion of equivocation: in other words, they set out from the principle that the interlocutor may have a different perspective, meaning the person must remain aware (or wary) of this potential difference the whole time. Here, a better explanation of the Wari’ concept of double (jamixi’) is needed for us to comprehend more clearly the idea of equivocation, which, though a positive mode of knowledge, contains the possibility of making mistakes that can prove fatal.
Jamixi’ is not an immaterial component of the person located in some part of the body; it is a capacity to transform, the ability to assume the form of other bodies, which is characteristic of every human being. Hence when a person becomes sick after being shot with an arrow by a capuchin monkey, the Wari’ say that it is the monkey's jamikon (jam- + -kon male suffix) that caused the illness by acting as a person instead of an animal. Simultaneously, the Wari’, through the shaman's vision, know that the victim, his or her jam-, is transforming into a monkey due to the fact the double is among the monkeys, living among them as though they were kin, eating their food. As a result the sick person acquires a monkey subjectivity/understanding, which the non-Christian Wari’ translated as the loss of a Wari’ heart. A man once told Vilaça that during the disease process, “the animal enters us and starts to eat our heart.” If not rescued by the shaman, the victim would become completely monkey, ceasing to have a body visible to the Wari’ as human (humans would see him or her as a monkey). Hence, while jamixi’ is a capacity for transformation, it can only be objectified as a body, a different kind of body because of its links to other relations. The status of a person, whether human or animal, did not traditionally depend on self-perception, since everyone saw themselves as human, but on who sees the person as a person: in this example, the Wari’ see the victim as a monkey, while the monkeys see him or her as a person.
This kind of uncontrollable and undesired transformation was not a rare event dissociated from everyday life. Any illness was conceived as a process of transformation, with children being the most susceptible to complete transformation. Preventing these sudden transformations depended on constant care, attention, affection, and above all providing food. These forms of caring, combined with obeying various rules on the treatment of animals, helped deter counter-predation, and ensured that people constituted themselves as human. Those who had nobody to look after them were always potentially vulnerable to turning into an animal, or of going to live among the dead, since these other humans were always interested in attracting Wari’ and transforming them into their kin. To become vulnerable, it was enough to respond to their call, talk to them and, above all, eat their food or have sex: such actions could lead to the victim's perspective being completely subsumed by the other. By losing their Wari’ body/subjectivity/understanding, the person began to see the animals as humans and to be seen by the other Wari’ as an animal.
We can conclude, therefore, that the vitality attributed to the proper functioning of the heart was traditionally conceived as a sign or a guarantee of non-transformation, registering the person's stability. People without an appetite, lazy, and above all sad could be suffering from a process of transformation that needed to be reversed. The risk of making a mistake about people's identity was a constant feature of Wari' life and demanded constant attention. Nobody knew for sure who was who, an ambiguity that also applied, of course, to the person him or herself. There was nothing like a “self” identity to be possessed since this identity was contextual, produced on the basis of a relation determined from the outside.Footnote 12
From this we can gather that the Wari’ conceive of the secret—something that is not said, whose origin or container is the heart (“it is fixed in her heart; he/she does not speak it”)—to be extremely dangerous. The threat posed by equivocation means that people should give clear manifestations of their humanity, their wari’ way of being, thereby showing that their ‘heart’ is open, exposed (see Taylor Reference Taylor, Latour and Weibel2002: 462). Intense relations between people are essential because they fill the place otherwise occupied by relations with other kinds of people. In this sense, morality for the Wari’ was subordinate to relations with the outside. People became like one another, constituting a moral community, to avoid turning into others by being seduced or captured by these other subjectivities.
In the context of this paper, the primary question that arises at this point is that of whether or not adherence to Christianity is provoking a change in Wari’ conceptions of selves and relations? Is there a new way of conceiving the meaning of human action?
In the Christian world the act of creation, as narrated in Genesis, established a primordial and fixed difference between humans and animals: humans are urged to prey on animals. Given the equations made by the Wari’ between humans/predators and non-humans/prey, the act of creation means that animals are deprived of their human attributes: in Wari’ terms, they lose their double/spirit. It makes sense, therefore, that the Wari’ are interested in what we could call the ontological aspect of Christianity, since the latter provides them with an additional tool for the continual work of differentiating themselves from animals.
Another aspect of Christianity that interests them in particular is its moral code, which professes generosity and love, and which links persons through siblinghood: everyone is a child of God and a brother or sister in Christ. This Christian morality also matches the Wari’ ideal of community life, which excludes affinity, and the morally condemnable behavior associated with affines, angry people, and sorcerers. Affines, as stated in myths, rituals, and social practice, are kinds of enemies within the group, which are equated with the animal (karawa) position, constituting a constant threat to the stabilization of communal life. While during pre-Christian times the Wari’ looked to mask affinity using consanguineal terms of address or teknonyms for co-habiting affines, the terminology of Christian brotherhood today offers them an extra resource or vocabulary. Hence, it is as if Christianity, with its ontology and morality, had arrived as a response to the difficulties experienced by the Wari’ in their day-to-day lives, offering viable solutions to problems of selfhood that already beset them. Could we say, then, that Christianity has not induced any real difference to their lived world?
One change worth considering relates to the Wari’ conception of their given or innate world and, consequently, of the “control” or focus of their inventive process, to use Wagner's terminology. According to Wagner (Reference Wagner1975: 42–51), any human culture defines a domain of the innate, a primary or implicit context. For members of the urban and secular cultures of the Euro-American middle and upper classes, the domain of the innate is nature, individualities, the self, and personalities, while the artificial domain, to which action is focused, is formed by convention, rules, and society. For members of tribal, rural, and religious societies, among others, the innate is precisely the opposite—convention, rules, and relations—and the purpose of human action is to produce singularities and differences on the basis of this conventional context.
If what was once innate or given among the Wari’ was a world inhabited by an extended humanity that encompassed humans and animals, today this given world—given by God—centers on the difference between humans and animals. Hence although the Wari’ have striven to produce a moral community in both the traditional and Christian worlds, in the former context it was produced against the innate background of a generalized humanity, while in the Christian world, where this background ceases to exist, the production of a moral community, though similar to the former in terms of its overall characteristics, no longer comprises a movement of differentiation and particularization of a Wari’ community on the basis of a common human background. Today it seems to occur independently of this context, and the Christian Wari’ frequently describe the moralizing actions in a way reminiscent of our own, that is, as the production of a moral collective on the basis of individual idiosyncrasies. Christianity has led them to reconfigure the given world, in a way that became similar to our own. Moralizing action therefore gains another reference point. Or rather, by ceasing to be differentiating, it becomes conventionalizing, counter-inventing the innate as particularities or individualities, along the lines of the Euro-American model (Wagner Reference Wagner1975: 45–46).
This helps explain why the concept of heart, related to personal singularity, has undergone a kind of hypertrophy. Though found exclusively in the Christian context and concentrated in the speech of the younger generations born after contact, expressions have emerged that are seldom used in other contexts, such as “let's look at our own hearts,” “he sees his own heart,” “my heart detests you,” or “you see me through my heart,” indicating a movement towards a singularization of the person through the constitution of an inner self. While in other discursive contexts the more generic term for the body is used to qualify and differentiate a particular person (“her body is like that, that's why she's always angry”), in the Christian universe these qualities are concentrated in the heart, rendering them more internal and less visible.
So far, though, this movement seems limited. The Wari’ experience conversion (and de-conversion) as a collective process, while sinning—judging by the content of the public confessions—is understood as a failure in relationships (for example, I cheated on my husband because he traveled a lot; I did this because I was attracted by the devil) rather than as a personal failure generating guilt. This suggests a non-individualized idea of morality. Moreover, the inexorable fact of death, which arises from the fact of occupying the position of prey, forces them back into their ontological dilemma where humanity is only a transitory position. Christianity also provides them with the figure of the devil who is embodied in animals and thereby returns human agency to them. With the passing of time, however, relations with the devil began to concentrate in the heart: in other words, they have a moral effect on the person, who acts in a conventionally condemnable way, and not an effect of metamorphosis, though this too may happen. In this sense, nowadays the innate world reconstituted by the devil bears closer proximity to a morally diverse human nature rather than the extended humanity found before. From bodily metamorphosis the Wari’ have shifted to moral transformations.
CHRISTIANITY AND CHANGING NOTIONS OF SELF AND BODY AMONG THE URAPMIN OF PAPUA NEW GUINEA
The Urapmin are a group of 390 people living in the West Sepik Province of Papua New Guinea. Due to their remote location, they were never directly missionized by the Australian Baptist missionaries who settled in other parts of their region in the late 1950s. Observing the effect missionization was having on their neighbors, the Urapmin sent some of their young people to live in Telefomin, a community about six hours walk to their East, to attend a mission school. Over the course of the 1960s and early 1970s, these young people brought a sophisticated understanding of Christianity back to Urapmin and taught most members of the community the main tenets of the faith. Yet even as these young people, who were all converts, taught others the outlines of Christian doctrine, they did not lead many of their elders to convert. Then, in 1977, a charismatic revival movement swept through the highlands of Papua New Guinea. When some of the young Urapmin converts brought the revival to their home, people began to be possessed by the Holy Spirit—to shake, feel hot, and become convicted of their sin. Those who had such experiences, or watched their relatives have them, became convinced of the existence and power of the Christian God and converted. Within a year of the advent of the revival in Urapmin, the entire community had become Christian.
Robbins has elsewhere discussed in detail the conversion of the Urapmin and the nature of their Christianity (Robbins Reference Robbins2004a). Here we recall only the main outlines of this previous analysis. Among the Urapmin, the religious domain has been profoundly transformed along Christian lines, and in their daily lives Urapmin people are preoccupied with issues of human sinfulness and the possibility of overcoming it in order to attain salvation. Crucial to Robbins’ analysis of the Urapmin focus on sin is the claim that while Christianity has radically transformed aspects of the Urapmin understanding of self, it has not succeeded in transforming indigenous models of sociality and human relatedness, both of which draw on older notions of the self. The Urapmin sense of sin, Robbins argues, is Christian in character but is given its felt force by the difficulties the Urapmin have in reconciling their Christian religiosity with their more traditional ways of approaching the social domain. The pivot between Christianity and the Urapmin understanding of sociality is the conception of self. This conception is currently composed of both Christian and traditional threads and is thus a key site in which the relationship between these two broad cultural logics is being worked out.
We begin by discussing the set of ideas we call, for purposes of exposition, the traditional Urapmin notion of the self, a notion that is still in play in the way Urapmin conduct much of their daily lives. Urapmin speak of a set of personal inclinations and faculties (emotions, thoughts, and the will or intention) that they see as crucially involved in producing a person's behavior. The Urapmin locate these inclinations and faculties in the ‘heart’ (aget). Crucial for present purposes is the internal quality of this location. The heart is “inside of the body” (ibak tem, ‘body interior’). It is far from the ‘skin’ (ipnal), which is figured as the surface or outside of the self. Thinking and remembering are things the heart does inside the body (aget fukenin and aget fenung, respectively), while feelings are doubly internal, being located inside the heart (aget tem, ‘heart interior,’ but also the generic term for ‘emotion’), which is in turn inside the body. In Urapmin, the will (san) is not as clearly located by the verbal formulas used to refer to it, but everyday conversation makes it clear that it is also in the heart.
The Urapmin heart and the faculties it contains are not only markedly internal, they are also “private.” This private quality of the heart and its states is evidenced by people's ubiquitous assertions that one cannot know what is in another person's heart. Located inside the body, people's hearts are “opaque” to one another (Robbins and Rumsey Reference Robbins and Rumsey2008). Furthermore, even the thought of trying to guess what is in other people's hearts is treated as repellent. Urapmin would react with something like disgust when Robbins asked them what others were thinking or why they had acted as they had. Their reaction is based on a strong notion of “psychic privacy,” which holds that even if it might sometimes be possible to guess with some confidence what others think or feel in their hearts, it is improper to attempt to do so. Taken together, the notions that people's hearts are opaque to or hidden from one another, and that it is improper to try to see past this opacity, go beyond the general Western notion that the self has privileged access to its own mind: the idea that I know my thoughts, feelings, and intentions in a way others do not. In Urapmin, this doctrine of privileged access often becomes something more like a doctrine of “exclusive access”—only the self legitimately knows its inner states, others are normatively completely ignorant of them.
This doctrine of exclusive access sets up what we can call the Urapmin problematic of selfhood—a problematic that has played an important role in the Urapmin encounter with Christianity. The Urapmin problematic of selfhood turns on the question of how selves are to be connected, given the barriers that separate hearts from one another. The Urapmin do not address this question in the terms that Westerners use to confront a much milder version of this problem presented to them by their own notions of the interiority of the self. The paradigmatic way Westerners solve the problem of relating inner selves to one another is via language. Western language ideologies, based in important respects on Protestant ideas, dwell both on the importance of sincerity and on the need to interpret speech and written language on the basis of the intentions of those who produce it (Robbins Reference Robbins2001a; Keane Reference Keane2002; Reference Keane2007). In this language ideology, people reveal themselves to others by means of language, and by interpreting correctly what people say we come to know about their inner selves. Among the Urapmin, key language ideological tenets do not support this way of understanding contact between selves, holding that speech can never reveal what its producers think, feel, or intend, and that its interpretation hence cannot be based on knowing such things. For the Urapmin, the interpretation of speech involves the listener deciding how they want to understand the words spoken. It involves, that is to say, the listener's relationship to his/her own heart, not that to the person who produced the speech (Robbins Reference Robbins2001a). In light of this kind of language ideological emphasis, it is fair to say that language does little to connect selves/hearts to one another in Urapmin.
How, then, are people connected to one another in Urapmin, given that they are not connected by their hearts? They are connected in two ways that tend to operate simultaneously. Both means of connection relate people primarily as bodies, rather than as hearts or minds, or at least they suggest that it is only the body that can ever evidence for others that an inner self exists (cf. Strathern Reference Strathern1979). The first way selves relate, which is clearly lodged in the body, is on the basis of the fact that as bodies they are born related to one another through the shared bodily substances that constitute local definitions of kinship. As a general claim about Melanesian sociality, this point is so well known by now as to need little discussion (see Wagner Reference Wagner1975; Strathern Reference Strathern1988). The cognatic Urapmin exemplify it in a very simple, rather poorly elaborated form. They understand a person's body to be made of a woman's menstrual blood and a man's semen. By virtue of being made of these bodily substances, a person is born related to numerous others on the sides of both parents and will grow up to take part regularly in exchanges that realize and reinforce these relationships. Selves as bodies have such relationships regardless of the states of their hearts. Though ideally the heart will be preoccupied with “thinking about” these relations and with willing its own participation in the exchanges that they entail, the relations will exist regardless of how the heart behaves. On the level of the body, then, relations are taken for granted in a way that is not challenged by the opacity and privacy of the heart.
The second way Urapmin solve the problem of bringing selves into relationship is through the use of material objects. Exchange of such objects, which is ever-present in Melanesian societies, serves to realize and enhance the relationships between people already given by their links of bodily substance. When exchange is of foodstuffs, it contributes directly to bodily relatedness by fostering bonds of shared substance. But in all cases, whether people exchange foodstuffs or other objects, exchange serves to socially display and fix certain kinds of intentions that ground relationships between people. Perhaps it is better to say exchanges “constitute” such intentions, because the fact of exchange makes it irrelevant whether such intentions actually exist in the participants’ hearts—simply having carried out the exchange commits one to the relational intention it entails, whether one experiences that intention internally or not (see Robbins Reference Robbins2001b; Rappaport Reference Rappaport1999). In this way, exchange can relate people as if it were their intending hearts that are related, even if the precise content of people's hearts remains opaque.
The heart and the body clearly accomplish different things in Urapmin understanding. It is also true that as the two key parts of the Urapmin self, their relationship can often be one of conflict. Put simply, the bodily, relational aspect of the self demands that people act in accordance with what Urapmin call “the law” (awem). The law requires people to honor the obligations of existing relationships by treating everyone in reciprocal ways, particularly in material exchanges. By contrast, the will, a key component of the private, relatively non-relational part of the self that is lodged in the heart, often drives people to disregard existing relations in favor of pursuing new ones or otherwise seeking self-aggrandizement. By virtue of its cognatic character, Urapmin society demands that people both attend to the law in maintaining existing relations and deploy the will in making new ones (Robbins Reference Robbins2004b). People's moral success depends upon their skill in balancing lawful and willful behavior in ways that allow both the hearts and bodies to work together in the self's construction of social life.
The Urapmin have understood the process of conversion to Christianity to involve in important respects a transformation in their notions of selfhood. As an initial formulation of this transformation, it might be said that in indigenous Urapmin thinking, the body connected selves and the heart separated them. In their understanding of Christianity, by contrast, it is hearts that should connect selves and bodies that should separate them. This chiasmic framing is in some respects too schematic; for example, in traditional Urapmin thinking, the heart should at times work to maintain the connections the body establishes by foregoing willfulness in favor of lawful intentions. But it can still serve usefully to orient the account of the transformation of Urapmin ideas of selfhood that follows.
The Urapmin had little difficulty aligning the descriptive aspect of the Christian concept of the internal self with their traditional notion of heart. Thus they have not felt a need to adopt special Christian terms to talk about the heart or its contents of thoughts, feelings, and intentions. Standard Urapmin terms work fine for this purpose since just like the indigenous Urapmin heart, the Christian one, as the Urapmin understand it, is internal, private, and the seat of thought, emotion, and will. But in contrast to ease with which Urapmin terms can express the descriptive features of the concept of heart, conversion has fostered a profound transformation in the moral content of that concept. Urapmin Christianity defines the heart as the primary source of connection between people. Good feelings and thoughts in one's heart should drive one to relate to others, and sincere communication of these feelings and thoughts in speech should provide the content of these relationships. It is by sharing thoughts and feelings with others, by being “in agreement” (wanbel) with them that one comes into relationship with them. This notion of relationship demands that people read the hearts of others by listening to their speech.
Having defined the heart as central to relationships, Urapmin Christianity directs a good deal of moral energy at insisting that the heart contain only “good” thoughts and feelings. Anger, covetousness, and desire are considered sinful. And sins of the heart—sins of bad feeling (aget tem mafak) and bad thought (aget fukunin mafak)—are the ones Urapmin worry most about committing. Controlling the heart has thus become one of the key moral practices for Christian Urapmin and a constant, quotidian preoccupation.
This emphasis on the relational work of the heart has not only changed the way Urapmin relate to their thoughts and feelings, it has also fundamentally transformed the dialectical relationship of law and will that is at the center of indigenous Urapmin morality. In their indigenous understanding, Urapmin selves have to balance willful intentions and actions against lawful ones—finding a way to negotiate between them so that old relationships can continue to develop even as new ones come to be born. As Christians, however, Urapmin must “suppress” (daunim) their wills, or surrender them to the will of God. In Christian terms, the will is what drives bad thoughts and feelings toward active expression, and in so doing definitively destroys relationships and brings whole groups of people into sin. For Christians, there is no room at all for the will in social life, only for lawful intentions and actions that offend no one and ensure that “agreement” between hearts will give social relations a placid or “slow” tenor. Self control then not only works to prevent bad thoughts and feelings, but also to extinguish all aggressive or self-regarding aspects of the will.
A final point needs to be made about Christian conceptions of the heart in Urapmin. Even as Christianity defines the heart as central to the way a self relates to other selves—even as, that is to say, it addresses the Urapmin problematic of the self by allowing hearts to connect directly to one another—it also downgrades the importance of relationships in defining the moral status of the self. Indigenous Urapmin morality evaluated the self on the basis of its ability to both create and maintain relationships. Urapmin Christian morality does not place the creation and maintenance of relationships in this paramount position. Instead, it insists that what matters for the moral state of a self is its ability to suppress its will and have a heart that is calm or “slow.” Even anger provoked by others counts as a sin for the self that experiences it. People will only be saved by their own “belief” and the moral self-control that follows from it; having many enduring relationships does not in itself contribute to salvation. For this reason, we can say that the Christian self in Urapmin is an individualistic rather than a relational one. It acknowledges the need for the heart to connect with others through sincere speech, but it defines social relations as a distinctly secondary matter in comparison with the need to control the heart.
What, then, becomes of the body in Urapmin Christianity? Traditionally the source of relatedness, and along with the heart one of the two key components of the Urapmin self, it is somewhat of a minor player in Urapmin Christianity. The ways in which it connects people—through shared bodily substance and material exchange—remain in force, but they are not matters of ultimate importance. More than this, they can be sources of conflict over the distribution of material resources that can damage people's efforts to maintain hearts that are free of sins of anger, covetousness, and desire. For this reason, some people now try to remove themselves as much as possible from the bodily aspects of social connectedness, withdrawing in particular from the major ceremonial exchanges that represent the most public displays of bodily relatedness in Urapmin. Participating in exchanges such as brideprice payments, they say, can “destroy” their “Christian lives.” In such cases, our schematic claim that for Christians the body should separate people becomes an explicit guide for life.
To summarize our findings, the indigenous Urapmin self has two components. One, the heart, is interior, hidden from others, and is the place in which mental and emotional life transpires. The other, the body, is external, visible to others, and is the primary force in the construction of the self's social relations. The process of conversion has transformed this notion of self by basing social relationships on the possession of shared thoughts and feelings. It has correlatively diminished the importance of the body in the constitution of the self, and of social life as well. Having moved the heart to the center of both selfhood and social life in this way, Urapmin Christianity has made the control of the heart the key goal of Christian practice. It should be said in conclusion, however, that the transformation from the indigenous to the Christian notion of self is not yet complete in Urapmin. It is the contradictions that hold between the two of them that provide the sense of routine failure that drives the felt sense of sinfulness that so profoundly shapes the lives of most Urapmin.
CONCLUSION
The purpose of this article has been to explore new avenues for cross-cultural comparison opened up by the development of the anthropology of Christianity. Our work has also unfolded under the influence of recent thinking about ways to revive the project of comparative anthropology. Borrowing from Fox and Gingrich and from Keane, we have highlighted the ways in which people living in three unrelated societies hailing from two different world regions have grappled with the similar affordances offered by several strains of Evangelical Christianity. On the basis of our prior knowledge that this kind of Christianity tends to make strong claims about the nature of the self, we chose changing ideas in this domain as the focal point of our comparison, searching, as Detienne suggests, for ways in which Bosavi, Wari’, and Urapmin converts, once they took on Christian identities, have worked with a logic of evangelical selfhood. Taking another lead from Detienne (Reference Detienne and Lloyd2008: 26), we have defined the domain of selfhood both inductively and comparatively, allowing more standard definitions to “fracture and disintegrate” as our project developed. On the basis of this method, we have ended up treating selfhood as a set of linked ideas about the interior of the person, the body, and the nature of exchange and communication. It is around these subjects that we have compared processes of the cultural change that have followed from conversion in our three cases.
In conclusion, we would like to briefly review our comparative findings. But it is perhaps appropriate to begin by noting something we somewhat surprisingly did not discover. In the end, differences between the disparate branches of Evangelical Christianity and the distinct missionization processes involved in our three cases failed to register strongly on our comparative work. The Bosavi and the Wari’ were both missionized by Fundamentalist Evangelicals, while the Urapmin were influenced by more mainstream Baptists and then converted to a Charismatic form of Evangelicalism (see Robbins Reference Robbins2004b for some of these differences). In terms of missionization, the Bosavi were partially missionized by Western Fundamentalists. The Wari’ experienced this kind of missionization as well, but then de-converted, only to convert again almost twenty years later, still directly influenced by the missionaries, who stayed throughout this whole period. The Urapmin, for their part, were never directly missionized by Westerners at all, but sought out contact with Baptist missionaries active in their area and then converted in a charismatic revival movement that was widely influential throughout Papua New Guinea. When we began our comparative project, we assumed these kinds of differences would loom large in our account. As our analysis unfolded, however, we found that as regards changes in notions of selfhood, commonalities greatly outweighed differences, and therefore these variations in denominational influences and missionization experiences faded into the background of our comparative work. Our conclusion in this regard is that the elements of selfhood we take up here are common throughout different kinds of Evangelicalism, and are a widely diffused aspect of the “logic” (in Detienne's sense) of this kind of Christianity—an aspect that is available to be taken up and elaborated in the locales to which different kinds of Evangelicalism travel in various kinds of ways.
What, then, are these elements of the Evangelical notion of selfhood that we found in play in all our cases, and what have we learned about how they are taken up in the course of conversion? At the broadest level, our primary finding is that conversion in all our cases radically transforms notions of the inner self, the body, and relations between people. The most striking commonality is the strong emphasis converts come to put on the inner self. In all of our cases, in Christian terms this self is rendered as the heart. For the Bosavi, the development of the notion of the heart (bel) as the center of personal experience, and of a highly salient notion of the inner self more generally, was something Christians had to develop anew. For the Wari’, a notion of the heart as a meaningful site of thought and emotion was traditionally available, but it experienced a hypertrophic development under the influence of Christianity. In Urapmin, the notion of the heart as the key interior locus of thought and feeling was traditional, but as in the Wari’ case conversion caused it to become even more important to Urapmin understanding of the self than it had traditionally been. While these differences are of interest, what is most striking is the extent to which after conversion the heart moved to the very center of people's conceptions of selfhood.
Along with important transformations in people's understanding of the heart and its relation to selfhood, conversion in all three cases also led to changes in the understanding of the body. For the Wari’ and the Urapmin, the status of the body as a key focus of moral evaluation and relational construction was greatly diminished. Among the Wari’, this led to a focus on changes in the status of a person's heart becoming as morally momentous, if not more so, than bodily transformations between species-perspectives had once been. For the Urapmin, moral success came to be based on control over the state of the heart and the actions to which it led, rather than primarily on one's ability to meet the demands of bodily relatedness. Among the Bosavi, the body in one sense retained its pre-Christian moral importance, but the expectations laid on it changed radically. No more was the self's goal the cultivation of a hard, assertive body. Instead, the self sought to project a sick, soft body that withdrew in important respects from the rough and tumble of social life. In this case, the body remains focal, but only to the extent that it enacts its own diminished power. Given that the Christian body in Bosavi is committed to its own diminishment in this way, the changes in Bosavi notions of selfhood are not so different from our other cases after all. As with the heart, then, we find interesting differences in how conversion has transformed ideas about the body, but we also find a very robust common pattern of subordinating the moral status of the body to that of the heart.
A final dimension along which we have tried to examine the self is that of relatedness to others, understood both as connections made through verbal communication and those based in bodily linkages. Our guiding question has been how does the self connect with, or disconnect from, other selves? We have found key changes in notions of relatedness in all of our cases. First, relatedness has come to be based much more importantly, or at least ideally, on shared understandings of what is in the heart, rather than on shared bodily connection based either on exchange, shared kinship substance, or bodily transformation between perspectival positions. Second, in keeping with what has been said about changing notions of the heart and the body, there has been a shift in moral emphasis away from a moral focus on the status of relations between persons or conspecifics (in the Wari’ case), to one on the state of the heart and its relationship with God. This is the kind of moral shift anthropologists and other scholars often reckon as a move toward individualism. While it has not been our primary goal here to argue for the connection between Evangelicalism and individualism, in this respect our conclusions do support widespread efforts in the anthropology of Christianity to develop more nuanced accounts in this area (see Bialecki, Haynes, and Robbins Reference Bialecki, Haynes and Robbins2008; and Mosko Reference Mosko2010 and the responses published with it).
On the basis of the foregoing observations, we can conclude that at least in our three cases there does appear to be a “logic” of Evangelical selfhood that is making itself felt in otherwise diverse processes of conversion. This is a logic that ties a growing emphasis on the inner self to a devaluation of the bodily contribution to selfhood. This focus on the inner self is further linked to a decreased (though never wholly absent) moral interest in the state of social relations in favor of one placed on the inner self, particularly as it is known by and related to God. At a very general level, such findings are not perhaps newsworthy simply as observations about Evangelical Christianity itself; as we noted above, Mauss and many others have recognized the importance of such notions of the self in this religious tradition. Yet it is also true that the comparative account we have provided of the different ways the Evangelical self has developed in our three cases highlights the links between the inner self, the body, and the importance of various kinds of relations between selves in some original ways. And what is of greater importance for our project is our demonstration in detail that this Evangelical logic of selfhood has in profound and subtle ways come into play in the three cases of conversion we discuss, and this despite the different starting points from which the Bosavi, the Wari’, and the Urapmin came to Christianity, and despite the differences in the ways they encountered it. This makes a crucial point about conversion processes, and processes of cultural change more generally—a point about how Christian conversion can lead to substantial changes in people's conceptualizations even of core domains of cultural understanding such as that of selfhood, domains anthropologists sometimes imagine are rarely subject to such profound transformation. It is also our hope that on a more general level the paper as a whole demonstrates the value of carrying out comparative studies of diverse societies based on the appearance in each of them of some or other form of the Christian tradition. The anthropology of Christianity has made promises about the value of such comparison from the beginning, and we hope to have redeemed some small portion of them here.
