Print version ISSN 0100-8587
Relig. soc. vol.3 no.se Rio de Janeiro 2010
Conversion, with versions: exploring models of religious conversion
Conversão, com versões: a respeito de modelos de conversão religiosa
Doctoral candidate on the Post-Graduate Progam in Social Anthropology, Museu Nacional, Federal University of Rio de Janeiro. email@example.com
Translated by David Rodgers
Translation from Religião e Sociedade, Rio de Janeiro, v.29, n.1, pp. 200-221, 2009.
Preferindo entender que não cabe ao antropólogo desautorizar os discursos nativos, quaisquer que sejam suas proposições, este trabalho pretende considerar afirmativas díspares a respeito de processos de conversão religiosa. Diferentes fenômenos de conversão são mobilizados não para serem explicados pelos modelos apresentados, mas para evidenciar quais os pressupostos e a aplicabilidade destes últimos. Deste modo, e calcando-se em exemplos etnográficos, almeja-se reposicionar certas perguntas a respeito daquilo que seria propriamente uma conversão e de como ela aconteceria. Por fim, analisa-se de que modo os movimentos de conversão contemplados implicam um desafio à consideração das noções de aculturação ou mudança social.
Palavras-chave: conversão religiosa, modelos de conversão, aculturação, mudança social.
By maintaining that it is not for the anthropologist to disempower native discourses, whatever the propositions of the latter may be, this work aims to take into account diverging statements concerning the processes of religious conversion. Different instances of conversion are not mobilized to be explained by the models presented, but to reveal both the presuppositions and the applicability of these models. In this way, and drawing support from ethnographic examples, the work looks to resituate certain questions about what a conversion is and how it occurs. Lastly, it analyses how the movements of conversion contemplated involve a challenge to the notions of acculturation or social change.
Keywords: Religious conversion, models of conversion, acculturation, social change.
Make the heart of this people fat, and make their ears heavy, and shut
their eyes; lest they see with their eyes, and hear with their ears, and
understand with their heart, and convert, and be healed.
Conversion can, of course, mean other things.
Peter Wood (1993:319)1
When examining discourses and practices relating to religious conversion, it is not unusual to encounter controversies over its legitimacy, its modes of operation or even the very meaning of conversion. The coexistence of divergent information on the topic is the stimulus for the present text, which looks to problematize the diverse ways in which the idea of religious conversion is conceived. First it describes those analytic models that interpret conversion through the prism of cultural change. Next the article focuses on situations where the event of conversion itself proves controversial, attempting to include all of the agents involved in interpreting the phenomenon. Finally it looks to establish a framework for exploring the complex question of conversion by making explicit and reworking some of the premises encountered in the presented examples. Hence the overall aim of the text is to resituate various questions concerning what conversion is and how it occurs.2
Generally speaking early anthropological discourse interpreted the theme of religion by enclosing it within a cultural system, a strategy that led to culture itself acquiring characteristics previously taken to be typical to the concept of religion (Viveiros de Castro 2002:191). In other words: a culture came to be understood either as a set of beliefs in which individuals placed their faith, or as an aggregate of representations held in common. A similar productive contamination can be observed in approaches that insert discourses on the theme of religious conversion within the conceptual framework of cultural change. This is the case of the work of Joel Robbins (2004), drawing from the anthropology of Marshall Sahlins (1985), which puts forward various potential models of conversion. In his reworking of the latters schema, Robbins (2004:10-11) presents three different ways of thinking of the encounter between two cultures, determined by the modifications that each culture experiences or ceases to experience over the course of contact. Following Robbins, a process of cultural change can be seen as: 1) assimilation, when certain groups, responding to new circumstances, adapt the latter to the categories of the previous culture; 2) transformative reproduction, which reflect the attempt to relate older categories to the contemporary world, with a consequent transformation in the relations between traditional categories; 3) adoption, which admits the possibility of adopting a new culture entirely, relinquishing any conscious attempt to adapt it to traditional categories. Robbins subsequently emphasizes the first and third models in his account of the conversion to Christianity of the Urapmin of Papua New Guinea.
As an exercise, my aim here is to examine how different theories of conversion approximate each of the identified models of cultural change, as well as to explore the possibilities for reinterpreting them, drawing support from different ethnographic accounts of religious conversions, especially those of indigenous populations.3 It is not my explicit intention to explain diverse situations of conversion through the expounded models and theories, but to understand their range of applicability, their premises and the ways in which they can be continually rethought by being placed in contact with distinct ethnographic situations. The cases presented here are not intended to be exhaustive in any way: rather, they are chosen in response to the common theme of conversion.
Models of conversion
According to the assimilation model, the outcome of a process of social change primarily depends on the original substance to be converted, that is, on the native substrate absorbing the external influence. The theory propounded by Robin Horton (1975) provides a clear example of this model. Horton argues that human groups cannot be taken as a tabula rasa, absorbing outside cultural influences in an unreflected form (1975:221). According to the author, a specific prior configuration predefines the terms in which the absorption will take place. Horton writes:
Given the same Muslim or Christian stimulus, some people remain unmoved while others respond [...] Here, it stares one in the face that the crucial variables are not the external influences (Islam, Christianity) but the pre-existing thought-patterns and values, and the pre-existing socioeconomic matrix. (1975:221)
Horton builds his theory on African responses to the so-called world religions. For the author, it matters little what the outside influence modifying a specific substrate is, since the outcome of the interaction if any exists is already given: almost irrespective of the questions, the answers are already known, they are already in the air as he puts it (Horton 1975:234). A similar line of reasoning can be found in other studies. In Frédéric Laugrands work on the Inuit, the author claims that acceptance of the new element always depends on a structure that, far from being destabilized, incorporates the novelty in its own terms, eliminating aspects incompatible with the native framework (Laugrand 1999:105). An analogous idea can be encountered in a text by Paul Schultz and George Tinker (1996:62-63), which argues that the interpretation of Biblical stories made by North American Indians is primarily determined by the importance given by native conceptions to all narratives. Returning to Africa, Birgit Meyer (1999:76) shows that, despite the emphasis of Pietist missionaries on faith above all else, the Ewe of Ghana understood the procedures introduced by the latter via a traditional mode of thinking that prioritized ritual action aimed at serving and influencing the gods. In sum, according to this reading of the assimilation model, seniority prevails.4
As Robbins himself points out, Hortons theory has been heavily criticized (2004:85-86). According to Robbins, the explanations given for conversion are generally either meaning-based, or, as in the African case, utilitarian. Hortons approach sets out from the premise that changes occurring in the world make conversion a necessary process for living in the new environment. The suggestion, then, is that rather than conversion depending on individuals motivated by the calculations of economic rationalism, it involves calculating peoples who select the best alternative for adjusting their cosmology to a new sociocultural situation the latter being responsible for the need for the superficial conversion in which they engage.5
Additionally, though, as the above quotation from Hortons text makes clear, the exacerbation of one of the poles i.e. that of the traditional frameworks makes it necessary to combine very different religions in the other pole. If the determining factor is actually the native framework alone, it makes very little difference to this theory whether a people convert to Islam or Christianity insofar as both can be grouped under the label of world religion. Here, though, I would argue that claiming that the specific religion to which people convert matters little entails a loss of intelligibility, not a gain. In the case of Hortons theory, explaining seems to involve eliminating factors that fail to suit the model, rather than fully considering what those involved actually present. Both converts and missionaries are implicated in the latter distinct people dedicating their existence to one particular religion.
Moreover the idea of assimilation also prevents us from taking into account the potential reciprocal effects arising from the process of conversion: while autochthonous cultures are eventually converted albeit in a very particular way and in their own terms the exogenous culture is taken as a block that remains unmodifiable by contact. One of the things that emerges strongly from studies of conversion, though, is how a religion wishing to disseminate itself more widely needs to develop a variety of techniques specifically adapted to those it aims to convert (Birman 1996:90, 92; Viveiros de Castro 2002:192). This in turn poses questions specific to the canon of the missionary religion involved.
Anthropological studies employing the assimilation model thus run the risk of deauthorizing native enunciations about religion, whether by being highly selective in their use of evidence, or by contesting the natives directly. For example, when explaining situations in which converted peoples claim that God was a decisive agent in their adoption of the new religion, some authors are compelled to refute their informants, arguing that it would be impossible for an element that in their view only comes into existence after conversion to function as its cause.6 Since the assimilation model emphasizes the former pole, the assumption is that the natives cannot be persuaded to convert by categories that they only imperfectly understand, constrained as they are still by the traditional cosmology.
Inuit natives, for example, today claim that they already used to perform a Eucharist ritual traditionally, having converted to Christianity long before any contact with missionaries (Laugrand 1997:109, 113). As the author summarizes:
When the missionaries arrive and introduce Christianity, the latter is already there, it has already been received. Hence literally speaking the missionaries do not teach anything new to the Inuit, who claim, on the contrary, to already know the Creator, the first two ancestors, the figure of Satan and even certain precepts. (Laugrand 1999:104)
Furthermore, contemporary Christian Inuit insist that by practicing their traditional religion they were actually, albeit unknowingly, worshipping Satan (Laugrand 1999:103). The assimilation model would be content to categorize this discourse as simply a reinterpretation of traditional experience. But could we not take the Inuit testimonies as a new symbolic interpretation of an old fact? Imagining this experience as a kind of performative re-experiencing (Crapanzano 2000:123; Segal 2003:241), for example, enables another conception to emerge: the Inuit become actors who reflexively reelaborate themselves and their past.
Allied to the first position, which prevents the emphasis from falling on native creativity, is a certain historical identification between anthropologists and missionaries (Stocking 1983:74), one usually rejected by both (Van der Geest 1990:589). After all, both groups think that they know the true essence of the religion in question and how any given culture moves towards or away from this ideal. That the missionaries externalize these positions as part of their catechizing is obvious. By assuming a similar stance, though, researchers engender theoretical and practical consequences for their research, ceasing to presume an at least initial lack of knowledge concerning what interests the natives the condition of possibility of anthropology. Peter Gow (2006:211-212) specifies the problem:
The questions I address here, why Piro people should have converted to evangelical Christianity and subsequently forgotten about it, and why they asserted to me that they had always been Christians, are clearly not Piro questions. They are the sorts of questions that anthropologists ask, and they contain a hidden danger. Most anthropologists have historically come from societies in which Christianity has been the dominant religion, and the discipline of anthropology has been formulated within intellectual traditions strongly marked by Christian thought. Because of this, anthropologists are likely to find themselves asking questions that are far closer to the questions asked by Christian missionaries than to those asked by the people the former study and the latter seek to convert. Indeed, the sorts of questions that interest missionaries and anthropologists about Christianity are quite similar, or at least are much closer to each other than they are to what Piro people find interesting in Christianity.
The distance between indigenous peoples and anthropologists in relation to the ideals of conversion should be seen, then, more as a warning for us to give due consideration to the claims made by natives. Robbins himself, though he mostly argues in favour of the adoption model in his analysis of the Urapmin case, is compelled to divide the conversion process into two distinct stages. In the first phase, conversion takes place for traditional sociocosmological motives, while in the second the natives can be thought to convert for religious motives connected to the religion they have begun to embrace (Robbins 2004:87). However this is the authors own inference, since this is not what the first known Urapmin converts are said to have claimed. Today they insist that it was God who led them to convert from the outset (Robbins 2004:112-115), something the author takes not as a fact, but as a claim to be deconstructed. So although Robbins advocates the adoption model in his analysis of the Urapmin case, it only applies as a stage subsequent to an initial phase of contact understood via the assimilation model, explained as a weakening of the traditional Urapmin division of ritual work. Here I am not discarding Robbinss interpretation: I merely wish to point out that it may be insufficient to account for what the Urapmin themselves profess. I return to this point later.
While in the assimilation model the somewhat determinist emphasis concerning the outcomes of conversion falls entirely on the pole of indigenous culture, what really matters in the adoption model are the singular features of the outside culture. Studies framed by the adoption model primarily seek to observe the post-conversion continuities between the religion that a population effectively begins to practice and the original religion brought by the missionaries. The idea of adoption implies that something essential can be transmitted without necessarily being modified in the process. As Peter Wood writes (1993:321): If Christianity has any single meaning in this welter of cultural contexts, it lies in the promise of a truth that transcends them all.
Usually when agents from the missionary culture speak of successful religious conversion, they mean the result of a complete process of adoption, or, more precisely, in certain cases, a process of substitution. Sometimes the arriving religion may be posited as an absolute novelty, when the assumption is that the peoples reached by the missionaries previously lacked any kind of religion. This is the case, for example, of Jesuit catechism in colonial South America: Christianity was a blessing offered to the Indians, who were expected to adopt it in a non-traumatic form given that they themselves were unfamiliar with anything resembling a religion (Viveiros de Castro 2002:192 note 12). However improbable this construction may seem nowadays, it cannot be ignored since it defines what the missionaries of the period understood as conversion: a procedure that could not be thought of as violent in itself since it did not conflict with any traditional values of the same importance.
In almost diametrically opposite fashion, certain missionary groups, in general Catholic, currently preach respect for traditional ways of life, considered in themselves an expression of religious values with which they agree, albeit expressed in an inadequate form. This is the case, for example, of the Missionary Indigenist Council (cf. Vilaça 2002:68). Perhaps we can therefore differentiate one adoption model from another, which can be labelled the substitution model, each with its own characteristics. Robbins (2004:10-11) hints at this possibility, although the author is apparently unconvinced that full cultural substitution is possible at least, definitely not in the case of the Urapmin.
This substitution model frequently presumes a radical transformation of the person, more specifically of the persons subjectivity an event usually identifiable in the accounts of rebirth accompanying conversions to Protestant Christianity (cf. Burch 1994:84). In these cases, it is indeed difficult to conceive the process in terms of adoption, since the latter contains the notion of two religions living in parallel something that would fail to make sense within the canon of exclusivist religions. This does not mean, though, that adoption does not occur in certain cases, and these should not be ignored either. As Steve Charleston shows (1996:78), the fact that North American Choctaw Indians, for example, assert the existence of a native Old Testament in parallel to a Jewish Old Testament may be problematic to others, but this does not stop them from considering both to be part of their own religion and this, in turn, to be legitimately Christian. Yet for the substitution model which assumes Christianity possesses a single unique sense, irrespective of the cultural realities that it must necessarily transcend (West 1996:33) this would be entirely impossible.
Earlier I mentioned some of the potential repercussions of confusing the programs of anthropologists and missionaries. But from the viewpoint adopted here, subsuming all religion in advance under the all-encompassing metanarrative of science has equally serious effects. Arguments that reduce conversion to a play of interests whether invoking a banal utilitarianism or the trading of irrelevant concessions for precious goods remove from analysis any religion that the natives already profess, thereby preventing consideration of any process of conversion, whether adoption or substitution. Yet the exact opposite may well occur: [The] foreign culture was frequently seen in its entirety as a value to be appropriated and domesticated, a sign to be assumed and practiced as such (Viveiros de Castro 2002:223).
The adoption model in itself does not seem quite so problematic as the assimilation model, but it may carry with it conceptions that prevent it from being legitimately used, at least in its entirety. For example, although Robbins (2004) admits that the Urapmin adopted Christianity in toto, this adoption could not have been instantaneous since, in his view, at the outset there existed a fundamental lack of knowledge about the new religion, an analytic stance that prompts him to divide the event of conversion into two distinct stages, as we saw earlier.7
Consequently the adoption model, in which two distinct cultures coexist as totalities, may require a notion of synthesis for it to become intelligible, even though this reduction is not necessarily imagined to occur between two hermetically sealed and immutable cultures. Here, as Robbins himself notes (2004:332), there is a difference in relation to the idea of cultural or social integration: it entails an ideal of predictability and the simplification of controversies, even though these are intrinsic to the native discourse. For the author, the moral torment of the Urapmin of which they are undeniably victims, it should be stressed stems from the coexistence of two essentially disparate cultural values battling for predominance, namely traditional relationalism and Christian individualism. However, the Urapmin see no problem in possessing two distinct and contradictory logics, that is, in being Urapmin and Christian simultaneously (cf. Robbins 2004:175-177). Indeed, it is precisely through Christianity that they can overcome their worst, closely inter-related problems the fact that they are black and poor and thereby attain a life free of torment (cf. Robbins 2004:xxvi-xxvii, 171-172). Their existence is morally conflict-ridden not because of a clash between cultural values per se, but because they lack the means to meet the demands of the situation in which they now live. Although the diagnosis may be the same, the attribution of different motives for their torments indicates distinct forms of dealing with the question: after all, the option in the first case would be to ignore the precepts of Christianity, which for the Urapmin would mean no less than eternal damnation.
Finally, the transformative reproductive model looks to provide a compromise between the two extremes. In this model an original culture is altered in response to its own precepts and to the specificities of the culture impacting on it. People from different cultures converting to distinct religions arrive at singular combinations, depending on the initial configuration just as much as the religion they start to follow (Wood 1993:305, 320). As Robert Hefner points out (1993:4), conversion to an exclusivist religion does not always demand apostasy: a more pacific combination of elements is possible, though generally involving the censuring of the traditional religion. In fact the transformative model can also be found in Meyers study (1999:110-111), in the process of diabolizing the traditional Ewe religion. Despite the use of categories linked to tradition to indicate their continuing presence, the relation of the Ewe with these categories is undeniably altered by conversion. By using the Ewe vocabulary, for example, the missionaries changed the meaning of commonplace words in transmitting the Christian message, though it was not possible to abandon them entirely or at least some of their acceptations.
Whatever the case, the four models of conversion or cultural change proposed here retain something in common. Without exception, they presume the existence of an original substrate that is subsequently influenced by a foreign culture. The choice of which model to use ultimately seems to depend on how the native culture is imagined: on one hand, cultures that only accept combining with others in their own terms; on the other, cultures that wish to adopt an alterity, conserving it as well as possible and to some extent extinguishing themselves in the process; and between these two extremes, mutual concessions in apparently more tolerant processes. Aside from the substitution model proposed here, which in any event does not seem to encounter any ethnographic examples to substantiate it, native culture is not extinguished during the transition. The question becomes the degree of interpenetration between the two cultures, as well as the type of coexistence that becomes possible, with or without conflicts.
One way or another, though, none of the models considers that the process of conversion may have reciprocal effects: both for the religion of the converts, and for the religion of who converted them. There may be no problem in recognizing, albeit somewhat timidly, how Christianity in Ewe or Urapmin can tell us something about the Ewe or Urapmin. However we can extend the question by taking seriously what Ewe Christianity or Urapmin Christianity have to say about Christianity in general (cf. DAngelis 2004:212; also see Capiberibe 2004:81, 96). Consequently even a so-called world religion can cease to be seen as a monolithic and immutable block shaping people in identical form wherever it is taken, becoming affected in the most diverse ways possible by its new collectives of worshippers (cf. Calavia Sáez 1999:49-50 note 10).
How do natives themselves explain their religious conversion? As we have seen, very often the motives given for this transition are discounted by the missionaries. However the anthropologist willing to consider them without pre-judging their validity can learn much from this information. It is not rare to find, for example, that conversions occur to obtain material benefits that will become available following religious adherence (Calavia Sáez 1999:43, Capiberibe 2004:75, Hefner 1993:5, Meyer 1999:11, Wood 1993:312). Nonetheless, as indicated earlier, reducing conversion to the logic of a simplistic economic rationalism can mean losing sight of the importance of religion itself, a fundamental part of the process. In the case of the Achuar studied by Taylor (1981:657), for example, who converted to obtain access to beads of divine origin, would it be fair to say that it involved merely the actions of calculating individuals? Here it is worth observing that the missionaries themselves usually bring with them an elaborate lay support infrastructure, enabling access to schools, hospitals and industrial goods as a whole (Burch 1994:84, Capiberibe 2004:59, Hefner 1993:38 note 14, Meyer 1999:22, Sahlins 1985:38, Wood 1993:320). Claiming that this paraphernalia is not part of the religion in contexts where the indigenous people think precisely the opposite means acting in the same way as the missionaries, assuming a monopoly of knowledge over what religion actually is.
In some ways this process can be seen as a form of conversion not to a religion, but to the idea of community (Gow 2006:213; Pollock 1993:66; Wood 1993:308; Viveiros de Castro 2002:190). This view, however, seems to be found more among the conceptions that the missionaries themselves have of the motives supposedly leading the natives to convert (Vilaça 2002:69; Viveiros de Castro 2002:192). Asserting that people embrace not a religion but what it represents can lead the anthropologist to adhere to the missionary perception of what constitutes authentic conversion. As Sjaak Van der Geest warns:
One could say that in most cases the anthropologist deprives religion of its original meaning and redefines it as something which is relevant and interesting within anthropological discourse. Religion thus becomes ritual, social control, a survival strategy, an etiology, a philosophy. [...] In other words, it becomes something that makes sense to the anthropologist. (1990:591)
Another reason for conversion can be seen, at least in Amerindian cases, in the inconstancy of the savage soul, to use Eduardo Viveiros de Castros formulation. The author notes that the constant indigenous desire to become other allows the possibility of convergence with the missionary desire to make the other identical to the self, thus resulting in the conversion and equally the deconversion8 of the natives (Viveiros de Castro 2002:193). Without doubt missionaries and Amerindians would both agree that the latter wish to be Christian like them. But while in the case of the whites, the emphasis falls on the idea that the natives supposedly want to be Christian, for the Amerindians it is more a case of being like them.9 As Viveiros de Castro points out (2002:224): In their own inconstant style, of course; the becoming white and Christian of the Tupinambá failed to correspond in any way to what the missionaries wanted, as shown by the resort to the shock therapy of compelle intrare.
Using Amazonia as a paradigmatic case, it should be recognized that by adhering exclusively to the western view of conversion as an interiorized and psychologized phenomenon the anthropologist is unlikely to observe any kind of orientation among native populations towards a new religion (Vilaça 2002:58). Hence the very understanding of what is deemed to be the religion worshipped by another people is fundamental to comprehending what the process of conversion is imagined to involve. And as we have seen, it is very possible that missionary agents and indigenous peoples have completely different ideas about this process. As Donald Pollock points out (1993:192 note 1, 172), in the 16th century most European colonizers considered performance of the sacramental rituals of Catholicism to be enough for someone to become a convert.
The Christian message of individual salvation, for example, may be altered to embrace communal needs (Hefner 1993:5), sometimes involving the conversion of entire tribes. Frequently observed (Capiberibe 2004:87-88; Gow 2006:219, 13; Laugrand 1997:109; Remie & Oosten 2002:113; Sahlins 1985:37; Schultz & Tinker 1996:66; Shapiro 1981:143; Vilaça 2002:64), this event may occur as a result of either native demands or the strategies typically used by those promoting conversion, with different effects in terms of the kind of change that comes to be considered. As a rule, when the emphasis on conversion tends towards the western paradigm indicated above, the missionaries refuse to recognize its occurrence en masse, undertaking painstaking work to win over peoples souls one-by-one. Even so the natives may well assert that they converted as a group, an event that, if ignored by the anthropologist, may terminate the investigation prematurely.
Horton (1975:395) observes similar turbulence among West African peoples, a constant oscillation between world religions like Christianity or Islam and the traditional African religions. For the author, though, it would be incorrect to speak of conversion in this case hence his decision to place the term in quote marks all the time since, by sharing the same general cosmological framework,10 both pagans and Moslems basically believe in the same things (Horton 1975:219, 394). Here we can imagine that the Moslems, to say the least, would tend to disagree with the author.
Presuming that all conversion can be limited to a specific event, a singular temporally demarcated moment, would also be another way of ignoring situations in which the natives claim to have been converted. As shown by the diffuse and continuous nature of the experience narrated by one of Vincent Crapanzanos informants (2000:104), the wish to define a precise moment when conversion occurred amounts to trying to control an experience that, by definition, cannot be controlled. The instant in itself matters less than the result of the process, a rebirth into a new life.
Likewise when missionaries claim that particular groups have yet to be converted, or that the process happened incorrectly, this constitutes another fact for the anthropological investigation, not one of its axioms. Ignoring the highly ingenious interpretations that converted natives produce, for example, on the basis of Biblical texts (in a similar way to those made by more orthodox fundamentalist Christians) can impoverish the anthropological research. The Ewe, for instance, focus on the ambiguity of the figure of the devil (Meyer 1999:41) in the scriptures in order to thematize the lack of certainty of their own lives, continually threatened by the arrival of the Final Judgment. The Wari emphasize the rules of conduct and eradication of affinity (Vilaça 2002:65) proposed by Christianity, delighting in the idea that everyone is a sibling. The Muscogee recall that, as Christ said, even the stones can cry (Maxey 1996:45), something which echoes their cosmology profoundly by not denying the agency of objects taken by others as inanimate. So can these explanations be taken as authentically Ewe, Wari, Muscogee? Undoubtedly. But this should not prevent the emergence of another question: pursuing a kind of symmetrization, can we not see these readings as authentically Christian too? Put otherwise: although these interpretations may very well be prefigured in some form in the native culture (Viveiros de Castro 2002:194), reducing them completely to reflections of a prior essence would be to ignore the originality of the indigenous constructions themselves in their constant thematization of the other as an other (Viveiros de Castro 2002:223).
Leaving aside for now the cultural transformation models described above, we can ask how conversion should be understood ethnographically, that is, studying the ways in which it appears in native discourse. First of all, my preference here is to avoid limiting the idea of conversion from the outset to something like its usual Protestant interpretation, that is, the idea that a profound reorientation of subjectivity is required for the process to occur effectively (Hefner 1993:35 note 2). Susan Hardings proposal (1991:380) concerning what she calls the representational event may be a good starting point. The author argues that this kind of event should be conceived as a complex, polyvalent and open discursive process, taking place at multiple levels, in which those involved including the self-proclaimed observers create and contest representations of themselves, others and the event itself.
Consequently what is initially taken as the same fact, conversion, can be seen in highly distinct ways. The case of the conversion of South American native peoples illustrates this point: while the Europeans sought to conceptualize the indigenous peoples within a typically western cosmology, the Indians, for their part, wanted to incorporate this alterity fully (Viveiros de Castro 2002:206). In any case, faced by the possibility of conversion, the very form in which Christianity thought of itself had to be transformed in the endeavour to answer the question of whether or not the natives had souls. Amerindian cosmology, for its part, cannot cease to be seen as indigenously perspectivist, continually desiring to exchange points of view. So where exactly do assimilation, transformation, adoption and substitution begin and end? The answers are not self-evident.
In any event, it is not my contention here that this dissolution of boundaries should be used to erase the idiosyncrasies of processes of conversion. While for some collectives conversion was equated with a kind of conjugation, encounter, commitment or interpenetration, ultimately resulting in a con-fusion (in the multiple senses of the word),11 for others, the missionary impulses characteristic of the world religions were more interested in the dimension of spilling outwards, spreading, expanding, increasing their contingent so as to enlarge their own borders while they themselves remained unchanged their chalice was not to be mixed with other nectars. According to Meyer (1999:134), here the question is not one of focusing on one of these acceptations at the cost of the other, but of actively preventing the reduction of the process of conversion to any one of these currents, instead perceiving the phenomenon in the way it is presented: multifaceted, complex.
The endeavours of missionaries among indigenous peoples suggest that the notion of perspective is important to understanding conversion processes on both sides. Among the Achuar and Wari of Amazonia, to pick just two examples, a close connection is observed between the act of seeing and the way in which one sees and the body one possesses. Inhabiting a particular body means participating in a specific world, distinct from the many other forms of the world occupied by beings with different bodies. According to Amerindian cosmology, entering into different worlds only takes place by exchanging perspectives,12 which is made possible by exchanging bodies properly speaking (as in the case of shamans). How would it be possible, then, to understand the idea of an omniscient god, like the god of Christianity, in a perspectivist cosmology? It only makes sense when correlated with the conception of divine immateriality. The fact of not having a body comes to be seen as the obviation of a constraint that would limit the capacity to access different worlds (Taylor 2002:464). Further still: it clears the way for indigenous multinaturalism itself to transform, at least at one level, into a mononaturalism, conceiving the existence of a single world under the constant vigilance of God. A kind of flattening of perspective thereby results (Vilaça 2003). Christianity, then, presents a way of seeing the world foreign to Amerindian cosmology, but which nonetheless can encounter a meaning in its terms; an authentically ex-otic point of view, to borrow Ordep Serras formula (1995:179), a gaze meant to be ubiquitous.
The phenomenon that certain Catholic missionaries call incarnation can also be rethought in a similar way. Some missionaries believe that before winning over the native souls to their religion, they themselves have to embrace traditional customs; only then is a dialogue possible that is in principle ecumenical (Shapiro 1981:141). The missionary vocation, pursued in this case through the imitation of the life of Christ, suggests that the outsiders should almost literally adopt a body adequate to the transmission of the Christian message in the environment in which they find themselves: The missionary must take on the flesh, the experience, of the Indians with whom he lives; this way, it is felt, the message he brings will be the answer to their own questions. (Shapiro 1981:143). The idea that bodily metamorphosis is the Amerindian counterpart to the European theme of spiritual conversion (Viveiros de Castro 2004:476) seems to be recognized by the missionaries. If it is not unusual, in this case, to imagine a missionary proposing to adopt an authentically perspectivist way of thinking, it can be asked to what extent a white person can possess a body similar to that of an Indian, the reply to which is also not immediately self-evident. In any case, by attempting at incarnation, the missionary does not abandon his world entirely: his logic does not involve exchanging perspectives with another, but adding other points of view to his range of possibilities to be used as necessary: he engages in a change of point of view. Unsurprisingly this proposal does not always find a lasting echo among Amerindians since in their cosmologies perspectives are not addable, only commutable: to obtain another perspective one has to lose ones own, even if only momentarily.
On one hand, as Judith Shapiro suggests (1981:146), the search to become an Indian may be no more than the missionarys attempt to encounter problems familiar to himself within the indigenous culture.13 In the final instance, a missionary, qua recruiting agent, has to convert someone to something, even if that means converting himself (Shapiro 1987:136). On the other hand, as Pollock suggests through the notion of religious diversion (1993:176),14 neither can we ignore the processes of indigenization undertaken, for example, by caboclos to indigenous cosmology, or by members of urban populations to African ethnic groups via religions like candomblé (see too Serra 1995:104). Finally we must avoid the tendency to hierarchize the significance of this kind of conversion, seeing it as either more true or less true because from the outset it involved a native movement without the presence of missionaries.
In turning to Amerindian cosmology, I have no intention of claiming that conversion processes show a perspectivist quality in each and every cultural change. What I propose is that the chosen ethnographic examples and the perspectivist theory inferred from them can function as emblematic cases by making intelligible what are in principle highly diverse situations. Here Crapanzanos reference to the experience of North American fundamentalist Protestants (2000:97) appears particularly resonant: [I]t is not so much a change in the way the world is experienced subjectively, but in the world itself, as it comes to be known, as it presents itself objectively. This makes evident how the presumed existence of a single, natural, unquestionable world is linked as Bruno Latour indicates (2005:116-117) to the notion that facts are incontestable and independent of ones relation to them. Alternatively, Latour suggests, recognizing that facts should be approached via their processes of construction enables us to conceive a plurality of incommensurable worlds (as in the case of multinaturalism). Consequently the existence of multiple truths, not necessarily mutually exclusive, also becomes conceivable. The diverse and apparently contradictory claims concerning situations of conversion discussed over the course of this article can acquire a new meaning if we analyze them in terms of exchanging perspectives. Or put otherwise: the Urapmin claim that their conversion was based on authentically Christian motives from the outset is unproblematic as long as the anthropologist treats this native claim as more than a retrospective perception, taking it, rather, as an enunciation made from another perspective, one grounded in another ontology in the same way we should read the motto of another of Latours books (1991) in which the author claims that we have never been modern. In other words, what the Urapmin say, in some ways, is that from a particular historical moment onwards they began to have always been Christian, which precisely matches the idea of conversion being a re-experiencing, a rebirth. At the same time, it is impossible to ignore the typical missionary insistence on converts showing exclusive fidelity to the monotheistic religious precepts. While the generalized constructivism to which we alluded can be especially useful when discussing gods in religions like candomblé as Latour himself shows (1984) whose practitioners insist on the made condition of their divinities, the same cannot necessarily be said of Christianity. For instance, it would probably be fairly unusual for a Christian to agree with a constructivist lingua franca in which his god is as constructed as any other. On the contrary, he would retort that God (capitalized) is a single, perfect being independent of any relation that humans have with him. Here it would be a case of re-reading, under a different light, the questions posed by Latour elsewhere: Might not the nearly fanatical attachment to the non-constructed character of the unity of God be largely a response to the unifying role of nature, which the negotiations have agreed to limit? If the latter becomes negotiable, why not the former too? (2002:45). While the authors question is pertinent in terms of how modern western scientists conceptualize nature, for Christians themselves it would make more sense to say that nature is non-negotiable insofar as it is a divine creation, and not the other way round. Ultimately this may lead to a questioning of the usefulness of unambiguous notions of cause and effect, at least in relation to the topic in question.15
At any rate, as Latour (1984) suggests, the problem can be circumvented by declining to choose between two interpretations, since strictly speaking there are not two but a multiplicity of distinct approaches. A fundamental question also comes to the fore here: the existence of power relations implied in the activity of restricting the diverse cosmological elaborations continually reconfigured by the natives. Extrapolating from one of Latours formulations, perhaps we can speak of an ontology with an invariable geometry that is frequently accompanied by violent processes of ontological constraint during the conversion process. Adopting a radical approach, it would make sense to doubt even the character of conversion to be read in this kind of situation, since, after the obliteration of perspectives, only a single ontology would remain that would have subsumed all the others, not any kind of exchange. It would be possible to conclude, based on the observation made by James West (1996:35), that while the contemporary Christian mission is fully aware that a pluralist world exists, it cannot in any way accept the existence of plurality of worlds. Ontological multiplicity, for its part, would be related not to the proposition of a unique, extensive truth but to existential, intensive truths, according to each situation and experience, in the words of Godfrey Lienhardt (1961:250).
In sum the constant movements of conversion imply a challenge to the notions of acculturation or social change (Viveiros de Castro 2002:191). Anthropologists can and sometimes really should be seen as participants in the process, whether, for example, as the target of native actions (Crapanzano 2000:164-165), or as a vector for transformations (Van der Geest 1990:588-589; Wagner 1975: 7 note 1). I am not suggesting here that the presented models of conversion lack any explanatory value in themselves, nor am I arguing for a kind of fusion between them which would only leave their premises and consequences intact. As stated earlier, the limitations result from the fact that none of the models in question appears to allow for the possibility of a purportedly exterior culture also experiencing alterations as a result of the contact. In other words, although the effects are registered by each model in different ways, none of them affords a symmetrical view enabling us to see that the colonizing culture may be or perhaps inevitably is transformed and that this occurs precisely because of the challenges posed by the inadequacies stemming from this approximation.
Following the models of conversion described earlier, both the cultures in question, native and foreign, can be imagined via the paradigm of original culture versus diaspora culture proposed by Manuela Carneiro da Cunha (1987:99). Following this view, a culture has to be conceived as a kind of self-contained totality, occupying a determined size. When contact occurs, elements may even be offered to other cultures, but there is an undeniable process of loss in both. Additionally, in a similar way to what happens with the concept of society (Barth 1992:18), taking cultures as closed units enables a simplistic separation of endogenous and exogenous processes, as well as the subtle implication of models belonging to the nation-state as the organizational matrices of all human thought. Here we can cite Stratherns argument (1992:77) that the modernist and pluralist vision of a world full of discrete whole units has dissolved into a post-plural world that requires other discursive aesthetics. In analyzing a conversion process, therefore, we can speak of a common historical experience in which the incorporation and re-elaboration of a new cultural repertoire has taken place [...]. (Serra 1995:101).
Here, it should be emphasized, it is not a matter of rejecting any of the interpretations highlighted in our survey of different models of conversion, but of relating them in some way, making them collide with each other, converting them, and in the process encountering a way of not ignoring any of the native assertions taking native here to mean all those implicated in the construction of the practices related to conversion. Setting out from the principle that the anthropologist must avoid deauthorizing the indigenous peoples in question, whatever propositions they may make, the question becomes how we can take seriously what are often highly disparate affirmations concerning conversion. Anthropologists usually ignore these religious phenomena, while missionaries are in the habit of exaggerating them (Pollock 1993:190-191; Vilaça 2002:57), though this is not a general rule. After all, as we have seen, there are cases in which some natives claim to be converts, reborn, and others in which they say they never underwent any kind of conversion, having always known and been part of the religion in question. There are also other situations in which the indigenous peoples are postulated to have converted easily (something which they themselves may well deny), or that their conversions were not legitimate (which the natives may also contest). The discourses themselves may be absolutely contradictory if seen in conjunction, but this does not impede the anthropologist from considering them simultaneously: incommensurability does not mean irrelationality. Doing justice to the diversity apparent in research involves avoiding reducing particular discourses to the terms proposed by another (including anthropology itself). There is no problem, therefore, in asserting that two apparently self-excluding arguments may both be true. Indeed it is possible for an anthropologist to say simultaneously that, yes, conversion took place and, no, conversion did not take place, as long as this is supported by the claims of the natives: after all, they themselves may say both things in some circumstances. And as Meyer suggests (1999:xixxx), an anthropological study of conversion needs to account for all those involved in the process.
Here we need to avoid thinking in terms of cultural change since this notion requires assuming discrete units, complete in themselves, that enter into communication in order to alter each other in some way, albeit in the most diverse forms possible. Obviating the idea of cultural change enables us to understand culture as something in permanent transformation, continually constructed, conferring a rich meaning to conversion as a specific event. It is only by assuming that there is something fixed, solid and crystalized that we can think of moments of flexibility, transformation and change:
Conversion in its most usual sense seems to presume that religious beliefs and practices form an internally coherent and comprehensive whole that is appropriately acquired (if not always acquired) en bloc by converts. Moreover, religions, in this view, are preferentially exclusive; [...] syncretic religions are interesting precisely because they seem to violate these basic assumptions. (Pollock 1993:170)
By abandoning the missionary requirement to substantialize units, it ceases to make sense to speak of continuity (or worse still of survival) in opposition to the notion of change. If instead the anthropologist sets out from the idea that his or her subject matter consists of complex and partially connectable multiplicities, another picture emerges a scenario that does not prevent us from thinking in terms of conversion, as long as this term can mean something different, or even possess a multiplicity of meanings, as shown, for example, in the work of Calavia Sáez (1999:47). Conversion would come to designate, perhaps, a kind of transformation or translation, a relation between versions, themselves in continual effervescence.
BARTH, Fredrik. (1992), "Towards greater naturalism in conceptualizing societies". In: A. Kuper (ed.). Conceptualizing society. London/New York: Routledge, pp. 17-33.
BIRMAN, Patricia. (1996), "Cultos de possessão e pentecostalismo no Brasil: passagens". Religião e sociedade, 17 (1-2): 90-109.
BURCH, Ernest S., Jr. (1994), "The Iñupiat and the christianization of Arctic Alaska". Études/Inuit/Studies, 18 (1-2): 81-108.
CALAVIA SÁEZ, Oscar. (1999), "Os ‘homens sem Deus’ e o cristianismo: para um estudo dos fracassos missionários". Religião e sociedade, 20 (2): 39-53.
CAPIBERIBE, Artionka. (2004), "Os palikur e o cristianismo: a construção de uma religiosidade". In: R. M. Wright (org.). Transformando os deuses. Volume II: igrejas evangélicas, pentecostais e neopentecostais entre os povos indígenas no Brasil. Campinas: Editora da Unicamp, pp. 55-99.
CARNEIRO DA CUNHA, Manuela. (1987), "Etnicidade: da cultura residual mas irredutível". In: Antropologia do Brasil: mito, história, etnicidade. São Paulo: Brasiliense, pp. 97-108.
CHARLESTON, Steve. (1996), "The Old Testament of Native America". In: J. Treat (ed.). Native and christian: indigenous voices on religious identity in the united states and canada. London/New York: Routledge, pp. 68-80.
CRAPANZANO, Vincent. (2000), Serving the Word: literalism in America from the pulpit to the bench. New York: The New Press.
D’ANGELIS, Wilmar. (2004), "O SIL e a redução da língua kaingang à escrita: um caso de missão ‘por tradução’". In: R. M. Wright (org.). Transformando os deuses. Volume II: igrejas evangélicas, pentecostais e neopentecostais entre os povos indígenas no Brasil. Campinas: Editora da Unicamp, pp. 199-217.
GEERTZ, Clifford. (1973), "Thick description: toward an interpretive theory of culture" In: The interpretation of cultures. New York: Basic Books, pp. 3-30.
GOW, Peter. (2006), "Forgetting conversion: the Summer Institute of Linguistics mission in the Piro lived world". In: F. Cannell (ed.). The anthropology of Christianity. Durham: Duke University Press, pp. 211-239.
HARDING, Susan. (1991), "Representing fundamentalism: the problem of the repugnant cultural other". Social research, 58 (2): 373-393.
HEFNER, Robert W. (1993), "Introduction: world building and the rationality of conversion". In: R. W. Hefner (ed.). Conversion to Christianity: historical and anthropological perspectives on a great transformation. Berkeley: University of California Press, pp. 3-44.
HOLBRAAD, Martin. (2006), "Response to Bruno Latour’s ‘Thou shall not freeze frame’". Rede Abaeté de antropologia simétrica. Disponível em: http://abaete.wikia.com/wiki/Response_to_Bruno_Latour%27s_%22Thou_shall_not_freeze-frame%22_%28Martin_Holbraad%29.
HORTON, Robin. (1975), "On the rationality of conversion". Africa, 45 (3): 219-235, 373-399.
LATOUR, Bruno. (1984) , Reflexão sobre o culto moderno dos deuses fe(i)tiches. Bauru: EDUSC.
____. (1991) , Jamais fomos modernos: ensaio de antropologia simétrica. São Paulo: Editora 34.
____. (2002), War of the worlds: what about peace?. Chicago: Prickly Paradigm Press.
____. (2005), Reassembling the social. Oxford: Oxford University Press.
LAUGRAND, Frédéric. (1997), "Le siqqitiq: renouvellement religieux et premier rituel de conversion chez les Inuit du nord de la Terre de Baffin". Études/Inuit/Studies, 21 (1-2): 101-140.
____. (1999), "Le mythe comme instrument de mémoire. Remémoration et interprétation d’um extrait de la Genèse par un aîné de la Terre de Baffin". Études/Inuit/Studies, 23 (1-2): 91-115.
LIENHARDT, Godfrey. (1961) , Divinity and experience: the religion of the Dinka. Oxford: Oxford University Press.
MAXEY, Rosemary McCombs. (1996), "Who can sit at the Lord’s table? The experience of indigenous peoples". In: J. Treat (ed.). Native and Christian: indigenous voices on religious identity in the United States and Canada. London/New York: Routledge, pp. 38-50.
MEYER, Birgit. (1999), Translating the Devil: religion and modernity among the Ewe in Ghana. Trenton: Africa World Press.
POLLOCK, Donald. (1993), "Conversion and ‘community’ in Amazonia". In: R. W. Hefner (ed.). Conversion to Christianity: historical and anthropological perspectives on a great transformation. Berkeley: University of California Press, pp. 165-197.
ROBBINS, Joel. (2004), Becoming sinners: christianity + moral torment in a Papua New Guinea society. Berkeley/Los Angeles/London: University of California Press.
REMIE, Cornelius H. W.; OOSTEN, Jarich. (2002), "The birth of a Catholic Inuit community. The transition to Christianity in Pelly Bay, Nunavut, 1935-1950". Études/Inuit/Studies, 26 (1): 109-141.
SAHLINS, Marshall. (1985), Islands of history. Chicago, London: The University of Chicago Press.
SCHULTZ, Paul; TINKER, George. (1996), "Rivers of life". In: J. Treat (ed.). Native and Christian: indigenous voices on religious identity in the United States and Canada. London, New York: Routledge, pp. 56-67.
SEGAL, Alan. (2003), "Text translation as a prelude for soul translation". In: P. G. Rubel & A. Rosman (eds.). Translating cultures: perspectives on translation and anthropology. Oxford/New York: Berg, pp. 213-247.
SERRA, Ordep. (1995), Águas do Rei. Petrópolis: Vozes.
SHAPIRO, Judith. (1981), "Ideologies of Catholic missionary practice in a postcolonial era". Comparative studies in society and history, 23 (1):130-149.
____. (1987), "From Tupã to the land without evil: the christianization of Tupi-Guarani cosmology". American ethnologist, 14 (1): 126-139.
STOCKING, George W., Jr. (1983), "The ethnographer’s magic". In: G. W. Stocking, Jr. (ed.). Observers observed: essays on ethnographic fieldwork. Madison: The University of Wisconsin Press, pp. 70-120.
STRATHERN, Marilyn. (1992), "Parts and wholes: refiguring relationships in a post-plural world". In: A. Kuper (ed.). Conceptualizing society. London/New York: Routledge, pp. 75-104.
TAYLOR, Anne-Christine. (1981), "God-wealth: the Achuar and the missions". In: Norman E. Whitten, Jr. Cultural transformations and ethnicity in modern Ecuador. Urbana, Chicago, London: University of Illinois Press, pp. 647-676.
____. (2002). "The face of indian souls: a problem of conversion". In: B. Latour & P. Weibel (eds.). Iconoclash: beyond the image wars in science, religion and art. Cambridge: The MIT Press, pp. 462-464.
VAN DER GEEST, Sjaak. (1990), "Anthropologists and missionaries: brothers under the skin". Man (N.S.), 25 (4): 588-601.
VILAÇA, Aparecida. (2002), "Missions et conversions chez les Wari’: entre protestantisme e catholicisme". L’Homme, 164: 57-80.
____. (2003), Big Brother Wari’. Os efeitos da idéia de Deus em uma cosmologia perspectivista. Comunicação apresentada no 51º Congresso Internacional dos Americanistas. Santiago (Chile), 14-18 de julho de 2003.
VIVEIROS DE CASTRO, Eduardo. (2002), "O mármore e a murta: sobre a inconstância da alma selvagem". In: A inconstância da alma selvagem: e outros ensaios de antropologia. São Paulo: Cosac & Naify, pp. 181-264.
____. (2004), "Exchanging perspectives: the transformation of objects into subjects in amerindian ontologies". Common knowledge, 10 (3): 463-484.
WAGNER, Roy. (1975) , The invention of culture. Chicago, London: The University of Chicago Press.
WEST, James L. (1996), "Indian spirituality: another vision". In: J. Treat (ed.). Native and Christian: indigenous voices on religious identity in the United States and Canada. London/New York: Routledge, pp. 29-37.
WOOD, Peter. (1993), "Afterword: boundaries and horizons". In: R. W. Hefner (ed.). Conversion to Christianity: historical and anthropological perspectives on a great transformation. Berkeley: University of California Press, pp. 305-321.
WRIGHT, Robin M.; KAPFHAMMER, Wolfgang. (2004), "Apresentação". In: Robin M. Wright (org.). Transformando os deuses. Volume II: igrejas evangélicas, pentecostais e neopentecostais entre os povos indígenas no Brasil. Campinas: Editora da Unicamp, pp. 7-32.
Received July 2007
Approved December 2008
1 All citations in the present article refer to the year of publication of the original version of the work in question. The date of the version actually consulted, where different, is found in brackets in the bibliography.
2 My thanks to Aparecida Vilaça, Bruno Marques and Marcio Goldman for their comments on an earlier version of this text.
3 Thereby allowing us to contemplate the specificities presented by these processes during the encounter between missionaries and these populations (cf. Wright & Kapfhammer 2004:14-17) across the world.
4 The texts by Birman (1996:94 note 9, 98-99) and Capiberibe (2004:61, 68-69, 84, 95-97) can be read in the same key, though they show a number of divergences in relation to the former.
5 Robbins (2004:339 note 2) also makes another persuasive criticism: if, according to Hortons theory, the process of conversion involves accepting only those features compatible with the traditional cosmology, would the latter not already provide an adequate response to the questions posed by the new sociocultural environment? If so, there would be no reason to convert in the first place.
6 The discourse of some Christian North American Indians, for example, states that they had always known God, even before converting to Christianity (Schultz & Tinker 1996:57-58). According to them, the novelty introduced by missionaries was not the figure of God, but that of Jesus.
7 The Urapmin themselves, it is true, also speak of two different conversions, but not in the same form as Robbins. While for the anthropologist the second conversion involved the understanding of the adopted religion in its own terms, for the natives the new conversion occurred after a profound ecstatic experience, the result of a religious rebirth (Robbins 2004:87, 131). In any case, according to the Urapmin, both conversions can, and should, be understood through the terms of Christianity itself.
8 Which does not necessarily signify a return to being what one was or what one always was, though these are possibilities.
9 For Amerindians, then, difference is valued in itself. This is not just a case of contrastive identity, the need for an other to enable the construction of the self. Rather, this type of relationalism implies the mutual and concomitant existence of an other-self. Hence what emerges as fundamental in Amerindian cosmology is not identity but alterity, including the specific alterity of white-becoming.
10 According to Hortons theory, the condition of possibility for religious alternation, as stated previously.
11 An approximation between conversion and conversation can be found in Meyer (1999:54). See too Clifford Geertzs comments (1973:13, 24) concerning the importance of not only speaking and listening, but talking.
12 Here I use the construction exchanging perspectives, rather than a change of points of view, a difference extracted from a text by Marilyn Strathern (1992:90, passim). See the idea of commutation below.
13 Which is only questionable for the anthropological program adopted here, as indicated earlier, not for the missionary.
14 Diversion, which may also be read as a detour or divergence, as long as we do not associate this with the idea of there being a correct route to be followed. The author employs the term diversion in opposition to aversion, forming the set conversion, aversion, diversion.
15 Cf. a similar argument in Martin Holbraad (2006).