Clothes for spirits

This work is licensed under the Creative Commons | © Diana Espírito Santo. ISSN 2049-1115 (Online). DOI: http://dx.doi.org/10.14318/hau6.3.010

Clothes for spirits

Opening and closing the cosmos in Brazilian Umbanda

Diana ESPÍRITO SANTO, Pontificia Universidad Católica de Chile

In this article I argue for alternative models to understanding the relationship of spirit cosmologies to their social surround. In my ethnography, I show that fluidity and plasticity are central to a particular urban sector of the Brazilian spirit religion of Umbanda. Some contemporary Umbandists, particularly in the city of São Paulo, see that “culture,” read as the Brazilian historical imaginary, becomes the stuff by which the spirits self-reflexively “clothe” themselves. However, using Don Handelman’s division between cosmoses held together “from the inside” or “from the outside,” I also argue that Umbanda practitioners take varying degrees of openness and closure of their cosmology, including the urban sector I focus on, whose plastic cosmos is essentially ruptured, and thus needs to be “believed” in.

Keywords: Umbanda, spirit cosmos, possession, plasticity, identities, perspectivism, rupture

In an article on Melanesian anthropology, Ron Brunton argues against Peter Berger’s assertion that man has a universal propensity for order (Berger cited in Brunton 1980: 113). Order is often more a product of the institutional and intellectual drives of the researcher, he says, than of the culture he studies (Brunton 1980: 113). This is a critique by and large now familiar in anthropology (see, for example, Barth 1987). But Brunton suggests more attention should be paid to fluidity, incoherence, inconsistency, and idiosyncrasy. This may be even more imperative in contexts where religious experience is born from just these idiosyncrasies, uncertainties, and immanent transits of deities, such as in the African-inspired traditions of Cuba and Brazil.

However, that scholars have tended to come to terms with these ontological fluidities through recourse to their psychological, sociological, or political counterparts is arguably telling of the ultimate frames of such analyses, which assume that [86]cosmological dissonances can be understood with reference to other, more “real” aspects of political, social, and cognitive life. Thus, for Brunton, “cognitive consistency may only be necessary in contexts where inconsistency leads to political vulnerability” (1980: 122, my emphasis). In his fascinating ethnography of Kalahari Bushman religion, Mathias Guenther argues that a whole series of social-structural factors explain the ambiguous and highly variable quality of people’s belief complex, in which Tricksters figure prominently (1999). Guenther describes the first of two orders of Kalahari existence (the first being the “primal,” mythological order, and the second being the “present,” historical order) as peopled by entities whose capacity for mutation is absolute, such as the “Trickster,” who disguises himself in this or another skin, with one or another face, at once animal, human, and spirit (1999: 66–70). Indeed, Guenther describes these orders not as separate or sequential; the primal, mythological time pervades the historical present (1999: 66).

This scenario would not be unfamiliar terrain to anthropologists of Amerindian worldviews, who are accustomed to documenting the ontologically transformative and expansive capacities of shamanic technologies, nor indeed to scholars of Native American folklore, which is replete with transgressive Coyotes and other tricky creatures (Radin [1956] 1973; Hyde 2010), and so on. The question that arises is how to theorize these aspects of ontological mutability in themselves. This implies understanding them not as epiphenomena of other social and political forms of uncertainty, but as alternate (and alter) cosmological orderings. For Guenther, the malleability of the Bushman’s cosmology is a function of an equally malleable social matrix, where uncertain social classifications coupled with the absence of a defining political system yield broader metaphysical manifestations of uncertainty. Both a strong individualist ethos and the absence of ritual specialists to render complex beliefs systematic, result in interpersonal variation, diversity, and change (Guenther 1999: 81–83). Ambiguity, he suggests, or chaos, is a form of living, and thus also a form of thought.

But while this functionalist tack is food for thought (functionalist in the sense that an understanding and experience of cosmology is predicated entirely on social structure), the presence and persistence in otherwise non-“chaotic” cultural universes of immanently transgressive entities such as the African or Afro-Caribbean inspired Trickster reminds us that what is at stake is not always a lack of cultural consensus or even broader social uncertainty—nor even the absence of an expert orthodoxy—but the existence of cosmologies, or facets of the cosmos, that are uncontained by the categories or tropes reserved for them. This means placing the idea that spirit cosmological order is a precondition of religious experience under scrutiny: first, perhaps, by paying closer attention to the transformational logics underscoring its specific mutations, fluidities, and their clearly also social effects; and second, by not assuming that people experience spirit cosmology in only one way.

In the ethnographic case I will describe, Trickster also tells us that there is more trickery than meets the eye; we should pay closer attention to instances of ontological malleability in a broader—and not simply exceptional—sense. Our anthropological obsession with finding exceptions to order goes to the heart of the limitations of our representationalist thinking—where, as Karen Barad says, we assume sharp edges between entities in the world, and relations of absolute exteriority between them, as also between knower and known (2003). This is particularly the case, I will [87]argue, with spirit possession practices, whose scholars must deal not simply with the social life of invisibles but also invariably with complicated embodied histories, mimesis, concepts of persons, things, and material possessions, that are assumed in some form to be shaped by their historical surround. In his recent volume on genealogies of spirit possession languages, Paul Johnson argues that “spirit possession is a theatre of potential pasts, selectively activated by ritual performers for purposes in the present” (2014: 15). But that it is a far from straightforward theater is emphasized by Michael Lambek (2014) in his afterword to the same book and elsewhere: possession may be ambiguous, just as discourses surrounding it ironic. Further, it is not just people who animate or empower spirits but also the other way around— “power goes both ways,” precluding any simple binary distinction, for instance, between master and slave, defeat, conquest, or resistance (Lambek 2014: 263). In this essay I will think through my ethnography of a Brazilian religion called Umbanda and imagine other conceptual languages with which to understand a moving spirit universe and its relationship to people, one that does not reduce a spirit repertoire in a direct form to a cultural and historical “imaginary,” but introduces more creative means of grasping the relationship between the two.

Perhaps no other spirit mediumship tradition in Brazil has evoked so forceful and direct a relation between cosmology and the particulars of the society in which it exists than Umbanda, described by its early scholars as the first truly “Brazilian” religion (Bastide [1960] 2007; Ortiz 1978). The most common origin myth among Umbandists is that of seventeen-year-old Zélio de Moraes, who, upon involuntarily incorporating an indigenous “Caboclo” spirit in a white Kardecist Spiritism session in Rio de Janeiro in 1908, much to the displeasure of the mediums present, revealed the birth of a new religion that would embrace spirits of all “races” and degrees of “evolution.” Against the backdrop of what was an elitist white Spiritist community on the one hand (Spiritism is a nineteenth-century “scientific” religion designed by Frenchman Allan Kardec), and the largely tight-knit, African-inspired religious traditions on the other, Umbanda appeared conciliatory at least and revolutionary at best. While scholarly consensus would ultimately place Umbanda’s formalization in the first decades of the twentieth century as a middle-class attempt to recover the magical efficacy of the Afro-Brazilian religious universe while largely erasing its ties with “a backward and uneducated Africa” (Capone 2010: 74–75; Brown [1986] 1994), anthropologists have repeatedly sought recourse to a vision of Umbanda as a repository of a national symbolic imagination, reiterating a native view of Umbanda as quintessentially “Brazilian” (if not always democratic).

Certainly, the identity make-up of its spiritual denizens seems to point to an inclusivity of social stereotypes. Many of those entities in wider circulation among Umbandists today—among which are the Pretos Velhos (Old Black Slaves), Caboclos (Indians, native Amazonians), Crianças (child spirits), Ciganos (gypsies), Pombas Giras (women of “ill repute,” prostitutes, courtesans), Bahianos (natives of Bahia, migrants from the north), Boiadeiros (cowboys of the arid Brazilian backlands), Marinheiros (sailors), and others—reflect a porosity so evident in spirit mediumship traditions elsewhere, particularly those where strong historical and cosmological entanglements obtain.

However, in this article I propose to explore the overlooked dimensions of what we could call ontological “plasticity,” which in the first instance remits us to the [88]domain of the Tricksters of the Afro-Brazilian cosmos, the Exus. Exus are especially relevant here because they tend to be the darlings of an anthropology that sees an immediate and unproblematic correlation between a certain class of entities in a popular ecstatic cult and a widely shared imaginary on the persistent problems and contradictions of Brazilian social life and its underbelly—poverty, inequality, sexism, exploitation, and corruption (cf. Prandi 1994), even if this correlation is fraught with ambiguity, irony, and inversion. But Exu’s transgressiveness must be understood not as a departure from order (or its inversion), which would reify this very order, but as indicative of this order’s very logic, a logic of plasticity rather than transgression. I take my understanding of plasticity from Catherine Malabou, for whom it means being transformed without being destroyed. For her “plasticity” is that which “renders possible the appearance or formation of alterity where the other is absent. Plasticity is the form of alterity without transcendence” (2010: 66); it is, thus, an internal property of that which is transformed. One of the overarching points here is that we can and should experiment with conceptual alternatives to causal understandings of spirit identities and their changes, and that as a concept, plasticity allows us to understand this topological mutability of forms without necessarily positing an external, reductive frame of reference.

The argument in the first section is that ontological plasticity, as it is articulated by an elite group of contemporary practitioners in urban [89]São Paulo, has a long history stretching back to the very inception of Umbanda’s orthodoxy in the 1930s and 1940s. These first Umbanda intellectuals posited spirits as anonymous beings of light that could “camouflage” themselves with different appearances or “clothes” generated by Umbanda’s main intent: the embracing of racial and ethnic diversity. However, their intention was also to distance Umbanda from its “African” “uncivilized” origins. Contemporary intellectuals, particularly in São Paulo, posit a similar cosmography of metamorphosis but frame it in terms of symbolic “archetypes”—Jungian-type imagery seen to be sustained by a collective consciousness (and unconscious). Both dissolve the opposition between cultural construction and ontology by proposing that the latter (spirits) use the former (cultural imagery, archetypes) to constitute (or clothe) themselves. In contemporary São Paulo, this idea is particularly vibrant. Practitioners of Umbanda Sagrada (Sacred Umbanda), an intellectualist movement that began in the 1990s, in effect conceive that spirits “wear” culture, so to speak, a conceptualization betrayed by language that separates spiritual form (that which is worn) from essence or content (the wearer), a displacement that both has roots in early Umbandist writings and innovates from them.

But as Maya Mayblin argues (2014), there is an ultimate unknowability of divine forms that characterizes the movement between distance and proximity that people experience in relation to Brazilian deities. In an article on popular Catholicism in the northeast of Brazil, Mayblin argues that people understand saints and other sacred figures ambiguously—both like them (as flesh and blood, gendered) and not like them (ungendered, divine, not-human), and they shift between these poles depending on context and discourse. We could argue that this may be a sacred dynamic applicable to some of Brazil’s other cosmologies, seen, for instance, in the context of the contemporary Umbanda movements described in the first part. In the second section, then, using my ethnography from both Rio de Janeiro and São Paulo, I will show that practitioners experience varying degrees of “openness” and “closure” of their cosmology. This in turn either engenders conditions for an identification of spirits as ex-people, with particular stories and deaths, creating proximity, or it creates an ontological distance between person and spirit that swallows up their archetypical characteristics as biographically arbitrary and recycles them as moral “perspectives.”

I take my understanding of the differential ways of viewing cosmology from Don Handelman (2008), for whom some cosmoses are bounded, meeting their own limits, and held together from without themselves, and some are not (2008: 182). A good example of the former is a monotheistic cosmos where a barrier separating God and human beings creates an absolute difference that necessarily requires “belief ” to hold it together (2008: 184). Belief comes to the fore, according to Handelman, when the cosmos is constituted on a rupture, or is ruptured; it cannot know itself except through outside itself, which is why belief in a transcendent is imperative. A nonbounded cosmos, on the other hand, holds itself together “from the inside,” “intra-grating” through itself (Handelman 2008, 2014), and belief holds little importance. In this cosmos innovation is commonplace. Much like the Cuban spirit cosmos that I also study (Espirito Santo 2015), much of Umbanda’s cosmology is continuous with itself, with few abrupt shifts or ruptures between people and other-than-people. Handelman’s preoccupation, which is also mine here, is to show how “worlds are holding together through the metaphysics of the human, through the imaginaries of the human” (2014: 96). He refers to “cosmos” as the “entirety of the phenomenal lived-space of all entities—human and other-than-human—the entirety of a world of all dimensions of existence” (2014: 96), which is how I will take it.

But I also argue that in certain Umbanda contexts, this cosmos of spirits is essentially recursive—the spirits are aware of themselves as spirits and able to effect categorical changes on their own appearances. While this characteristic apparently places Umbanda in the “fuzzy” continuous-with-itself cosmos category, I argue that in the Umbanda sectors I mostly look at in this article, there is a move to transcendentalize Umbanda’s entities and thus to “close” the cosmos, making it impermeable to people. There appears to be an oscillation between different degrees of “encompassment” in an understanding of spirits, depending on practice context, and indeed, it is the practice context itself (the phenomenology of trance) that allows the cosmos to “open” again. My contention is that spirit cosmoses should not merely be regarded as a refraction of wider, more tangible processes but in their own terms as much as possible, and as such, irreducible in an immediate sense to sociological facts. Rather, the point here is the opposite: to acknowledge how Umbanda generates ontological possibilities that afford their own disentanglement from just such modes of social and racial determinism.

Trickster shapeshifting, now and then

Roger Bastide observed early on how Umbanda’s cosmology—based on seven lines of spirits headed by the African gods cultivated in Brazil, the orixás, but subdivided recursively into many others legions and sublegions bridging numerous identities [90]and ethnicities—is constituted by an “extraordinary syncretism” and system of correspondences ([1960] 2007: 324). Umbanda linked Spiritism with Candomblé, the popular West African–inspired religion of the orixá gods by placing the souls and spirits at the latter’s disposal, for orixás were “higher” types of beings on a “moral” scale (even though some Umbanda temples also call their spirit entities orixás); and it resolved the problem of syncretism with Catholicism by appropriating a popular, even magical Christianity, sublimating the former in the process (Bastide [1960 2007: 327–28). Umbanda has appeared, in this sense, as a both provocative and mollifying modern religious form. In a religious context where the benchmarks of “tradition” have been the Candomblé temples of Bahia on the one hand (Gois Dantas 1988), and European-derived Spiritism on the other, Umbanda has tended to be regarded as a “bureaucratic compromise” (Brown [1986] 1994), so predicated on an inherently syncretic constitution that early scholars alternately regarded it as a degraded form of Afro-Brazilian religion, and a “low Spiritism” (Giumbelli 2003; Maggie 1992).

But Umbandists have long found ways to resist its definition as a religion of mixtures, lacking in idiosyncratic roots or self-identity. For instance, in my fieldwork in Rio de Janeiro in 2010, I met the leaders of an Umbanda federation in existence since the 1940s who categorically refused what they called the “conspiracy theory” view of Umbanda’s history, promulgated by Diana Brown among others. “Umbanda is not a middle-class phenomenon designed to whitefy Candomblé,” they told me, annoyed at the thought, or “a white version of African religions, as some anthropologists would have it. It is wrong to see Umbanda as a mixture of pre-existing religions.” (For a detailed reading of how anthropologists politically and socially create the field of “Afro” religions, see Palmié 2013; Capone 2010; Gois Dantas 1988.) Rather, they place Umbanda’s beginnings among the “red people,” the Indians of Brazil who possessed sacred knowledge (but no more): the point of Umbanda is to revive it. In the early days of Umbanda’s codification, intellectuals backed by Umbanda’s first federations during the Estado Novo period de-historicized Umbanda—positing its genesis in distant lands, from India to Atlantis (Ortiz 1978). This occurred in the context of the continuous persecution of the religions of “African” inspiration in Brazil, which these leaders (partly successfully) hoped to extricate themselves from. Recent innovations are arguably more sophisticated.

As an ethnographic example of this contemporary state of affairs, I forward a brief description of an Umbandist temple I observed on several occasions in 2011, whose philosophies trascendentalize spirit identities in a slightly different way to the Umbanda federation mentioned above. Just before the end of an unpaved road in a wooded, lush corner of Jacarépaguá in Rio de Janeiro is Templo Estrela Guia, headed by a husband and wife team in their mid-30s, and whose Exu sessions made a lasting impression on me. Estrela Guia’s pai-de-santo (father-of-the-saint, priest), a large white man called César, would oscillate during any one evening in his “incorporation” of his two main Exus: Lalu and Meia Noite (Midnight).

On one of these evenings when he did come, Seu Meia Noite, as he was fondly referred to, released a loud, wide cackle that resonated through the bodies of all the audience members. One of his ritual children swiftly draped him with a black, silky cape with a purple interior and a cosmogram of two pitchforks sewn in green on the exterior. A necklace made of thick, shiny, black beads was placed around his [91]neck, a pendant hanging in the shape of a skull. Meia Noite had stripped off César’s black t-shirt, leaving his round belly showing. He would strut around the terreiro for the next couple of hours, torso naked, and caped. “Good evening,” he said slyly to us. “Is that song meant to not wake up the children?” he chuckled ironically. Then he strutted outside with candles in hand, and placed a lit one outside each of the spirit “houses,” ending at the Exu house, which looked much like a cave, inside which were statues of Zé Pilintra and Maria Padilha, two iconic Exu spirits. Exus are also known as “people of the street” (o povo da rua), a category bridging prostitutes, pimps, hustlers, conmen, tricksters, and otherwise street-savvy personas characteristic of the 1920s and 1930s in Rio, and are widely understood to form one of Umbanda’s metaphysical cornerstones.

Save the people of the crossroads!” said Meia Noite upon returning. People began to play songs more intensely now on the drums. Meia Noite paced around the terreiro slowly but dramatically, whiskey bottle in hand, cigar in mouth. He had perfumed himself generously from a bottle he had received as a gift. We all clapped and sang. The tunes were catchy, the lyrics simple and repetitive. Meia Noite stood at the center of the terreiro and gazed at all of us. He then said:

Terreiros are just the stage, the theater where things happen momentarily. But when the entities leave the heads of their mediums, and the mediums leave the stage, the entities keep on working. We work all the time, at every hour. When you get here, we’re already working. We don’t get tired, people, tiredness is of the dense, material world. . . . I travel with the speed of thought. I can be here now, with you. But I can be incorporated in 10, 20 mediums at the same time, in different places.

Later in the same evening, after a switching of guard, so to speak, Exu Lalu had piqued my interest by somewhat sarcastically observing, at several moments of the gira (the possession rite, literally meaning “to turn”), how misunderstood Exus were in the public’s imagination. “People think we’re the devil, but we work against devils. We are given access to the dark zones so we have to look like we belong there.” Lalu’s statement points to something other than that implied in debates on what Exus are, where popular representations portray them as malandros, scoundrels, and untrustworthy types, and sometimes, accomplices of the devil (cf. Hale 2009; Negrão 1996). His allusion to camouflage, mimesis, and metamorphosis— couched in moral terms (you need to look like the devil to fight him)—pointed instead to the importance of what Exus do. Indeed, at another point in the evening he said, chuckling, “There are many Exus that people think are Caboclos,” leveling his criticism at another temple leader who was attending the ceremony; “I am Lalu, but as I explained to you, I come as Caboclo, I come as you wish, Preto Velho, Criança. Exu gives form, spirituality.” This concern with the inherent arbitrariness of spirit appearances is prevalent among the more scholarly urban Umbandists, who have taken Trickster’s message close to heart.

From the 1990s onward there has been a theological “turn” among educated (mostly white) elites in urban Umbanda, especially in the city of São Paulo, indexed by a proliferation of Umbanda theology courses and literature (among other ways), some of which has provided an implicit critique of Umbanda’s traditional cosmological models. In part, these have done so by forwarding a universe of spirits [92]whose identities and appearances dialogue reflexively with a national archetypal repertoire, rather than merely reflect it. For some members of a new generation of urban Umbandists, Umbanda’s cosmos reveals itself as aware of itself and its capacity for categorical and thus symbolic shifts in both perspective and form. Far from just “bits” of the Brazilian past harnessed into the present through forms of mediumship, spirits are thought to borrow their appearances from a human, historical, cultural repertoire. So, for instance, this group of Umbandists would argue that it is precisely because the Brazilian population has an image of the slave as a historically passive, humble, subservient type that certain spirits come “as” Pretos Velhos. This Umbanda category of spirits manifests as elderly, bent-over, needing a cane, and exceedingly good for their insight into domestic life. They are referred to affectionately in kinship terms: auntie, uncle, grandmother, grandfather. But according to some Umbandists, this may be only their disguise, modeled on human concepts; behind these benign, simple appearances are spirits of exceeding vitality. Thus, the spirits articulate a version of Umbanda that is highly conscious of its own categories and their culturally (humanly) constituted nature.

One group of people for whom this is true is followers of São Paulo’s Umbanda Sagrada (Sacred Umbanda) movement. Among other things, this movement has reappropriated the Brazilian popular imaginaries of identity in order to regenerate Umbanda’s cosmology in sophisticated ways. In their view, it is not the spirits of the (Brazilian, national) dead that wield symbols, but symbols that wield the (often anonymous) spirits. In proposing this, Umbanda Sagrada subverts dominant representations of Umbanda as the recipient of “lesser,” more “popular” spirits (in comparison, say, to Candomblé, cf. Capone 2010), as also, perhaps unwittingly, anthropological renditions of Umbanda as a nostalgic reproducer of some distant historical or romantic imagination. As a collection of theological precepts, Umbanda Sagrada was “received” by inspiration from the spirits (specifically, from a Preto Velho spirit called Pai Benedito) through the middle-aged medium Rubens Saraceni, its charismatic leader and promoter before his death in 2015. Fighting against what he saw was an insular, elitist culture of secrecy and mystification in Umbanda, Saraceni opened Brazil’s first public Umbanda “college” in 1999. He also began to publish many volumes of books of complex theology on the genesis and spiritual organization of the universe (Saraceni 2013), which have arguably triggered a renewed interest among the young in what was a declining religion in the 1980s and 1990s. The core of this success is his reformulation of Umbanda’s understanding of the “symbol” and its dividends. Saraceni has written almost fifty books, which have sold over one million copies.

Some people regard Umbanda’s spirits literally—as spirits or souls of people who have lived, namely, as a slave, or an indigenous person, or a child, or a prostitute—and who now labor through mediums, whether for their own moral “evolution” or that of others. Umbanda Sagrada, in contrast, regards spirits as manifesting an aspect or quality of God, called a “mystery,” and as such, not as intermediaries between a higher entity and a human world but an end, or message, in themselves. According to Saraceni, “Caboclo” is a degree of being, rather than an identity, one that refracts itself as qualities of the same degree in the different lines; for example, Caboclo of Oxossi is a guide that shapes itself as a Caboclo with the quality of the orixá-saint Oxossi. Their shapes, Saraceni maintains, are given by ideas sustained in [93]a kind of national collective consciousness (or indeed unconscious), the archetypes: patterns of thought and behavior, imagery conforming to ideas of distinct ethnic and professional identities in Brazil. Saraceni calls these forms, formas plasmáticas, which would roughly translate as “expressed forms.”

Umbanda Sagrada’s young theologians and leaders understand that spirits have the ability to shape or mold themselves according to particular imagery in Brazilian imaginary; that they do so through metaphysical figures that speak to the fundamental values of humility and simplicity is seen as evidence not of Umbanda’s inferiority vis-à-vis an Afro-Brazilian religious “continuum” (Camargo 1961), or of its mediatory position between moral and religious poles (Negrão 1996), but of its ontologically sophisticated design as the religion of the people, accessible in form and language. This new theorization has as one of its corollaries the de-coupling of cosmos from nation, by way of posing the former as internally and generatively conscious of the latter. Saraceni writes the following about this relationship:

The spirit, the deity, even the saint has shown herself in a particular way. A specific way of dressing, a color, a shape, in spirit. These “clothes” are symbolic. It is all a process of showing the living the power to whom he must direct his thought or prayer. . . . All divinities do this. The living must have some kind of reference. How would I be able to focus on Jesus if I don’t have a physical, material reference in mind for what he looks like? In this case, it’s a human form. But culturally this may vary. If the spirits were to present themselves in Brazil as dragons, as some do in China, people would be scared. They would run, not interact. We have protective spirits that take shape and description according to the religion. And according to the country and its archetypes. . . . If there is no image inside the mind, there is no reference point for that shape. (Saraceni 2013: 17–18)

Since 1999, Saraceni’s Colégio de Umbanda has taught leaders and followers from hundreds of Umbanda temples in the São Paulo area. Alexandre Cumino, a young Umbandist theologian and teacher, is one. For him, Umbanda’s “archetypes” result from the strong prejudices of Brazil.

This prejudice in relation to race, sexuality, freedom, to women’s power. . . that’s why Umbanda has archetypes. None of the spirits need to make an elaborate speech to teach you to love the Indio, to love the black man, to love the child. But when one of them says, I am a Preto Velho, I am an Exu—you will never know who he is. What matters is what he wants to convey. (Interview with Cumino, October 2013, São Paulo)

The spirits’ individual identities become a language of manifestation rather than its precondition, a language determined by but not reducible to historical contingencies: a language that can change constantly. More important here is the idea that the spirits themselves become anonymized, separated by an ontological gulf from their human counterparts who will never see or experience them as persons, or rather, ex-persons-now-spirits. In this separation, according to Cumino and others, lies the effectiveness of their moral message—freedom, tolerance, peace, and so on. It is only by virtue of the first that the second is conveyed.

These Trickster ideas are not new, however; they have a long history in Umbanda, tracing back, in fact, to its very inception.[94]

Some of Umbanda’s most important and influential early twentieth-century texts also reveal foundational concerns with what I call “ontological plasticity”— the property of a cosmos (in this case of spirits) to change, re-form, and recreate itself, in this case, in relation to a social, human reality, to which it also gives shape. At stake, I would argue, is not just the metamorphosis or differential identity politics of spirits for the sake of efficacy or deceit but a self-reflexive spirit perspective that articulates a purposeful and ontologically impactful relationship to human “culture.”

One of the first Umbandist authors and intellectuals to tell his audiences of “trickster” spirits was Antônio Eliezer Leal de Souza, who warned not just of false or charlatan mediums but of imposter spirits, too (1933: 28). Like many others of his time, Leal de Souza saw Exus as malefic, irritable creatures that exercise their whims in centers that practice black magic (1933: 38), and that forge cosmic battles with their opposites: the White Lines of Umbanda. More interesting, however, is his characterization of the “protectors” of the White Lines: the Caboclos and Pretos Velhos. Leal de Souza defends the thesis, based on Kardecist Spiritist concepts of metempsychosis, that:

Between the Caboclos, many were Europeans in previous incarnations, and their reincarnation in the heart of the forest does not mean a regress, but the beginning, through their identification with their environment, of a mission that as spirits, after a learning period in space, they would have to carry out on earth. Others belonged, in their last existences on earth, to white or Western peoples, or to the yellow Asian peoples, and never came through our tribes. (Leal de Souza 1933: 58, my translation)

While clearly subject to the racial prejudice of his time, Leal de Souza nevertheless gives us what is perhaps the first glimpse of what we can call the disjuncture between spirit’s “real” identities and their appearances, namely, by way of an appeal to their past lives. “The protector, in the White Line, is always humble, and, in his incorrect or clumsy-sounding language, he causes a sorry impression of ignorance, but frequently, in practicing the duties of his mission, surprises the consultants with his high level of knowledge” (Leal de Souza 1933: 58, my translation).

For Leal de Souza, this “camouflage” is part of Umbanda’s message in itself. He notes, for instance, how the spirits often come speaking in a tongue-rolled way, apparently disfiguring the language. This is not because they cannot do better, Leal de Souza suggests, but precisely because they wish to give the impression that they are backward, namely, so that the “individuals who think of themselves as superior and are forced to appeal to the humility of inferior spirits,” or so they think, “perceive and understand their own inferiority” (Leal de Souza 1933: 46, my translation, emphasis added; cf. also Giumbelli 2013). He tells of an occurrence he witnessed once whereby a Preto Velho, while treating a woman with his pipe smoke, candidly told her that she had a heart problem that should be attended to. Upon being examined by a doctor that was present, the Preto Velho proceeded to explain the medical problem in precise, technical terms, surprising the doctor, who disbelieved that the spirit could really be “African” or black. The Preto Velho then explained that indeed he was “black,” but that he did not “come down” in “roda de doutores” (Spiritism sessions with doctors), lest he become as arrogant as them (Leal de Souza 1933: 59, [95]my translation). With this story, Leal de Souza places his loyalty firmly within Umbanda and its exaltation of the principle of humility and anonymous charity, articulating it with his conception of the spirits’ forms and identities. This is done both with Spiritism, through an appropriation of the notion of multiple lives and existences, and against it, through a denunciation of its arrogance vis-à-vis the heterogeneous spirit world that is not always what it seems. Indeed, Leal de Souza denounces not just the arrogance of assuming a Preto Velho knows little of medicine but also of taking the spirits and their appearances at face-value.

It is tempting to see in the stances explored in this section an example of what Diana Brown ([1986] 1994) has identified as a primary drive in the movements of early Umbandists—an appeal to the expansion of Spiritism to its “outer” unacceptable edges but by way of the redemption of the “other’s” spirits and rites. In what would then become the best of both worlds, the Pretos Velhos, Caboclos, and others, could be still seen in this light as “evolved,” perhaps even white, spirits, under the guise of symbolic spirit “costumes.” Thus, spirits divested of their “primitivity.” This was probably the case for the older writers, such as Leal de Souza and W. W. Matta da Silva, among others. Furthermore, the concept of a spirit “body” or “form” subject to changes of appearance or to shifts or unfoldings of qualities is not one specific to Umbanda proper in its milieu, but inherits from and expands Kardecist Spiritism’s notions of the “perispirit,” or “astral body.” In his dualistic cosmology of the person, the perispirit is conceived of by Kardec as a subtle mediatory principle or body, a semimaterial substance linking the perceptions, sensations, and experiences of the time-framed physical body to the recesses of the immortal spirit or soul (cf. Cavalcanti [1983] 2008: 28–29). In the Book of spirits, Kardec’s spirit informant states, “the spirit changes its envelope as you change a garment” (Kardec 1857: book 2, chapter 1, question 94); the perispirit “can assume any form that the spirit may choose to give to it” (Kardec 1857: book 2, chapter 1, question 95).

There is a sense in which we could ask the same of newer versions of this argument, such as those wielded by Umbanda Sagrada, but I have no evidence ethnographically or historically to sustain that it is a will to “cleanse” the “African” aspect proper. However, whatever their intentions for doing so, both camps de-essentialize identity-driven understandings of spirit forms, symbols, functions, and meanings, namely, by introducing what appears to be a fundamental principle of ontological plasticity in a reading of the cosmos. As I will explain below, however, this plasticity ends up divorcing this same cosmos from the immediate grasp of its mediums.

Fracturing the cosmos

One of the easiest levels of analysis for the material presented in the last section is a perspectivist approach. For instance, in North Asian contexts, according to Morten Pedersen, there is “a potential interior spiritual quality in things” (2001: 414), whether natural “things” or animals or even spirits. In an article on Mongolian and Siberian animistic ontologies, Pedersen and Rane Willerslev ask what exactly is meant by soul (2012: 466). The animist soul, they say, cannot be treated as fixed referent but as “an inherently relative or deictic phenomenon, whose form depends on who perceives it and from where” (Pedersen and Willerslev 2012: 467).[96] For the Siberian Chukchi—contrasting with the Judeo-Christian conception of the physical body as substantially different to the immaterial soul—there is a sense in which souls are bodies. They can eat, drink, and be hunted by other creatures. These authors argue that a perspectivist ontology is better applicable in their ethnographic context than one that separates soul from body. The reader may know that perspectivism is a term-concept-theory employed by some Amerindianists, and most notably developed by Eduardo Viveiros de Castro to describe a cosmos “inhabited by different sorts of subjects of persons, human or nonhuman, who apprehend reality from distinct points of view (1998: 469), subject to a common “soul” and distinct “bodies” (thus, multinaturalism). Through the work of Aparecida Vilaça, Pedersen and Willerslev argue that the Amerindian “soul,” conceived as a quality that identifies it with all other beings, is characterized by its capacity to take on new bodily appearances (Pedersen and Willerslev 2012: 471). According to them, “the soul is not simply one body that ego cannot presently see . . . ; it is all the potential bodies that ego might take at some future point” (2012: 472). When we apply this concept to their North Asian context, this means that the soul of the soul is another body, or all the other bodies that a body could be.

Among the Umbandists I have spoken about in this section, the Umbanda “entities” are also bodies, of sorts, relaying a moral perspective on the world of humans through a message or archetype specific to that “skin.” The “soul” beneath or under these bodies may take on a multiplicity of perspectives and shift, perhaps at will. For these Umbandists, “culture” (social and psychological history) constitutes the materials by which the spirits fabricate their “bodies” or “clothes.” This means that “culture”—articulated by my interlocutors as a substance that exists and is sustained in a collective consciousness—is not understood representationally, epistemologically, as something produced and accumulated by representing minds; it must be seen in the light of its effects on a cosmos of beings. If we combine these insights with Marcio Goldman’s articulation of the concept of “virtualities” (2007, 2009), we can conceive that spirits can change, adapt, and multiply not just as a function of time and space but of their own internal metamorphic properties.

Everything in the universe of the Candomblé that Goldman studies is a diversification and a crystallization of axé, an all-encompassing force of nature (2007): people, deities, stones, and so on. In this “total” cosmos, multiplicity is an inherent property, although relations must be worked into existence, for example, through sacrifice and possession. The point of underlining Goldman’s ideas here is that virtuality, or the capacity to become, is forthcoming from the very definition of an entity in certain temples of Umbanda, that see it not as a god or a deity, or even a self-contained soul or spirit, but closer in kind to a vibration of sorts, one that has the ability to refract itself beyond a single form. The form itself, then, could be considered the spirit’s perspective, from a moral and archetypical point of view.

For instance, Bernardo, a young São Paulo–based Umbanda Sagrada medium, says the following:

You go to an Umbanda terreiro and you chat with Caboclo Pena Branca (White Feather). Do you know how many Pena Branca exist? Thousands! Everybody incorporates a Pena Branca. Who are they? We don’t know who they are. Do you know why? Because it doesn’t make the slightest difference who he is. He doesn’t come to bring his personality. . . . He brings a particular strength. (Interview with Bernardo, September 2013)[97]

But this is an understanding of cosmology that is characteristic of a relatively small, socio-economically elite, highly literate group of people throughout Brazil’s twentieth-century history. Indeed, perspectivism locks the Umbandist’s cosmological gaze in a very specific type of relation, one that is simply not always the case, as I will briefly show in the next section. Instead, we may need to experiment with a more plastic, pragmatic, and even individual understanding of cosmology, then, that does not see cosmology itself as a static, collective sort of representation, with each “cosmology” a whole unto itself relating or encompassing other aspects of a culture’s ethos.

Allen Abramson and Martin Holbraad have argued that classical understandings of cosmology were holistic in character, where ethnographers went “outside” their own culture, conceived of as a modern, complex, and relatively homogenous core, and “imagined themselves as charting the outer reaches of the social universe” where “this same whole curled inward in myriad places, organizationally similar (as worlds or totalities), but culturally differentiating and diverse” (2012: 37), and inevitably irrational, albeit not within themselves. Indeed, “their variously fanciful, less-than-true character was felt to be the abiding theoretical problem that they presented” (2012: 39). The point of this cosmos, they say, is that it is based on a uniform nature, described by science, subject to different epistemological viewpoints.

This idea that cosmologies are totalizing orderings of the world is rooted in the notion that societies themselves are naturally integrated (Abramson and Holbraad 2014: 6); thus, in a classical image, cosmology was imagined more or less as a reflection of other aspects of the socio-cultural order—kinship, politics, economy, and so forth. The anthropology of spirit possession is so exemplary of this ethnocentric holism that it is practically impossible for scholars to conceive of a spirit cosmology in its own terms, without recourse to other more “real” causes. Instead, “these approaches present indigenous cosmologies as inventive reactions to, or refractions of, more encompassing—and in that sense also more real—global processes” (Abramson and Holbraad 2012: 40). We are reminded of Charles Taylor’s argument on the historical emergence of the humanist “buffered self,” impermeable to outside forces, translated into a secular modernity that is closed in on itself, immanent, amenable only to differences in thought (2007: 35). Spirits are, then, essentially mental constructs, projected onto the often-unconscious body somehow, the differences of which can be located in various aspects of the environmental, cultural, and political landscape. As Erika Bourguignon—considered by many as the premier authority on trance, possession, and altered states of consciousness— has written, spirit possession “beliefs are present in large portions of the contemporary world and find expression in ways that closely tie in with various aspects of cultural and social change,” whether economic displacement, epidemics of disease, or anything else (2007: 374).

Morten Pedersen argues that the ambivalence with which certain authors have dealt with spirit possession, shamanism, witchcraft, and other occult phenomena in their respective ethnographies has led to a view of these as “symbolic language,” which in no way accounts for the Siberian shamanistic cosmology that he studies.[98]

These phenomena are typically analyzed “within a neo-Marxist model of social reproduction, where some cultural-ideational stuff is projected onto some political-economic stuff, or vice versa, in ways that are always in, or against, the interests of the powers that be” (2014: 176). In his ethnographic case, spirits were not caused by turmoil and uncertainty in Mongolia; they induced it. Pedersen asks whether it makes sense to speak of “non-representational cosmologies—of cosmologies that are not ordered reductions of the world but are instead specific enactments of its complexity” (2014: 178).

In a similar vein, what I am suggesting in this article is that we lend scope for a nonrepresentational understanding of spirit cosmologies that are not uniformly ordered or holistic, but inherently fractured and variable in their capacity to dialogue with their socio-cultural surround. The idea is to allow for a multiplicity of concepts to arise according to the specific ethnography at hand. This means being willing to treat spirits as relations, effects, and other phenomena that do not rely on notions of substance, for instance. I am reminded, for example, of Aisha Beliso-De Jesús’ excellent monograph on Santería transnationalism, where spirit entities and oricha-saints are conceived not substantially, but assembled and disassembled as electric impulses and currents, which travel through people but also televisual and electronic media, constituting copresences that transcend the normative boundaries of space-time (Beliso-De Jesús 2015).

Some of these languages locate spirits in the “interior” of people, where they can be understood as constituents of an expansive self in the process of becoming. For example, in an article on spirit possession in Mozambique, Morten Nielsen asks how “seemingly exterior entities might relate through a process of what could best be described as ‘interior swelling’” (2012: 435), thereby increasing people’s agentive potential from within themselves. As Alcinda Honwana (Honwana cited in Nielsen 2012: 438) explains, not only do people assume the personalities of spirits, the “spirits live and grow in people’s bodies,” for instance, as ancestors. This means, according to Nielsen, that a “dynamic interior assemblage of spirit and human is consequently established, infusing both with a social meaning that is projected outward as a reflection of the cosmological principles guiding social life in the local universe” (2012: 438). Among some of the consequences of the reconstitution of the acting agent as a “singular interiority expanding outward” (440), a sea of interior relations (with family spirits, for instance), are new assemblages on the outside, such as cleared land (441) to be claimed by means of these new “inner socialities” (Handelman 2002). In this particular context, spirit cosmology is understood to fold, swell, expand, travel outward from inner ontological components.

Other languages see a “curving ” of cosmology (and ritual). In an innovative article inspired by Don Handelman’s framework of “ritual in its own right” (2004), Matan Shapiro analyzes the ritual dynamics of two apparently very different religious spaces in Maranhão, Brazil—a context of oração (prayer) in an Evangelical church, and possession in an Afro-Brazilian temple (2016). According to Shapiro, Handelman “develops a unique gaze: rather than assume a priori a symbiotic relationship between the flow of everyday life (the ‘surround’) and the activities that take place during demarcated ritual events, he treats those events as distinct phenomena that hold together coherently and cohesively from within themselves” (2016: 48). Rituals self-organize to degrees linked to different levels of complexity[99]. Handelman conceives of rituals as literally “curving” or “folding” in the continuity of space and time (Shapiro 2016: 49); the more complex and self-organized, the more they “curve.” Shapiro says, paraphrasing Handelman (2004: 12–13), that “the ‘curve’ enfolds as suspension of regular social order, a distinct dynamic that participants feel on both sensorial and cognitive levels, and it is the experience of this ‘enclosure’ that makes the event temporarily autonomous from its surround” (2016: 49). According to Shapiro, the point of “curves,” for Handelman is that we can relinquish the idea that ritual events surround themselves with concentric fields of significance, only against which they are understood (Shapiro 2016: 49).

In the Afro-Brazilian Tambor de Mina, Shapiro discusses the spiritual biography of Carlos, who first experienced uncontrollable episodes of possession, then organized his orí (his head, the site of multiple spirits, once unruly), and then opened a temple. If at first possession was isolating, “curving” inward and suspending Carlos’ mental capacities, in a second instance, his entities began to interact and exchange with others in his surround, as Tambor de Mina entities are wont to. “Possession thus engendered new intimate linkages between Carlos and meaningful others, both humans and nonhumans” (Shapiro 2016: 54). This is contrasted substantially to the Evangelic rites also in Maranhão, which manifest a comparatively “flat” curve. The act of praying in this context—oração—itself entails a slow transformation of subjectivities (55), invoking an “ideological polyphony shrouded under the authority of a unitarily authoritative voice” (55). Consciousness awakes slowly; oração unfolds flatly (58). In Shapiro’s ethnography then, cosmology enfolds and outfolds, curves, twists and torques. By no means is its relationship to its social surrounding simple.

In the next and final section I wish to make a similar cosmological distinction from the point of view of Umbanda, also by way of Handelman. I take my argument from the article “Returning to cosmology—Thoughts on the positioning of belief ” (2008), which extends the argument made in 2004 on ritual to cosmology more broadly. I will argue that while most Umbanda can be modeled on Handelman’s open, organic, intraconnective cosmos approach, in cases such as the Umbanda Sagrada movement I describe above, there is a tendency to close off cosmos, creating breaches between persons and (unknowable) transcendent spirits that require a theological “belief-system” to hold them together from the “outside.” It is not God that encompasses the cosmos here (although He also inevitably does) but a whole array of higher-level entities whose designs for humans are largely unknown and whose appearances are extrinsic to them. However, this ontological gap is difficult to sustain in practice, and in possession sessions, mediums incorporate the same archetypical spirits as everyone else in Umbanda, manifesting them in their bodies and justifying this as “moral perspectives” in spirit bodies. The paradoxes that threaten to consume and unify this divided cosmos are thus exactly related to possession experiences.

Discussion: Opening and closing the cosmos

Handelman describes the “open-ended” cosmos as continuous with itself and, “so long as its entities are interrelated, can go on and on without meeting its own [100]boundedness, its own limits, since it is not held together from outside itself ” (2008: 182). He gives an example of this cosmos through Henry, a Native American shaman he met thirty years before. Henry inhabits a cosmos that is sentient, conscious, and densely interrelated, so that “beings affect one another—a change in one is likely to generate changes in others” (2008: 182). This is quite consistent with much of Umbanda itself, which demonstrates an ontological creativity characteristic of organic, porous spirit possession cults worldwide that dialogue closely with their immediate surround. Thus, José Bairrão (2004) argues that the recent increase in child Exu spirits—Exus mirins—being incorporated in trance and worshipped in Umbanda temples/centers reflects growing concerns in the 1990s and early 2000s with urban street children’s delinquency and propensity for crime. These mirins, he says, are the antithesis of childhood innocence, purity, or beauty; generally described as ugly, sensitive, and needy, but dangerous street creatures (2004: 63). One mãe-de-santo describes these spirits in terms close to those used for child criminals: “They’re afraid, but they make you more afraid than them, so they can intimidate you, exactly like a street punk” (2004: 63). Another interlocutor tells him that the spirit she “passes” (is possessed with) known as Mirim da Mata da Beira da Estrada (Mirim of the Bush of the Edge of the Road), had no mother or father, but was raised by prostitutes who sheltered and took care of him. He would steal from the wallets of their clients while they were in the act. Mirim da Mata da Beira da Estrada was murdered by the owner of a food warehouse that he stole from because he was so terribly hungry (65). Another group of child spirits of a temple Bairrão studied died, burned to death on a cold night. They were sleeping on the floor draped by newspapers when someone set them on fire. They didn’t react because they were high on drugs (65). Given Brazil’s level of child homelessness, violence, and substance abuse, it is entirely unsurprising that Bairrão links these Umbandists’ possession experiences of these spirits directly with the nation’s state of affairs. In this sense, Umbanda serves as an immediate, open-ended reflection on some of Brazil’s most pressing historical and social facts.

On the other hand, Umbanda also serves as a moral gauge or compass of its surround. It not only lives amid an ecology of what Stephen Selka has called competing moral orders (2010: 293), but has largely emerged as an organized movement through its selective appropriation of specific moralities. It is unsurprising that Exus are overwhelmingly described as lowly or inferior beings in this greater scheme. Kelly Hayes (2011), for instance, observes that Umbanda is riddled with references to the Devil or Lucifer, to Hell, and to demonic forces, particularly in Exu-related imagery, songs, and rituals. However, for her, Pomba Gira discourse (which she analyzes in her book) is not explicitly concerned with right or wrong, or indeed darkness or subversion, but “in dialogue with normative ideals,” which include feminine morality and sexuality (2011: 29). The very biographies of the spirits attest to the social inversions, albeit not always with resolutions or happy endings, made explicit and, to some extent, available by their presence in the lives of devotees. Take the story of Maria Molambo, a famous Pomba Gira:

In the story’s most basic form, Maria Molambo was the daughter of a wealthy landowner who fell hopelessly in love with a young man against her father’s will. Learning of his daughter’s affair, the landowner arranged for [101]her to marry a husband of his choosing. The girl and her lover ran off, were persecuted relentlessly, and eventually were captured. To restore his name and honor the father disowned the daughter and ordered his henchmen to kill her lover. With nowhere to go, Maria Molambo was forced to live on the streets, prostituting herself to survive. (Hayes 2011: 44–45)

For Hayes, Molambo’s story conveys messages that are consonant with women’s concerns with gender, sexuality, morality, and female agency (Hayes 2011: 46). The Umbandista protagonist in her book, Nazaré, gives form to the incongruities experienced by many women in Brazil: “Embodied in Pomba Gira, the holy harlot, these tensions are made dramatically manifest in the material world, thereby becoming subject to human reflection and manipulation” (Hayes 2011: 10). According to Hayes, all other Afro-Brazilian spirit entities provide a set of “symbolic resources” by which people like Nazaré become empowered. At first sight then, Umbanda is the prototype for Handelman’s “fuzzy” cosmos, which is continuous-with-itself in the sense that there is no rigid boundary between spirit cosmology and “real life”: identities, events, and characters transit easily between them, some crystallizing on the former end as spirit forms.

However, even in less prosperous zones of Rio de Janeiro, Umbanda’s entities seem to be far from clear-cut. Once, on a bus back from a beach ceremony with a traditional but poor temple in northern Rio, one with roots in Brazil’s Islamic slave ancestry, which Bastide called the “Mussulman religion in Brazil” (2007: 143), Francisca, one of the temple’s oldest and most trusted ritual children spoke frankly about the nature of her spiritual entidades. She said that while Caboclos were probably “higher” sorts of entities who had lived in the forests, the Exus and the Pretos Velhos were entities “of the earth” (entidades da terra) who had lived lives and made errors and who now came to tie up loose strings. However, she said, these spirits could have lived lives as professionals, doctors, engineers, tailors, or teachers—not just slaves, or people of the street, for that matter. At stake here for Francisca, we can infer, was not whether Maria Molambo lived the life and death as described for her and came back to work with Umbanda mediums but whether Umbandists conceive that other spirits (not related to her specific biography) may “assume” her identity in order to work. Francisca’s cosmos, in other words, has aspects both of an organic, intragrated universe that obviates the necessity of belief and a cosmos that has been fractured and is showing an increasing rift between the world of people and a world of the divine that remains somewhat inaccessible. Thus, the Preto Velho may not have been the spirit of a black slave at all, but a nineteenth-century French engineer, and Francisca would never know. In a similar vein, the leader of one of São Paulo’s Umbanda federations once asked a Preto Velho why it was drawing mandalas on the ground with chalk, instead of the usual cosmograms. He responded that he was a Buddhist but that he “came as” a Preto Velho because he liked “their energy.” Other spirits, such as Pombas Giras, however, were regarded far more literally—as women who had lived on the street. Here “belief in the unfathomable” is somewhat important to overcoming this discontinuity of identities (Handelman 2008: 181), which are not restricted in any way to merely Exus.

The closer we get to São Paulo’s intellectual Umbandist elite, characterized by Umbanda Sagrada among others, the more closed the cosmos becomes. Thus, [102]Adriano Camargo, a well-known Umbandist leader and book writer in São Paulo known as “the herbalist,” told me in 2013 that

Umbanda doesn’t have any personalism. It’s not like in Spiritism where you get messages from a spirit called José da Silva, or whomever. In Umbanda it’s the force of corporation, the battalion, the archetype, that counts. And the archetype is not a collection of beings but a thought form. So, the Caboclo is not a Caboclo—he has a symbolic name that determines his strength, his field of action. But there is no personalism. (Interview with Adriano Camargo, October 2013)

This lack of personalism has come hand-in-hand with sophisticated theological treatises that chart the organization and functioning of the universe according to certain spirit “masters,” such as Saraceni’s Pai Benedito de Aruanda, and which need to be studied, understood, and “believed” in before the cosmos can manifest in its fullest possibilities. In conversation with me, Saraceni asserted that Pai Benedito has merely provided the skeleton of a growing body of doctrine and knowledge. He added that while spirit manifestation itself is potentially infinite, there are certainly keys that are immutable, for instance, the “mysteries.” These “mysteries” of God may express themselves in myriad ways crossculturally, or even within Umbanda itself, but are inherent to nature of the universe, to the order of things, and to the character of human existence. Umbanda itself then becomes a symbol, or set of symbols, for a greater truth, manifest though plastic spirit identities.

But with the concept of “mystery,” among others, Umbanda Sagrada effectively has its cosmos encompassed by God and His unknowable aspects (spirits, forces). As Handelman says, “belief accepts (indeed, invokes) cosmic closure by believing in boundless, infinite God beyond the radical, paradoxical rupture between the infinite and the finite” (2008: 187). This particular Umbanda cosmos becomes thus much closer to a monotheistic religion than to a spirit possession cult proper, losing its plasticity on a more human level while gaining it on a transcendent one. But according to Handelman, even monotheistic cosmoses need some mode of passage from below to above—from the finite to the infinite—or they will lack integrity (2008: 187–88). While for Judaism, Christianity, or Islam it is belief that accomplishes this task, for a discontinuous Umbanda such as that manifest in the variations of Umbanda Sagrada, it is possession that does. It accesses the “inaccessible” through moral and archetypical perspectives wielded by spirit forms. But this, like Catholicism itself, which paradoxically promises to afford the believer some contiguity with the otherworldly, only to retract the boundary of the otherworldly to the world of humans (through the way the church changes its grounds for considering sainthood, for instance; see Handelman 2008: 186), is ultimately illusory too.

For Adriano (above), for example, the medium’s own history is an integral part of the message: he argues that it’s at this personal level, which determines the regional languages of spirit manifestation, that there are variations, or changes, not at the level of the archetype. For him, the force of the archetype was designed way back and remains fixed throughout Brazil’s cultural geographies, with the spirits manifesting in exactly the same ways with only slight variations—the Preto Velho will always come as a bent-over old man who needs a cane, the Caboclo as an energetic Amazonian Indian, the Criança as the prototypical child who wants (and [103]works through) candy, and so on. For Adriano, each temple projects its own energetic “designs” within these archetypes, which are what will determine the exact phenomenology of trance. “Umbanda has that freedom,” he says; “it resonates with your mediumship. The entities are not created individually, but by the force of a group.” However, these “perspectives” are not regarded as “evidence” of spirits but as the culturally sanctioned means by which these very (anonymous) spirits give evidence (in the form of advice, counsel, predictions, healing, and so on).

There is a sense in which this double-cosmos resembles a Catholic cosmology that has “always been peopled with anthropomorphically recognizable figures, with creatures shaped very much “like us”” (Mayblin 2014: 273). Corporeal identification with the divine always matters, Mayblin says (2014: 273). What lies beyond this identification becomes, however, largely a matter of faith. In monotheism, Handelman says, the boundary between the divine and the human is constituted as paradoxical; Mass shows this well. The first half prepares participants to approach the divine while on the level of the human; in the second half—through the ritual transformation of bread and wine, for instance—it is the Divine Presence that encompasses worshipers. In the end, “the agency is that of the Divine, not of the human” (2008: 186). In the “closed” Umbanda I have described, a similar case occurs: the divine “is, or becomes, entirely its own referent” (Handelman 2008: 186).


Abramson, Allen, and Martin Holbraad. 2012. “Contemporary cosmologies, critical reimaginings.” Religion and Society: Advances in Research 3: 35–50.

———. 2014. “Introduction: The cosmological frame in anthropology.” In Framing cosmologies: The anthropology of worlds, edited by Allen Abramson and Martin Holbraad, 1–30. Manchester: University of Manchester Press.

Bairrão, José Francisco M. H. 2004. “Sublimidade do mal e sublimação da crueldade: criança, sagrado e rua.” Psicología: Reflexão e Crítica 17 (1): 61–73.

Barad, Karen. 2003. “Posthumanist performativity: Towards an understanding of how matter comes to matter.” Journal of Women in Culture and Society 28 (3): 801–30.

Barth, Frederick. 1987. Cosmologies in the making: A generative approach to cultural variation in inner New Guinea. Cambridge: Cambridge University Press.

Bastide, Roger. (1960) 2007. The African religions of Brazil: Towards a sociology of the interpenetration of civilizations. Translated by Helen Sebba. Baltimore: The Johns Hopkins University Press.

Beliso-De Jesús, Aisha M. 2015. Electric Santería: Racial and sexual assemblages of transnational religion. New York: Columbia University Press.

Bourguignon, Erika. 2007. “Spirit possession.” In A Companion to psychological anthropology: Modernity and psychocultural change, edited by Conerly Casey and Robert B. Edgerton, 374–88. Oxford: Blackwell Publishing.

Brown, Diana DeGroat. (1986) 1994. Umbanda: Religion and politics in urban Brazil. New York: Columbia University Press.[104]

Brunton, Ron. 1980. “Misconstrued order in Melanesian religion.” Man 15 (1): 112–28.

Camargo, Candido P. F. de 1961. Kardecismo e Umbanda: Uma interpretação sociológica. São Paulo: Livraria Pioneira Editora.

Capone, Stefania. 2010. Searching for Africa in Brazil: Power and tradition in Candomblé. Translated by Lucy Lyall Grant. Durham, NC: Duke University Press.

Cavalcanti, Maria L. V. de Castro. (1983) 2008. O Mundo Invisivel: Cosmologia, Sistema Ritual e Noção de Pessoa no Espiritismo. Biblioteca Virtual de Ciências Humanas, at www.bvce.org.

Espirito Santo, Diana. 2015. Developing the dead: Mediumship and selfhood in Cuban Espiritismo. Gainesville: University Press of Florida.

Giumbelli, Emerson. 2003. “O ‘Baixo Espiritismo’e a História dos Cultos Mediúnicos.” Horizontes Antropológicos 19: 247–81.

———. 2013. “Amerindian and priest: An entity in Brazilian Umbanda.” In The social life of spirits, edited by Ruy Blanes and Diana Espirito Santo, 179–97. Chicago: University of Chicago Press.

Gois Dantas, Beatriz. 1988. Vovó nagô e papai branco: usos e abusos da África no Brasil. Rio de Janeiro: Graal.

Goldman, Marcio. 2007. “How to learn in an Afro-Brazilian spirit possession religion, ontology and multiplicity in Candomblé.” In Learning religion: Anthropological approaches, edited by David Berliner and Ramon Sarró, 103–20. New York: Berghahn Books.

———. 2009. “Histórias, devires e fetiches das religiões afro-brasileiras: Ensaio de simetrização antropológica.” Análise Social XLIV (190): 105–37.

Guenther, Mathias. 1999. Tricksters and trancers: Bushman religion and society. Bloomington: Indiana University Press.

Hale, Lindsay. 2009. Hearing the mermaid’s song: The Umbanda religion in Rio de Janeiro. Albuquerque: University of New Mexico Press.

Handelman, Don. 2002. “The interior sociality of self-transformation.” In Self and self-transformation in the history of religions, edited by David Shulman and Guy G. Stroumsa, 236–53. New York: Oxford University Press.

———. 2004. “Introduction: Why ritual in its own right? How so?” Social Analysis 48 (2): 1–32.

———. 2008. “Afterword: Returning to cosmology—Thoughts on the positioning of belief.” Social Analysis 52 (1): 181–95.

———. 2014. “Inter-gration and intra-gration in cosmology.” In Framing cosmologies: The anthropology of worlds, edited by Allen Abramson and Martin Holbraad, 95–115. Manchester: Manchester University Press.

Hayes, Kelly E. 2011. Holy harlots: Femininity, sexuality, and black magic in Brazil. Berkeley: University of California Press.

Hyde, Lewis. 2010. Trickster makes this world: Mischief, myth, and art. New York: Farrar, Straus and Giroux.[105]

Johnson, Paul. 2014. “Spirits and things in the making of the Afro-Atlantic world.” In Spirited things: The work of “possession” in Afro-Atlantic religions, edited by Paul C. Johnson, 1–22. Chicago: University Press.

Kardec, Allan. 1857. The book of spirits. http://www.spiritwritings.com/kardecspiritstoc. html.

Lambek, Michael. 2014. “Afterword: Recognizing and misrecognizing spirit possession.” In Spirited things: The work of “possession” in Afro-Atlantic religions, edited by Paul C. Johnson, 257–76. Chicago: University of Chicago Press.

Leal de Souza, Antônio Eliezer. 1933. O Espiritismo, a magia e as sete linhas de Umbanda. Rio de Janeiro.

Maggie, Yvonne. 1992. O medo do feitiço: relações entre magia e poder no Brasil. Rio de Janeiro: Arquivo Nacional.

Malabou, Catherine. 2010. Plasticity at the dusk of writing: Dialectic, destruction, deconstruction. Translated by Carolyn Shread. New York: Columbia University Press.

Mayblin, Maya. 2014. “People like us: Intimacy, distance, and the gender of saints.” Current Anthropology 55 (10): 271–80.

Negrão, Lísias. 1996. Entre a Cruz e a Encruzilhada: Formação do Campo Umbandista Em São Paulo. São Paulo: EDUSP.

Nielsen, Morten. 2012. “Interior swelling: On the expansive effects of ancestral interventions in Maputo, Mozambique.” Common Knowledge 18 (3): 433–50.

Ortiz, Renato. 1978. A Morte Branca do Feiticeiro Negro: Umbanda e Sociedade Brasileira. São Paulo: Editora Vozes.

Palmié, Stephan. 2013. The cooking of history: How not to study Afro-Cuban religion. Chicago: University of Chicago Press.

Pedersen, Morten. 2001. “Totemism, animism, and North Asian indigenous ontologies.” Journal of the Royal Anthropological Institute 7 (3): 411–27.

———. 2007. “Talismans of thought: Shamanic ontology and situated cognition in Northern Mongolia.” In Thinking through things: Theorising artefacts ethnographically, edited by Amira Henare, Martin Holbraad, and Sari Wastell, 141–66. London: Routledge.

———. 2014. “Transitional cosmologies: Shamanism and postsocialism in Northern Mongolia.” In Framing cosmologies: The anthropology of worlds, edited by Allen Abramson and Martin Holbraad, 164–81. Manchester: University of Manchester Press.

Pedersen, Morten, and Rane Willerslev. 2012. “‘The soul of the soul is the body’: Rethinking the concept of the soul through North Asian ethnography.” Common Knowledge 18 (3): 464–86.

Prandi, Reginaldo. 1994. “Pombagira dos candomblés e umbandas e as faces inconfessas do Brasil.” Revista Brasileira de Ciências Sociais 26: 91–102.

Radin, Paul. (1956) 1973. The trickster: A study in American Indian mythology. New York: Schoken Books.

Saraceni, Rubens. 2013. Doutrina e Teologia de Umbanda Sagrada. São Paulo: Madras.[106]

Selka, Stephen. 2010. “Morality in the religious marketplace: Evangelical Christianity, Candomblé and the struggle for moral distinction in Brazil.” American Ethnologist 37 (2): 291–307.

Shapiro, Matan. 2016. “Curving the social, or, why antagonistic rituals in Brazil are variations on a theme.” Journal of the Royal Anthropological Institute 22 (1): 47–66.

Taylor, Charles. 2007. A secular age. Boston: Belknap Press.

Viveiros de Castro, Eduardo. 1998. “Cosmological deixis and Amerindian perspectivism.” Journal of the Royal Anthropological Institute 4 (3): 469–88.

Willerslev, Rane. 2007. Soul hunters: Hunting, animism, and personhood among the Siberian Yukaghirs. Berkeley: University of California Press.

Des Parures pour les esprits: ouvrir et fermer le cosmos dans la religion brésilienne Umbanda

Résumé : Dans cet article, je propose des modèles alternatifs pour comprendre la relation des cosmologies incluant des esprits avec leurs entourages sociaux. Dans mon ethnographie, je montre la valeur accordée à la fluidité et la plasticité dans un secteur urbain de la religion spiritiste Brésilienne Umbanda. Des fidèles Umbandistes actuels, en particulier dans la ville de São Paulo, veillent à ce que la « culture », c’est à dire l’imaginaire historique brésilien, devienne le médium avec lequel les esprits, réflexivement, se parent. Cependant, si l’on a recours à la division, établie par Don Handelman, entre les cosmos qui se tiennent « de l’intérieur » et ceux qui sont maintenus « de l’extérieur », il apparaît que les Umbandistes adoptent des degrés variés d’ouverture et de fermeture par rapport à leur cosmologie, notamment dans le secteur urbain que j’étudie, dont le cosmos malléable est aussi essentiellement brisé, d’où la nécessité d’y croire.

Diana ESPÍRITO SANTO (PhD, UCL 2009) has worked both on concepts of knowledge and person in Afro-Cuban spirit mediumship and on notions of cosmology and change in Brazilian Umbanda. She is author of Developing the dead: Mediumship and selfhood in Cuban Espiritismo (2015, University Press of Florida), and she is coeditor of three books, including The social life of spirits (2014, University of Chicago Press). She is currently assistant professor of social anthropology at the Pontificia Universidad Católica de Chile in Santiago.

Diana Espírito Santo
Programa de Antropología Instituto de Sociología
Pontificia Universidad Católica de Chile
Vicuña MacKenna 4860
Macul, Santiago