The Territorialization of Difference

In Culture, Power, Place, Akhil Gupta and James Ferguson [1997: 4] underscore the need to undertake an anthropological study of the historical and political processes of cultural, as well as ethnic or national, territorializations. It is a study of this type that I now propose to present in this first part of my article, taking into account the importance of the role of the state in “the production and canonization of social classifications” [Bourdieu Reference Bourdieu2012: 24]. It should be noted that the state is to be understood here not as a monolithic construction, but as a set of institutions, each of which pursues its own, sometimes conflicting, aims. As regards more particularly the spatialization of culture [Gupta and Ferguson Reference Gupta and Ferguson1997: 3] as found in Brazil, it will be noted that, from the second half of the twentieth century on, the creation of separate geographical units based upon ethnic singularity has gone together with a diversification of the organizations entrusted with the management of difference, each of which has adopted its own specific norms. The solution of an “assumed isomorphism of space, place and culture” [ibid.: 34] initially applied to the Amerindian populations of the Amazon basin, then those of the Nordeste, was subsequently extended to the so-called quilombola groups and traditional populations. Henceforward, the “processes of legitimation and authentication” [ibid.: 14] of ethnicity involved a territorial status and recognition by an administrative organism.

The first stage in the territorialization of difference occurred in Amazonia in the 1960s when the Brazilian state took the decision to create three parks for Amerindian populations. Antonio Carlos de Souza Lima [2015] associates the context in which this decision was taken with the emergence of a “utopia”: the conviction that it was possible for contact with these groups to be non-confrontational. Without abandoning long-term assimilationist objectives, the aim was to pave the way for a less brutal transition towards modernity. The grant of vast geographical spaces was intended to enable Amerindian populations to “undertake a progressive, self-managed social transformation” [ibid.: 437]. Again according to this author, the internationally known Xingu park, established at this time, came to “showcase” Brazilian indigenism, presenting a place in which Indians could maintain “a way of life which was no longer possible in regions that had been colonized for longer” [ibid.]. The action of both the Indian Protection Service (IPS), and the National Indian Foundation (FUNAI), which succeeded it in 1967, is aimed at ensuring the territory’s protective function.

The establishment of groups in the Amazonian lowlands as a paradigmatic marker of otherness—referring to “veritable” Indians, not very different to those of the “Discovery” [ibid.]—was an essential step in that it gave support to the relegation of the populations of the Brazilian Nordeste (who had laid claim to their Indianness in the same period) to the category of “remnants,” that is, of pale copies of the past, who had lost their traditions and their languages. Moreover, as João Pacheco de Oliveira points out [1998: 53], the territorial question took on very different forms in the two regions: whereas in the Brazilian Amazon the issue was one of preventing the invasion of indigenous lands by outsiders (that is, one of facing up to a present threat), in the Nordeste it had to do, more fundamentally, with the “re-establishment” of territories that had disappeared under the impact of the “flows of colonization” (hence, with undoing the past). As Oliviera points out, re-creating these territories, in what was considered to be their original state, implied the departure of its non-Indian population; it should be added that the undertaking also involved the representation of a possibility of undoing a “mixture” [mistura]^{Footnote 15} until then incarnated in individual bodies and collective practices. These populations, ordinarily described as “Indians of mixed race,” could only obtain recognition as Amerindians if they were able to exhibit “the diacritical signs of Indianness,” the best-known example of which is the Toré ritual [Oliveira Reference Oliveira1998: 60; Arruti Reference Arruti2013]—in other terms, after visibly demonstrating to the administration the cultural contrast between themselves and the rest of the population. The Nordeste case in its own way helped to reinforce the idea that territorial integrity and ethnic purity were closely linked.

The question of the evidence of difference, which never surfaced with regard to the indigenous, historical lowland collectivities, whereas it was always problematic as regards the positions assigned to Amerindians in the Nordeste, has once again today become a pressing issue in connection with recent ethnic claims in Amazonia. Indeed, for the groups voicing these concerns, it is vital to be able to present a distinctiveness distinguishing them from their neighbours, who at times may also be relatives, and to make it visible to the authorities. The latter configuration can, it seems, be inscribed in the flow of models of singularization: just as Amazonian Indians such as those of the Xingu provided, for a time, an inspiration for their counterparts in the Nordeste, the latter in turn have served as an example for other inhabitants in the North region in that they embody the success of both their efforts to overcome internal difficulties and their struggle with the authorities to obtain recognition of a renaissance or ethnic resurgence. Progress, it would seem, can only be achieved if the Amazonian group sets the course towards the desired goal, and the Nordestino group embodies the means of attaining it.

The cascading proliferation of poles of reference, mirroring the Amazonian example with respect to “Indian purity,” and the Nordestino example with respect to the ability to “go back to one’s roots,” has thus fuelled the rise of ethnicity as the favoured form of expression and organization for social mobilizations. Moreover, the phenomenon has given rise to a, so to speak, thematic development, since it has proved capable of constructing not only an Indian but also a quilombola ethnicity. Indeed, in Amazonia, as in the Nordeste, otherness is no longer an exclusively Amerindian preserve. As Arruti notes, the case of the Indians in the latter region served as a “valuable example” [2006: 42] for both reflection on, and understanding of, present-day black “ethnicities,” of which the quilombola identity can be considered to provide an emblematic expression. Henceforward, ethnicity has become accessible to all sorts of populations provided, nevertheless, that a choice is made between the black and the Indian approaches, since the state has decreed their incompatibility of principle.

In appearance, the case of the “traditional populations” is different insofar as what defines them supposedly derives more from a specific way of exploiting natural resources (one respectful of biodiversity and consistent with sustainable development) than from a broadly “cultural” specificity (language, kinship system, religious practices, etc.). Their difference, then, is thought to derive above all from an “economic otherness.” It may be because of this non-ethnic nature that, unlike Indians and quilombolas who are presumed to be few in number, “traditional populations” are not associated with the idea of rarity, and this description could in principle apply to a large proportion of the inhabitants of riparian Amazonia. Thus, this more adaptable designation appears to be better suited to complex pre-existing situations. In reality, however, the legal label “traditional populations” is one that, like others, proves to be underpinned by the dual premise of exceptionality and territorial administration. Indeed, it is strictly reserved for those collective groups that have obtained official state recognition as such by virtue of their geographical inscription: the only beneficiaries of the rights associated with this categorization are those whose living space lies within the defined limits of what has been decreed a “Conservation Unit of sustainable use.”^{Footnote 16} Here again, it is the qualification of the territory that validates a recognized difference.^{Footnote 17}

From the perspective of the conditions of access to specific rights, again in connection with the legal status conferred upon occupied land areas, it therefore seems legitimate to consider the overall array of juridical categories referring to Indians, quilombolas and traditional populations. The fact that the recognition of difference in no way signifies a grant of autonomy is another consideration that militates in favour of this stance: whatever the territorial modalities in place, the associations, which the populations are obliged to set up, are systematically referred to public administrations at the Federal-state, state, and municipality levels for all their administrative procedures.^{Footnote 18}

From this more holistic perspective, it is not just the conception of otherness that changes: now multiform rather than singular, and gradual rather than radical. The reconfiguration of the ethnic model also influences expectations linked to the delimitation of “territory.” Thus, it could be argued that in the pre-1980s period the demarcation of a geographical space might be seen as the result of a state decision benefitting from input by Indians and their allies: the purpose of land attribution was to set a collective group apart from the rest of Brazilian society in order to “preserve” its singularity and its integrity. In contrast, as early as the end of that decade, the curtain came down on territorial delineation, and in public administrations the applications started piling up. The number of the latter actually dealt with reduced to a trickle, and their processing no longer took place collaboratively, but confrontationally, and as a result of pressure from militants and local populations. Thus, the more difficult it became for territory to emerge, the less able it was to fulfil its protective role.

How, then, can one explain that such slow progress in land certification has had little or no impact on the ever-increasing stream of new, ethnicity-based claims? It may of course be considered that such mobilizations will almost naturally be organized under the rallying cry of territory, since it is territory that underpins the development of all social life. It is indeed a fact that the reproduction of a social collectivity is dependent on its control of a geographical space. However this argument fails to take into account the discrepancy between juridical conceptions of territory—closed, contiguous, and peopled by a single type of population—and local conceptions of the notion in which spatial discontinuity is perfectly conceivable [Gallois Reference Gallois and Ricardo2004: 39, for instance, with reference to the Wajãpi]. We therefore now need to propose a re-definition of territorial demands in terms that are more political, and closer to local representations, as an attempt to have the group’s voice heard at every level of the state apparatus by speaking the language of the latter, appropriating its presuppositions, and offering it the homogeneous “us” which it expects, in order to obtain its protection.

Scrutiny of the term “protection” allows us to examine, at this point, a broader spectrum of rights associated with the different ethno-legal categories. Land-security rights are the most emblematic of these, and everyone is well aware of the land pressure to which local populations are subjected as a result of encroaching monoculture, expanding stock-rearing areas, illegal logging, and dam projects. Of all these issues, land security is also the most problematic, as it clashes head on with private and public interests, hence the length of time it takes for demarcation procedures to be concluded. But we also need to take into consideration another group of rights concerning so-called differentiated education and health-care (involving the creation of jobs as primary-school teachers or health-workers). I propose to describe these as “secondary” or “derivative” rights, insofar as the granting of such rights is conditional on the formulation of a territorial demand. Indeed, they are reserved for collective groups officially recognized as “different” from Brazilian national society, that is, for populations that state administrations circumscribe or envisage circumscribing within a given geographical space, that is, a territory. In reality, the implementation of these advantages is very often observed well before the demarcation project itself even reaches the first stages of its long passage through the administrative process.^{Footnote 19} Hence, these rights are seen by the populations concerned, on the one hand, as more accessible than landholding readjustments and, on the other, not merely as “secondary” considerations but as fundamental dispositions, (a point evinced by the recurrent nature of such demands).

In order to understand the meaning locally attributed to territorial demands we need to begin by clarifying the position of the organisms entrusted with the management of difference. In this connection, Souza Lima’s analysis proves extremely helpful; this author writes of “state-embodied power invested in agencies that seek to extend their scope nationally as an administrative network covering the totality of the political map of the country” [2015: 431]. Such agencies, he continues, endeavour to “label collectivities generically by linking them to supposedly distinct spaces and practices and by attributing to them a place in a codified, hierarchized system of positive and negative assignments” [ibid.: 432]. These public administrations, all of which seek to contribute to infrastructure implementation throughout national space, envisage territorial demarcation as merely involving the rubber-stamping of ethno-cultural singularity: in this view, the grant of land-tenure follows on from presupposed observability and self-evidence, and the initial filing of an application simply corresponds to the beginning of an administrative process that ends with a presidential decree.

The main conclusion drawn by local populations from this mechanical representation of the chain of actions is that “difference” is a criterion capable of setting in train what is seen as the overarching machine of the state. But to this consideration they add a further, constitutional criterion, according to which ethnic identity depends, above all, upon self-definition. This self-proclaimed difference has to be processed by the various administrations in accordance with approved dispositions, which are of a territorial nature. It is therefore the demand for land which, by giving concrete expression to the affirmation of singularities, actually obliges state institutions to interact with the various populations (by sending technicians and anthropologists to them). Thus, the articulation of the equation between difference and territory is powerfully distorted. It would only be a slight exaggeration to say that the relationship between the two terms ends up being reversed: rather than it being the case that the depiction of a singularity precedes and leads on to the delineation of a geographical space, it is the existence of a demarcation project, of itself, that is seen as constituting proof of a difference.

In sum, it is possible to suggest that populations, including those until recently still labelled according to a generic terminology (mixed-race, peasant, etc.), have seized upon this close association of “territory” with “difference,” while subverting its logic in order to be able to access other rights—of a non-territorial nature, but dependent upon a territorial claim. Now that the action of the state has led them to realize that spatial delimitation represents an administrative validation of their “difference,” they see such projects as indicators of a distinction which, in itself, enables them to acquire a certain visibility and a social existence: in other terms, a passport to reduced class-sizes, improved canteen services and health-monitoring. The fact is that, while the initial filing of an application in no way guarantees a successful outcome, it at least allows groups to demand the application of the “derivative” rights associated with the legal category relevant to their application. Local populations, therefore, are constantly obliged to territorialize in order to survive. To my mind, this largely explains why the stream of territorial claims has not dried up even though the state has clearly withdrawn from the process.

Criticizing the fascination for what he terms the “savage slot,” Michel-Rolph Trouillot writes that a “more perverse consequence of the eulogy of Otherness is the production of restrictive identities that give most Others few choices in defining themselves or in changing the terms of their relations with the unmarked” [2003: 74]. The factors that seem pertinent for social identities (religion, race, gender, etc.) ought to be even more powerfully relevant when identities are enshrined in law. Indeed, the more restrictive ethnic categories become, the more exclusive (from a legal perspective) they also become. And yet, as I would like to demonstrate in the following part, local populations in Brazil do not simply passively undergo the effects of these classifications—under certain conditions they are able to appropriate them for their own ends.

Ethnicity as a conversion of identity

The problem of “identity manufacture today” [Boccara Reference Boccara2005: 13] is clearly posed by Élisabeth Cunin in her introduction to a special issue of the journal Autrepart: should ethnicity be addressed from the perspective of localization—“its association with a place informing cultural difference [Wade, Reference Wade1997, p. 18]” [Cunin Reference Cunin2006: 3]—or should it be inscribed in a transnational space, taking into account “the local reinterpretation of globalized cultural signifiers [Warnier, Reference Warnier1999]” [Cunin Reference Cunin2006: 4]? Or, to reframe the question: does the patrimonialization of particularisms and their integration into a system aimed at signifying to various Others (geographical neighbours, public administrations, etc.) a singularity, constructed according to preconceived models—that is, as a sort of widely disseminated “showcase-ready” version of this singularity—fuel a process of base-level patrimonialization grounded in the everyday experience of various populations?

I propose to shed new light on these questions by addressing the subject of ethnic declarations through the prism of a comparison with religious conversion. Studies of adherence to Christianity are particularly relevant to the reflections offered in this article insofar as they draw attention to the importance of the appropriation and re-signification, by the populations concerned, of categories that are outside their social world. They are also relevant to understanding how such appropriation and re-signification can give rise to apparently innovative social groupings with their own hierarchies and places of sociability such as, for instance, “congregations,” “prayer groups,” etc.

Among the main advances resulting from discussion on conversion, we need to mention the uncovering of a narrative which was, retrospectively, centred on the radical turning point between a “before” of wandering in self-ignorance, and an orderly “after,” in accord with a truth held to be superior. As Danièle Hervieu-Léger writes, when the convert retraces his path, he “realizes the fundamental postulate according to which an “authentic” religious identity can only be a chosen identity” [Hervieu-Léger Reference Hervieu-Léger1999: 129]. Now, in the case of the Brazilian Amazon, not only has the theme of “choice” made its appearance in the most recent claims of ethnicity, but it has come to form their most salient feature: “We were like just any other inhabitant until we made this choice” [Eramos como moradores comuns até fazer essa escolha], is what, in substance, we hear over and over again. This choice, which is seen as personal and deliberate, is explicitly associated therefore with the exercise of a constitutional right that allows populations, previously seen by city-dwellers as being of “mixed-blood,” to assert their traditional, Indian or quilombola identity [Boyer Reference Boyer2015]. Thus we see that the articulation between choice and authenticity mentioned by Hervieu-Leger is also an underlying factor here: the fact of declaring a type of ethnicity in the name of the principle of self-determination is supposed, in itself, to guarantee its authenticity, and this is considered to be the case despite the accusations of the state, and sometimes of militants, who constantly draw claimants’ attention to the lack of evidence of their alleged difference [ibid.]. Once it is acknowledged that belief establishes a fact’s social existence beyond all doubt and independently of appearances, the object of enquiry shifts from sincerity to forms of adherence.

In this connection, researchers whose studies have addressed more particularly the topic of affiliation to Evangelical churches have underscored the need to take native categories of conversion into account insofar as the latter “inform the discourse and experience of those who describe themselves as ‘converts’” [Mary Reference Mary1998: 12]. From this standpoint, the notion should not be understood as an immediate, linear, generic change, but as a sinuous, complex and sometimes paradoxical process, which can only be grasped in the terms in which it is understood locally. Now, to stretch the religious metaphor to declarations of ethnicity, I consider it productive to interpret the foregrounding of the “rescue” [resgate] of “memory,” of “culture” and “tradition,” as the keywords of this conversion-like phenomenon, that is, as a reappraisal of practices and discourses, spaces and representations, which, however, does not necessarily affect all the latter, or in a uniform way. In the case of Indians, the designation of an official healer [pajé] does not mean that he alone fulfills this role nor even that he is the individual most often consulted by patients, any more than the construction of a community hut [maloca] means that guests are necessarily housed in it. As for quilombolas, their education in the African meta-narrative—encouraged by the teaching materials distributed in their schools—portraying that continent as representing their only roots and enslavement as afflicting the totality of their ancestors, has not led to the disappearance of the narrative of integration through union and co-parenting of outsiders to the villages, whatever their origins, any more than the promotion of religions from the “African matrix” has led them to prefer “orixás,” that is, African deities, to properly Amazonian “enchanted” spirits [mestres, caruanas].^{Footnote 20}

In a certain sense, these discrepancies between prescribed and effective behaviours are comparable to the divergence between the definition of what is appropriate behaviour for a good Christian and the Evangelical churches’ tolerance of moments of estrangement [afastamento]: the former is proclaimed in public space but also relativized in the private sphere, recognized as an ideal and adapted in practice. Similarly, the terms associated with the undertaking of resgate (like those related to salvation, moreover) make it possible to define intermediate positions between those who have already “made their choice” and those who have not, whether in the local arena (when dissident voices can be heard) or that of the surrounding area (when neighbouring villages do not assume the same “identities”). The adoption of these positions can have various implications: while they sometimes result in the isolation of a kin group or a relative who refuses a particular ethnic identification, they may equally well not lead to the interruption or restriction of relations within circles of sociability.

Among the new categories, that of “relative” warrants attention as it is abundantly present in the language of ethnicity: its definition includes not only all those who are linked by filiation, marriage or co-parenthood, but also all those who have made the same choice of identity. In order to illustrate how these two acceptations of kinship can co-exist, I will take the example of certain components of an ethnographic situation in the central region of the Brazilian Amazon, in which a conflict had recently arisen between Indians and quilombolas [Boyer Reference Boyer2015].

– The inhabitants of hamlet 1, like three other neighbouring localities, chose to self-identify with a well-known Amerindian group (ethnic group A), located about two hundred and fifty kilometres away as the crow flies, with which they had not had any contact before their recent “declaration” of their “Indianness.” Since then, they refer to members of group A as their “relatives.”

– Three kilometres away, the members of village 2, for their part, decided to join the struggle undertaken by four other hamlets for recognition as a black ethnic group (ethnic group B). They now all consider themselves as “relatives.”

– In hamlet 1 (henceforward self-declared as Indian), the cacique’s lieutenant, who declared himself as belonging to ethnic group A, was the father of the woman who presided over the quilombola association of village 2, which had declared itself as belonging to ethnic group B. No mention of his visits to see her was ever made in public meetings.

– In village 2 (henceforward self-declared as quilombola), the leaders denounced the hostility to quilombola mobilizations of certain “outsider” residents. When I met one of the latter, a woman whose husband was indeed born in another locality, she nevertheless declared herself a relative, while at the same time refusing any association with ethnic group B.

These examples highlight the power of the idiom of kinship to create a sort of optical illusion fusing together relationships with different social foundations. On the one hand, the term designates affectively charged relationships, partaking in the solidarities of everyday life or activated when needed, which people seek to reinforce and diversify though marriage and/or co-parenthood. These relationships, which are always established within webs of interacquaintance often broad enough to have backing in urban circles, create personal ties whose genealogy can always be traced by identifying the components of their support network. On the other hand, the expression is used of intentional coalitions between broad collectivities proclaiming their homogeneity. These alliances, which are mainly preoccupied with publicizing their action, and whose cohesion derives from advocacy for their demands, do not necessarily materialize into unions or sustained exchanges.

Thus there emerges a contrast between the plane of political action, turned towards the future and implying adoption of a common stance vis-à-vis the outside world, in particular state public administrations, and that of everyday interpersonal relations, nourishing memory and concrete collaboration in local space. In order to avoid the reification of identities, it is indispensable to keep in mind that the same term of “relatives” refers to two distinct registers: moreover, it would be all the more unreasonable to surrender to the illusion of their superimposition as not one of the anthropologist’s interlocutors confuses, except as the result of a situated discursive effect, the category of “relatives” as constructed in everyday life with that proclaimed in the context of militant activity.^{Footnote 21} Whereas the usual mode of self-definition—by mention of a place (“I was born in such and such a hamlet, I live on such and such a branch of a river, etc.”) or a collectivity (“I belong to this or that family, this or that community”)—inscribes people in the domain of the known, the use of generic labels opens up a field of challenging possibilities, in which the key issue at stake is precisely that of how to embody these designations successfully at the local level.

Favouring a political conception of the notion of “relative” allows people to distance themselves from “relatives” envisaged from the perspective of the domestic sphere, which in most cases also includes neighbours, and to disqualify them when there is disagreement over ethnic choices.^{Footnote 22} Conversely, putting forward family ties sometimes represents an argument legitimizing criticism of the shape of militant coalitions, or even of their very principle. It is worth noting that the same speakers might be led to foreground either of these perspectives in function of their perceived strategic interests at a particular moment in time or their addressees’ expectations. Vis-à-vis the outside world, it may be fair game to justify a rivalry between relatives, on the basis of difference, and a good idea to proclaim an alliance on the basis of likeness, even if, within the group itself, everyone is aware not only of ties of kinship linking enemies, but also of possible tensions, or even the absence of affinities between certain co-belligerents.

Social mobilizations occurring around issues of ethnic re-composition raise a double question. On the one hand, that of singularization which can be articulated as follows: how do populations that previously saw themselves as identical come to distinguish themselves from neighbours and relatives to whom they are close in the strict sense of the word? And on the other, that of the structuring of a space of multi-scalar identity-based demands, which can be formulated thus: how is it that collectivities that do not necessarily acknowledge a common ancestry, nor even share a long-standing sociability, come to embrace a common ethnic destiny?

The analysis of the differentiation and/or coming together of population groups needs to take into account not only local stances and anchor points, but also opportunities for access to wider political arenas. In other terms, it is essential here to take equally into account the general sociology of the situations under study (actors in presence, administrative framework, types of interlocution, etc.) as well as aspects of their internal dynamics (the ways in which hamlets were formed, economic collaborations, religious networks, etc.). In the case in point, by linking the competition between neighbouring villages for control of, among other things, the natural resources vital for their reproduction to the reconstruction of the chronology of individuals’ contacts with quilombola or Indian militants, it has been possible to show how, progressively, public expressions of separate ethno-legal identities follow on from each other. Thereafter, ethnicity can be seen as a resource to which collectivities resort as a way of coping with already existent tensions, and political kinship as a tool in its service: in the case mentioned above, the Indians assumed this new identity in 2010 as a sort of retaliatory measure for the shift in the balance of power brought about by their quilombola neighbours’ earlier ethnicity declaration, dating to 2005.^{Footnote 23}

This series of declarations of Indianness and quilombola-ness was certainly fuelled by the fact that these identities lay at the intersection of the very concrete problems and issues at stake facing the populations in question: offers of services and mediation at the regional level, the construction of a legal framework, and political decisions at the national level (which, as we regrettably see today, are always reversible^{Footnote 24}). However—and this refers us back to the hypothesis of the globalization of ethnicity and the circulation of transnational models mentioned at the beginning of this section—the general diffusion of a conception of “difference” as the royal road to emancipation would not have gained acceptance, had the images of autochthony not been the object of properly native reinterpretation and re-appropriation. Thus a rapid overview of the literature suggests that recourse to the criterion of “dark earth”—a soil quality evincing an ancient human presence—as a sign of Indianness or, more intriguingly, a reference to quimbanda—ordinarily associated with black magic—as a sign of negritude is only to be found in Amazonia or even, perhaps, uniquely in the group of villages mentioned above. Moreover, in many situations, positions do not yet appear to actually be ossified, as exemplified in a particular hamlet a ten-hour boat ride away from a medium-sized regional town,^{Footnote 25} in which members of one phratry articulated differently the Indian and the black lineages which they recognized as shared by all: one of the brothers constructed a family-history narrative centred on a certain Indian grandmother and a black grandfather, whereas a second brother favoured his black roots, and a third his Indianness. Yet, at this particular point in our interviews and taking into account the webs of interlocution of which they were then part, our interlocutors were in unanimous agreement over the adoption of a quilombola political identity.

It would not be unreasonable to object that the introduction of these nuances in no way invalidates the general framework of the institutional valorization of otherness at a global level. Indeed, its vitality is amply demonstrated by the consolidation of the multicultural model in all Latin American countries, to mention but this sub-continent. But it might also be thought that this interpretation fails to recognize sufficiently the convergence between identity and belonging, between political mobilization and the everyday experience of life: ultimately, the Indians, like the quilombolas, have earned recognition as such in the political arena because they have unceasingly asserted their particular identity. The way out of this impasse between constructivism and essentialism implies that we unhesitatingly consider ethnic declarations of whatever sort with all due seriousness, while questioning their historicization and their deployment in social space, along with their concomitant changes and continuities, innovations and numerous usages.

It further needs to be underscored that, when internal conflicts are seen to occur, they are never a logical, systematic consequence of the emergence of ethnic demands. Moreover, in some contexts, populations that consider themselves as different may come together in the face of common threats.^{Footnote 26} It would also be inaccurate to say that possible tensions are created solely by the implementation of a plurality of ethno-legal identities. Certain specialists, such as Alfredo Wagner Berno de Almeida and Roberto Sanches Rezende [2013], quite rightly prefer to point to the inefficacy of the various public administrations and call for the reinforcement of collaboration between these agencies. However, once ethnicity is understood as a language that, though obviously developed at the national level, is pressed into service on local issues, it also seems possible to contend that the populations may seize upon it as a further register in which to set out the points of disagreement (mostly of a political and economic order) that impact them directly, such as: the mode of attribution of the promised civil-service jobs, the drawing-up of electoral registers, the definition of land holdings, etc. To come out for or against the “chosen identity”—a stance which may change over time and is therefore seen as reversible—is, then, to take position on the legitimacy of an action that is determinant for the future of the collective group.

Such identifications stand in the same relationship to identity, then, as ethnogenesis to ethnicization [Boccara Reference Boccara2007: 71]: one of “endogenous creative processes” versus the “techniques of power” [ibid.] deployed by the state, and hence of a clearly different nature. Thus they may at times be expressed in terms of territory, or inclusion in a restrictive legal category, when populations consider such strategies advantageous, but they may also break free from this constraint via new mutations, including the adoption of other ethnic forms as exemplified, for instance, by certain traditional populations or quilombolas who now proclaim their Indianness. Examination of such instances of identity recomposition through the prism of religious conversion seems to me to offer a promising approach insofar as it encourages us to take account of the local usages of discourses and apparatuses formulated at the national, or even international, level.

The ethno-politicization of minority-group status

The study of Indians, “traditional populations,” and quilombolas is generally addressed through the notion of “singularity,” a distinctiveness that needs in every case to be identified and specified in relation to both other juridical categories and the rest of the population. This perspective often lends support to the idea—one seemingly self-evident and grounded in nature (or rather in culture)—that the demands of populations opting for a different label are also aimed at pursuing specific objectives on account of their ethnicity. Such emphasis on the supposedly particular interests of each of the three main groupings goes back to an image of fragmented social struggles, obscuring the fact that these populations are all caught up in similar dynamics, which lead to dependency and perpetuate their subordination.

The various demands formulated with the help of the ethnic register can nevertheless be considered as fundamentally identical insofar as they are all aimed at obtaining improved living conditions. This is evidenced by the common fronts put up against certain dam projects, for instance. From this perspective, parallels can be drawn between these mobilizations and others which, in certain places—irrespective of the local context (type of property, of agriculture, etc.)—came together, and continue to do so, under the single banner of peasant struggle. These considerations encourage a shift in perspective that reveals the fundamental difference—more of an argumentative than a sociocultural kind—between rural mobilizations and ethno-legal demands: whereas the former emphasize the largest common denominator in order to unify the different types of discontent, the latter choose to play the card of exceptionality, using particularism as the favoured norm and form of expression for social protest. The entitlement of all to land is thus replaced by the right of different groups to distinct territorial statuses. This observation leads us to reflect on the reasons for the success of what can thus be seen as a new standardization of the forms of political engagement.

We will pursue the comparison of ethnic self-identification with the conditions of religious adherence as this provides a useful starting point for our reflection. The theme of “conversion” draws attention to the fundamental role played by intermediaries. In the case of Evangelism, we find people coming from elsewhere, often working for themselves and seeking to create their own congregation. Associating Catholicism with a deviation from the original church, which they seek to restore, they claim to be motivated by the passion of Jesus and devotion to others. This in turn raises further questions: is it possible to identify, in what I call identity “conversions,” the presence of outside actors propounding a discourse of renewal, and what are the forms taken by their interventions? The answer to the first question is unquestionably yes: besides urban ethnic militants, various social-science specialists, members of non-governmental organizations, Catholics linked to progressive currents in the church, and members of educational teams particularly dedicated to their missions participate in these mobilizations, at least in their initial phase, if not over the longer term. The different types of action aimed towards local populations can be summed up as follows: certain actors help to create opportunities for new alliances by fostering the establishment of contacts with other groups; others help to improve knowledge of the procedures and rights available to the different ethno-legal categories by organizing numerous information meetings; other actors again organize “cultural” workshops devoted to language learning, home-made remedies, body painting, etc., and thereby promote awareness among these populations of elements that could signal a singularity that would back up the demands they address to public administrations.

The mechanisms involved in these interventions, promoting dialogue in a hopefully tight-knit collaborative collectivity, are not unlike what, in his study of the establishment of settlements [assentamentos or “agricultural colonies”) by the National Institute For Colonization and Agricultural Reform (INCRA) in the Nordeste region of Brazil, Benoît de L’Estoile terms “meetings,” and which involve, he writes [2015: 8]: “a ritualized form [of confrontation opposing] two socially and culturally heterogeneous worlds that find themselves in an asymmetrical interdependence relationship: the world of interpersonal relations between beneficiaries of agrarian reform, and the world of the state” embodied in its representatives, and which obeys a bureaucratic logic. Although ethnic intermediaries are not necessarily civil servants (though some are both), they perceive themselves—and, likewise, are perceived—as outsiders to the hamlets which they visit. As educated city-dwellers, they also represent an opening onto another world: that of “knowledge,” “authority,” and “power”—a world that now grants “tradition” a place within “modernity.” Lastly, as in the case studied by de L’Estoile [ibid.: 11], they have introduced forms of interaction that could be considered previously “unknown” to populations familiar with “informal, weakly ritualized” practices: public speaking, contradictory debating, arguing the case, are not behaviours that come naturally to populations accustomed to eschewing conflict through avoidance, absence and silence, and to adhering to forms of solidarity expressed principally in the language of kinship. This coming together of participants belonging to “heterogeneous worlds” is, then, equally visible and may even be pursued in the absence of intermediaries. For instance, in June 2017, in an Amazonian village, what was intended to be an handicraft workshop instantly turned into the setting for a formal meeting—quite unrequested by me—as soon as I went into the maloca: my presence provided me with the opportunity to hold a class on oral expression for the women there, who immediately interrupted their activities to listen attentively to my thoughts on how to speak to outsiders.

The “concrete mechanism of multiculturalism” has been illuminated by Guillaume Boccara and Paola Bolados [2010: 655] in an article addressing the formation of a “bureaucratic field of intercultural health care” in Chile. Among the features that “broadly characterize ethno-governmental practice” [ibid.: 661], the authors mention: 1) the depoliticization of social problems, 2) the role of indigenous civil servants as intermediaries travelling between different spaces, 3) the socio-ethnic division of labour which reinforces domination based on access to cultural, technical or symbolic capital, 4) the designation of some of the Indians present as legitimate representatives of their “communities,” and 5) the construction of the Indian population as a homogeneous whole [ibid.: 661-662]. While the features listed here apply to the case of Chile, they also can be found to coincide in many respects with the analysis of the Brazilian case put forward by de L’Estoile. Indeed, a common underlying triptych characterizes these parallels: an awareness among external actors of belonging to a different social group and of having come to deliver a message, a conception common to the latter of local collective groups as unanimous entities free from internal tensions, and a pressing demand for privileged interlocutors in these groups who are subsequently defined as “representatives of the community.”

On this last point, it is obvious that the viewpoint of these ethnic intermediaries is overarching. They seek to put their understanding of the language and demands of public administrations, their experience in the area of mobilizations, and their knowledges in the area of culture at the service of groups that they now only perceive through the prism of a supposed ethnic difference. Indeed, it is the supposed otherness of their interlocutors that justifies their presence and their attentiveness. As for the villagers, it is not in their interest to abolish a social distance between themselves and this heterogeneous group of actors insofar as it is this very distance that provides them with the possibility of accessing resources. Nevertheless, while all, intermediaries and natives alike, agree that the former come from outside [de fora], they are also recognized as full participating actors in local situations:^{Footnote 27} their outsider status is, then, a marker not of neutrality or non-intervention but of a social role.

On the second point, in his study of pioneer fronts on the Trans-Amazonian highway, Roberto Araújo [Reference Araujo1993] has masterfully demonstrated that the “communitarian illusion” is anchored in its Christian genealogy. This in turn prompts us to suggest that present-day social demands may represent a newly updated version—now less religious than ethnic—of the restoration of community purity: in the Amazonian cases studied in the course of our research, as in other instances addressed in Brazilianist literature, one can easily identify this idea of a “community” free from dissenting voices, headed by an uncontested leader, preoccupied with the “rescue” [resgate] of its “culture,” and thought to be aided by devoted, competent intermediaries—even when closer examination reveals internal conflicts and clientalist relations. Which brings us to a final point: these intermediaries, without necessarily being the instigators of possible redefinitions of local power relations, do in fact influence events in order to bring about the emergence of a “representative of the community.”

Our purpose here is not to cast doubt on the sincerity of the commitment of ethnic militants, whether outsiders or natives (indeed the violence of certain situations suggests quite the opposite), but to distance ourselves from individual reasons and passionate involvement in shared causes in order to examine the mechanisms that are then created. Firstly, it should be noted that the intervention of outside actors in the affairs of local populations is nothing new. More or less well-known scholarly or religious luminaries have over many years been their allies and lent them their support, defending their interests against the state, relaying their demands in the media, and echoing their tribulations. However, it would seem that, over time, some actors have begun to see their role not merely as one of providing logistical or citizen support, but also as being the inspiration for new forms of organization and the proponents of new ideas defining the outlines of an ethno-political space.

The notion of “institutional hosts” put forward by Peter P. Houtzager [2001: 3] in the context of his study of the modernization of agriculture in Brazil, in which he examines the effects on the rural union movement of what he terms intra-elite conflict [ibid.: 1] during the military dictatorship (1964-1985),^{Footnote 28} will perhaps allow us to better comprehend the ongoing changes. This political-science specialist defines the notion in the following terms:

The context of the 1990s, and even more so that of the 2000s, is patently not that of the 1970–80s: when ethnic demands first began to be voiced, the hope of radical change predominated, fanned by the 1988 constitution, broad mobilizations (a phenomenon termed “social movements”) and, upon Lula’s accession to the presidency in 2003, the implementation of public policies resolutely in favour of the most deprived sectors of society [the bolsa família]. During this period, the majority of Brazilian society seemed to back the government’s intention to put in place a vast project of redistributive reforms. Beyond these obvious differences, certain parallels can be drawn. One case in point is the creation of two mirror-image versions of the same project, at the time less in competition with each other—as during the military dictatorship—than working in complement with each other: one version, in the sphere of the state, involved the creation of public administrations dedicated to recognized ethno-legal categories, and the other, in the sphere of civil society, was concerned with incentives for the creation of new collective actors, mobilized to call for the application of constitutional rights.

The desire of ethnic militants to reach out to hitherto marginalized populations has contributed to this standardization of protest, by facilitating convergence between the expression of social struggles and the institutional framework, and thereby providing such populations with the means of acquiring visibility to public authorities.

With this in mind, I consider it possible to propose that the older alliances have indeed been transformed into what Houtzager, in order to distinguish them from “political linkages,” refers to as “structural linkages” [ibid.: 3], that is, institutional arrangements that drew “new issues into the public sphere […], created new collective interests and bases for alternative collective identities [… and] provided physical points of access to the state, facilitating petitioning, protesting, and negotiating” [ibid.: 4]. This movement is all the stronger as, besides the federal state, various states and even certain municipalities have espoused the ethnic watchwords, created corresponding official secretariats, and provided them with modest grants and/or premises. Furthermore, the gradual emergence of a class of paid professionals, some directly employed by state administrations (anthropologists entrusted by the latter with the assessment [laudo] of populations, for instance), others by non-governmental organizations wishing to support ethnic struggles, has also contributed to the institutionalization of political arenas.

What we then realize is that, for “collective identities” to be seen and heard, their formulation needs to build on the channels opened up by these mediators between state bodies and local populations. However, it is worth noting that certain militants—few in number, it should be pointed out—admit to being in two minds: while recognizing that this new configuration gives greater resonance to their cause, they nevertheless fear that their protest is at risk of being co-opted by the authorities. Indeed, ethnic actors from outside the groups concerned find themselves at the intersection of two strained expectations: that the voice of dominated populations should be relayed to the authorities, and that the presence of the state should be reinforced by the action of its public administrations. In answer to this offer of mediation, local populations are prepared to explore the new ways of extricating themselves from the very difficult situations that these connoisseurs of some, at least, of the unfathomable workings of the state,^{Footnote 29} seek to show them. If these new approaches involve the creation of an ethnic association, in replacement of an “inhabitants” [moradores] association—a structure dating from the 1990s that made it possible to obtain agricultural credit lines—or a “community,” populations are likely to turn towards the opportunities provided by affirmative action policies in favour of Indians or quilombolas as they allow access to citizenship based on an extension of social rather than political rights.

While the establishment of ethnic associations is directly linked to the action of a set of external actors and to changes in the legal framework, their creation should, however, not be interpreted purely and simply through the prism of the top-down imposition of a normative order. Local populations proclaim their ability to act according to conscience and forcefully insist that they consider their ethnic-based demands as the exercise of a “legal right” enshrined in the 1988 constitution. Which is why a multiplication of possibilities of interlocution is a pressing demand among villagers preoccupied with expanding their pool of information sources in order to be, in a historically situated context, in a position to make the best possible “choice.”