Encountering Images in Candomblé

This art icle was downloaded by: [ ROGER SANSI ] On: 14 Decem ber 2012, At : 09: 11 Publisher: Rout ledge I nform a Lt d Regist ered in England and Wales Regist ered Num ber: 1072954 Regist ered office: Mort im er House, 37- 41 Mort im er St reet , London W1T 3JH, UK Visual Anthropology: Published in cooperation with the Commission on Visual Anthropology Publicat ion det ails, including inst ruct ions f or aut hors and subscript ion inf ormat ion: ht t p: / / www. t andf online. com/ loi/ gvan20 Encountering Images in Candomblé Roger Sansi Version of record f irst published: 14 Dec 2012. To cite this article: Roger Sansi (2013): Encount ering Images in Candomblé, Visual Ant hropology: Published in cooperat ion wit h t he Commission on Visual Ant hropology, 26: 1, 18-33 To link to this article: ht t p: / / dx. doi. org/ 10. 1080/ 08949468. 2013. 735175 PLEASE SCROLL DOWN FOR ARTI CLE Full t erm s and condit ions of use: ht t p: / / www.t andfonline.com / page/ t erm s- and- condit ions This art icle m ay be used for research, t eaching, and privat e st udy purposes. Any subst ant ial or syst em at ic reproduct ion, redist ribut ion, reselling, loan, sub- licensing, syst em at ic supply, or dist ribut ion in any form t o anyone is expressly forbidden. The publisher does not give any warrant y express or im plied or m ake any represent at ion t hat t he cont ent s will be com plet e or accurat e or up t o dat e. The accuracy of any inst ruct ions, form ulae, and drug doses should be independent ly verified wit h prim ary sources. The publisher shall not be liable for any loss, act ions, claim s, proceedings, dem and, or cost s or dam ages what soever or howsoever caused arising direct ly or indirect ly in connect ion wit h or arising out of t he use of t his m at erial. Visual Anthropology, 26: 18–33, 2013 Copyright # Taylor & Francis Group, LLC ISSN: 0894-9468 print=1545-5920 online DOI: 10.1080/08949468.2013.735175 Encountering Images in Candomblé Downloaded by [ROGER SANSI] at 09:11 14 December 2012 Roger Sansi ‘‘Images’’ are a problem in the Afro-Brazilian religion Candomblé in two senses. First, the presence of Catholic images in shrines has been questioned by the ‘‘re-africanization’’ movement in the last decades, as a form of syncretism. Second, it is often said that taking pictures or videos of rituals is not allowed. However, in my ethnographic experience, religious images, pictures and videos are all used in many Candomblé houses. How can we understand this contradiction? What then is the problem with images in Candomblé after all? THE STATUS OF IMAGES In the literature on Afro-Brazilian religions, images appear as a problem, in two senses. First, the presence of figurative images in shrines has been questioned as a form of syncretism by the ‘‘re-africanization’’ movement. Second, it is often said that it is not allowed to take pictures or videos of rituals. However, in my ethnographic experience in Cachoeira, which is a city known for its Afro-Brazilian religious traditions near Salvador da Bahia, Brazil, religious images and figurative representations in general, as well as pictures and videos, are used in many Candomblé houses. What then is the problem with images? And what makes them suspicious and troublesome? Maybe we should start by acknowledging that the suspicion of images is not exclusive to the Afro-Brazilianist literature. In a recent text, The Future of the Image, Jacques Rancière wonders why images are still contentious and suspicious. On the one hand, they are often read as texts; but on the other hand, ‘‘we continue to regard them as a promise of flesh, capable of dispelling the simulacra of resemblance, the artifices of art, and the tyranny of the letter’’ [2007: 7]. The ambiguity between text and presence, what Barthes [1981] called the studium and the punctum, troubles our understanding of images; that is what the art historian W. J. T. Mitchell calls our ‘‘double consciousness’’ toward images [2005]. The scholarly study of the image, for decades, from iconology to visual culture studies, has often reduced images to ‘‘visual languages’’ or visual texts. But in recent years a literature that wonders about the ‘‘power,’’ ‘‘agency’’ and ‘‘presence’’ of images has started to build up in reaction to this ROGER SANSI is a Lecturer in Anthropology at Goldsmiths’ College, University of London. He has worked on Afro-Brazilian art and culture. He has recently published Fetishes and Monuments, Afro-Brazilian Art and Culture in Bahia [Berghahn 2007] and Sorcery in the Black Atlantic, ed. with Luis Nicolau Pares [Chicago 2010]. E-mail: [email protected] 18 Downloaded by [ROGER SANSI] at 09:11 14 December 2012 Images in Candomblé 19 ‘‘textualism.’’1 From Alfred Gell’s work on art and agency [1998], through Latour on iconoclasm [2002], Hans Belting’s [1997, 2005] approach to the question of presence in religious images, Birgit Meyer’s [2008] discussion of the use of new media in religious practice, to Webb Keane’s [2007] work on semiotic ideologies and fetishism, or W. J. T. Mitchell’s [2005] interpellation for what images are ‘‘wanting,’’ I would venture to say that one of the general questions that all these different authors address is: do we look at images not only as texts but also as persons? Are images not only fixations of collective representations, or symbols, but social persons who actively participate in social exchanges? This question would not necessarily imply a rejection of semiotics as a method, in the larger sense, as we find it for example in Peirce [1984]. We could start by saying that not all signs are symbols, that is, are abstract representations in a language that we need to decode; as Keane puts it [2008: 230]; a contemporary materialist semiotics stresses that signs can also be indexes, consequences, effects, causally related to their meaning. Gell’s and Belting’s ‘‘anthropologies of the image,’’ for example, owe a lot to Peirce’s approach. Alfred Gell’s take on artworks and images as ‘‘indexes’’ is based, though not explicitly, on Peircean notions of indexicality. Hans Belting’s anthropology of images is perhaps a more complex formulation of the same question. He insists on the mediation of images, the fact that they are made present, they happen through a medium. Beyond the dichotomy of signifier and signified, symbol and content, Belting proposes to understand images from the triad image–medium–body, in which medium ‘‘is to be understood not in the usual sense but in the sense of the agent by which images are transmitted, while body means either the performing or the perceiving body on which images depend no less than on their respective media’’ [Belting 2005: 202]. The relationship between image and body through the medium is not just an object–subject relation but can be a relation between persons, in which the image can enact the ‘‘presence of an absence,’’ the presence of another person. Images are sometimes confused or superposed to bodies; people can become images and in the same way images can become people because, as Belting says, the body does not only perceive the image but can also perform it, or in other terms, embody the image. In this sense the materiality of the medium and the body is stressed as key in the process of becoming of images.2 This dynamic understanding of the image, as something that happens, an event, more than a representation, is central to these new approaches to the question of the image. And perhaps, going further, but back to Rancière’s question, the ambiguity of this event is central to the dynamism of the image: the ambiguity between representation and a presence that is not quite there, text and ‘‘promise of flesh.’’ More than the fact that they can be social persons, it is the doubt that can be cast upon them, their ambiguous situation between text and body, the fact that they can be social persons, but perhaps not quite, just the presence of an absence, maybe just a trace or a document, what gives dynamism to images, its ‘‘power.’’ In Mitchell’s terms [2005], the question we often implicitly ask is what do images ‘‘want,’’ in the sense of what they are ‘‘wanting’’ or lacking to become full persons. This ambiguity is also what generates a backlash against them, iconoclasm. Iconoclasm, after all, is nothing but an attempt to cut images from their dynamic process of becoming, to pre-empt their Downloaded by [ROGER SANSI] at 09:11 14 December 2012 20 R. Sansi possible transformation into other images. But as Latour has explained [2002], by this very attempt to stop images from happening, iconoclastic acts generate yet other dynamics, other images: the iconoclastic act, as an event, becomes an image itself. These issues are present, I think, in the contradictory discourses and uses of images in and around Candomblé. In the next pages I will address these issues first in reference to religious images in shrines, and secondly in relation to reproduced images like pictures or videos. Bringing together these two instances I am also following Belting [2005] or Meyer’s [2008] approach to the relation of images and media, which does not only focus on the transformations brought forward by ‘‘new’’ media, such as video, but proposes to look together at ‘‘new’’ and ‘‘old’’ forms of mediation and objectification, not taking for granted that the ‘‘new’’ will inevitably bring radical transformations in the relationship of people and images. As Belting says, images are not produced by media, but only transmitted by them [Belting 2005: 205]. It is the relationship between image, medium and body, what makes images happen, and not the medium itself. In that sense he insists on using a very general notion of the ‘‘image’’ as disembodied, or separated from medium and body, bringing together ‘‘external’’ images such as sculptures or photographs, and ‘‘internal’’ images such as visions or dreams. By looking at the different instances in which the religious other (the saint, the divine, the holy) is seen, I propose a more general discussion of the role of images in Candomblé. RELIGIOUS IMAGES For some of the intellectuals and priests in the more traditional and powerful houses of Candomblé in Bahia, like the Opô Afonjá, Candomblé shrines should not have Catholic images. According to them Yoruba shrines, back in West Africa, rarely included anthropomorphic representations of the gods but only woodcarvings of initiates being possessed by the gods. The gods in themselves were not represented in human form: their presence appears in the ritual of possession, in the body of their initiates [e.g., Santos 1967; Thompson 1993]. Therefore anthropomorphic images in Candomblé shrines would be only peripheral, as opposed to those in Catholic shrines, which are precisely built around the figurative image of saints. In the 1980s Mãe Stella of the Opô Afonjá launched an anti-syncretism movement that included, amongst other things, withdrawing Catholic images from the shrines of the temple. Her argument was that Candomblé and Catholicism were two different things, and should not be mixed up. This move made some older members of the Opô Afonjá who were devoted to these images quite upset, but in the long run Mãe Stella’s position was accepted. Candomblé is not a unified religion, each house is self-organized, but the growing influence of traditional houses such as Opô Afonjá, especially in urban areas, and the general ‘‘re-africanization’’ of the religion [Capone 1998; Silva 1995], has resulted in many Candomblé houses now following her model of separating religions and images [Sansi 2007]. That was not the case at Madalena’s house, where I did a good part of my fieldwork, in Cachoeira. Likewise in most Candomblé houses in Cachoeira, Downloaded by [ROGER SANSI] at 09:11 14 December 2012 Images in Candomblé 21 Catholic images were (still?) integral to the shrines. How to understand their resilient presence there? Is this just an act of masking, or dissimulating, a ‘‘colonization of consciousness’’? Maybe ‘‘syncretism’’ has become tradition? Perhaps. But I think that there is something else to it, something more fundamental. Catholic images in Candomblé shrines are not seen as icons of Catholic religion, and are not worshiped as at Catholic shrines. They are a part of the indexical series composing Candomblé shrines, one which includes objects from the more random origins gathered because the devotees recognize in them aspects or qualities of their gods, and more than a physical resemblance. In this sense images are like pieces of cloth, jewelry, flowers or foods. But to explain this point a bit further, maybe I should talk more extensively about Candomblé shrines. The presence of Candomblé shrines in the houses of Cachoeira, and the practice of Candomblé rituals, is not necessarily evident to outsiders. In recent years the new public legitimacy of Candomblé would make one think that it would be possible for Candomblé houses to relocate to more public places, yet still most houses maintain their traditional concealment. This concealment is inevitable, not just because of the history of police repression and social disapproval toward the religion of former slaves. Concealment is necessary because Candomblé is based on initiation and secrecy. It does not proselytize like Christianity. The more open, or public, rituals of Candomblé are the celebrations when the santo descends into the body of the initiate and dances. These celebrations (festas) are public and everybody is welcome, because the Orixás are generous and flashy, and they like people to admire their dances and their beauty, and to have something to eat and drink in their honor. The people who go to these festas do not necessarily have anything to do with Candomblé. The rituals of possession are the public face of Candomblé; a public face that is ephemeral, taking place in a hall (sala or barracão) that remains empty most of the time. But Candomblé shrines are hidden in the backyard or on the other side of the wall from the Catholic shrine, which in most houses in Cachoeira is very publicly displayed in the front rooms. Candomblé shrines, in contrast, are kept in a closed room that nobody can see (the quarto de santo). The Candomblé shrine is very different from the Catholic shrine: Candomblé shrines are accumulations of objects related to the Orixá on a dais, together with pots and other containers, generally covered with cloths and hidden from public view, inside which there are other objects, generally stones (otã), that embody the force of the Orixás (axé). There the axé is seated, and the shrine is therefore called assento, or seat. They are not supposed to be seen, and thus a devotee is not supposed to look at them directly. Each initiate has her own assento, and the shrine is an accumulation of assentos around the central assento, which belongs to the mãe de santo, the mother of the santo, or priestess, who has initiated the others, their daughters of the santo (filhas de santo). Before the ritual of possession the shrines receive offerings, they are fed, and these offerings result in the transposition of the santo to the body of an initiate during the public celebration. The assento is not the image but the house of the santo. Shrine offerings open the circulation of force (axé) that will culminate in spirit possession. The body of the filha de santo, possessed, dancing and dressing in the santo’s clothes, is in fact the image of the santo. And this Downloaded by [ROGER SANSI] at 09:11 14 December 2012 22 R. Sansi image changes through time, as much as the relationship between santo and initiate changes. Progressively, as the initiate grows and deepens her knowledge, her body gets better adapted to the ritual of possession, which becomes less violent. At the same time the shrine becomes bigger, richer, full of presents, an objectification of this intense interaction. The shrines are also in constant transformation. The filhas de santo talk to the shrines, asking for help of their santo; continuously, offerings and presents are being added; and periodically, the pots are washed and their offerings renewed: the ‘‘life’’ of shrines is constantly renewed by the interaction with the initiates. We have to consider that the constant ritual feeding establishes a highly determined and determinant relationship between shrine and initiate, to the extent that it almost becomes an exterior organ of her body, a part of her ‘‘distributed person’’ [Gell 1998]. Shrines are indexes, as Gell [idem] would say, since they are also the trace or embodiment of a particular history of exchanges between the devotee and the saint. In this sense we could say that the shrine, as the body of the santo, becomes a person, a partner in mediated exchange, and not just an object of exchange. What Gell called the ‘‘distributed person": ‘‘As social persons, we are present, not just in our singular bodies, but in everything in our surroundings which bears witness of our existence, our attributes, and our agency’’ [ibid.: 103]. We could look at Candomblé as a dynamic system that builds persons. It does not only try to classify people through archetypes or reflect a repressed ego, as psychological interpretations of possession have often postulated, but its ritual practices produce new social persons. If we see the person as an opened process of becoming, we could think that the santos are active elements that collaborate precisely in the construction of a person that is always in the making, becoming both in bodies and in shrines. The objects accumulated in the shrines correspond to the attributes of the santo. Ogun, the Orixá of war and iron, whose color is blue, receives iron tools and blue robes. Oxum, the goddess of love, beauty, wealth and fresh water, receives jewelry, make-up, golden objects and clothes. But despite the fact that we often find similar objects in different shrines there is virtually no limit to what can be added to the shrine. The association with the santo can be extended to many kinds of object just because the filha de santo finds that they suit the santo. The house of the santo is not a recipe but it is something more personal: it is a present, a gift. And presents are difficult to make: they are meant to correspond naturally to a person, and it is not given that one finds what one is looking for, in these cases. It is often true that we find things that would make good presents by coincidence, more often that when we are actually obliged to make them. There is an element of chance in finding an object that we feel as corresponding naturally to somebody. This is why these presents are always, in some sense, ‘‘found.’’ For example, Madalena found an enormous iron pan while digging in her backyard [Figure 1]. This pot, according to her, was from the time of slavery, used to make cocada (a sweet made from coconut). She decided to put it on Ogum’s shrine, thinking that it was he who had made her find this pot in order to put it on his shrine. We can describe these processes of recognition of the sacred in everyday life in correspondence to surrealist theories of the objet trouvé and ‘‘objective chance,’’ situations in which the unexpected encounter of an Downloaded by [ROGER SANSI] at 09:11 14 December 2012 Images in Candomblé 23 Figure 1 Ogum altar, house of Mãe Madalena, 2005. (Photo # Roger Sansi; color figure available online) object is perceived as a moment of revelation, in which the arbitrary becomes necessary: here the fortuitous encounter of a giant iron pot had to happen. In other terms, the found object is an index of the santo, who left it there to be found by the initiate. That means that anything can become part of a shrine if the santo wants it to. An Ogum shrine can thus accumulate pots, tools, car parts or typewriters, if he wants it to. This is an important point, in order to understand the apparently random and disordered character of the objects accumulated on shrines. They don’t stand separately for specific qualities that make a logical ensemble; on the contrary, they are accumulations of presents, of objects that people have found, feeling that they fit naturally to the spirit that lives in the shrine. Shrines are the result of a personal history of exchanges between people and their santo.3 If we want to use a contemporary terminology, they are an assemblage, a jumbling together of discrete parts or pieces, objects and subjects, that is capable of producing effects; the assemblage in this sense is something more, or something else, than the mere accumulation of these parts. Thus the encounter of Madalena with the iron pan produces something more than just the accumulation of the object: it produces her Ogum, her santo. One of the objects we can find in shrines is figurative images. As mentioned before, for some Candomblé scholars like Mestre Didi, the gods of Candomblé Downloaded by [ROGER SANSI] at 09:11 14 December 2012 24 R. Sansi are not ‘‘represented’’ in sculptures: in West Africa, origin of the cults, the wood-carvings found in the shrines represented the initiates being possessed by the Orixás rather than the Orixás themselves [Santos 1967]. This argument has since been used to criticize the ‘‘syncretistic’’ use of Catholic images in Candomblé shrines, by anti-syncretistic movements that wanted to renew and re-africanize Candomblé. The fact is that in Cachoeira, still nowadays, we can find images, figurative representations in human form, that are Catholic or not, in many shrines, though they are not particularly central. Like other objects in shrines, the images can often be described as found objects (or found images, in this case), images that have been recognized as their saint by initiates. Thus they are not seen so much as representations of Catholic belief but as indexes of this personal encounter. That is probably the way that Catholic images were initially incorporated in shrines: some initiates recognized in already existing images the images of their Orixás. With time, Candomblé shrines have incorporated many different kinds of image, beyond the Catholic ones. If Catholic images were once dominant in Candomblé, it was not just because of an implicit project of ‘‘syncretism’’ between religions but also because Catholicism was overwhelmingly hegemonic in the production of public images in Brazil. But Candomblé has also borrowed from many other sources of popular culture. The images of the Caboclo, for example, were born out of the monuments to the Caboclo as a symbol of independence. Other iconographies of the Caboclos were clearly inspired from Western movies and more recently Amazonian Indians. Oxum is represented as a strawberry blonde mermaid. Exu, associated with the devil, takes the form of evil magicians from comic books, or gangsters. Some images even come from paintings or drawings of the 19th century, like Escrava Anastácia. This still happens today with all kinds of image. I was told of one case in an image factory, whose salesmen are always trying to understand the latest trends in the market of religious images. The owner of a bar decided to paint an image of a drunkard from Italy as a black man with white clothes. When one if his clients saw it, he recognized his ‘‘Preto Velho,’’ the spirit of an old slave. The factory decided to reproduce this image and it had a certain success. The iconographies of these images, therefore, can be completely strange to the world of Candomblé: in the same way that with objects, there is no limit to the images that can be incorporated into the shrine. In Madalena’s house, for example, we can consider the case of Sara the Gypsy, who also has a special story. When she was still a teenager, Madalena lived with a man in very poor conditions; but one day the man abandoned her for another woman. She was crying at the door of her house and then saw a beautiful Gypsy woman (zigana). ‘‘Why do you cry?’’ she asked her. Madalena explained her case, the Gypsy woman laughed and said she would help her. Some days later she knew that her husband’s lover had to go to the hospital . . . Then the Gypsy woman came back. ‘‘Do you want to stop suffering?’’ she said; ‘‘I can help you, then you have to take care of me.’’ With time Madalena forgot about that event. Some years latter, she had four children and her husband had abandoned her definitely. Her mother was sick and she had to take her to a hospital in Salvador, but had no money. Then the Gypsy woman appeared and said Downloaded by [ROGER SANSI] at 09:11 14 December 2012 Images in Candomblé 25 somebody would show up and would help her. Indeed, in the train station she found a man who gave her money. They became lovers later on. After this the zigana started to ‘‘seize’’ (pegar) her, and Madalena started to organize public celebrations for her. She talked to people and explained her story, and thus learned about the zigana: that her name was Sara and she had come from Egypt. Sara’s celebrations are very elegant: she smokes cigarettes with a long filter and drinks champagne. She does not like Candomblé music either, but listens to Gypsy music on the radio. Sara attracts a lot of men, because of her sophistication; but Madalena’s actual husband does not care much about it, since he knows that it is not Madalena but Sara who seduces men. One day, Madalena told me that she wanted to have an image of Sara. But she couldn’t find one at the markets of images. She wanted to ask one of the local craftsmen who make wooden sculptures of the Orixás to make a little image of Sara. She showed me a picture of her embodying Sara. She had debonair looks, was holding a cup of champagne and smoking a cigarette with a long filter. I don’t remember exactly how it started, but I ended making a drawing, modeled on the photography, of a Gypsy woman holding a cup of champagne and smoking like Madalena’s spirit. Madalena was amused with my drawing abilities, and surprised at the same time: she recognized Sara as she saw her in dreams, so she decided to send the drawing to a craftsman to produce a three-dimensional copy. The image of Sara is always in process, we could say: it is the result of several encounters, in dreams, revelations, incorporations, photographs, and a myriad of events that perform this image, and refine its profile, at the same time that her history becomes more elaborate, and the relationship between Sara and Madalena grows closer, more intimate. Eventually they will end up becoming the image of one another. CANDOMBLÉ AND CATHOLIC ALTARS However, I insist, these images are not central in Candomblé shrines. It is important to point out differences between these shrines and Catholic altars. The relationship of devotees with Catholic altars is in many ways different from with Candomblé altars, and their ways of seeing [Morgan 2005] are radically different (at least in principle). First, although offerings can be made to the Catholic altar with candles or flowers, these are not normally viewed as ‘‘feeding the ‘santo,‘’’ as is explicitly said in Candomblé. Secondly, the Catholic altar is eminently disposed to be seen, looked at and admired. The viewing of the Candomblé altar [Figure 1] is restricted, although it is also supposed to be overwhelmingly beautiful. The devotees have to prostrate in front of it in the offering, showing their submission, and never looking at it directly. The life of the shrine does not come from the lifelike quality of images, as we could argue for the Catholic image. The Candomblé altar is hidden and veiled and its life is a latent mystery, a hidden breath. And the offering can cause physical effects, because it is like a ‘‘tuning in’’ to the spiritual channels that bring the Orixás down to earth. The offering awakens the living force of the altar, the axé, causing the trance and possession of the devotees. While Catholic images seek ‘‘likeness,’’ the Downloaded by [ROGER SANSI] at 09:11 14 December 2012 26 R. Sansi Candomblé shrine seeks a hidden, inchoate ‘‘presence,’’ that becomes actualized through possession. I take the terms likeness and presence from Hans Belting’s homonymous work [1997]. He used these two terms to define the different ways of assessing the ‘‘life’’ of paintings in Late Antiquity and the Middle Ages. In terms of ‘‘presence,’’ the sacred art of Late Antiquity was conceived more as a ‘‘relic’’ of the sacred than as a representation: images of the saints were reproduced exactly as they were left, and they were supposed to reverberate sanctity by contagion with the sacred. The paintings that looked for ‘‘likeness,’’ on the other hand, would not be conceived as relics or as sacred objects in themselves, but as views, ‘‘windows’’ to the sacred, reproducing scenes in the Bible or in the lives of saints, which would move the imagination of the viewer; less as relics, more as texts. The move toward a more figurative and ‘‘objective’’ art in the early Renaissance is also a movement of subjectivization and internalization of faith, then. If Catholic altars in everyday life in Cachoeira are only based on ‘‘likeness,’’ or if there could be some aspiration of ‘‘presence’’ of the saint in the Catholic altar, could be a subject for discussion, but that cannot be addressed here. PHOTOGRAPHY AND VIDEO As mentioned, Candomblé has a problem with images at another level too: the visual reproduction of rituals, events and altars, that is to say, taking pictures or videos of sacred rituals or objects. Going to Candomblé houses visitors are often told that taking pictures or videos is not allowed. It is explained in different ways. Manny Vega, a Puerto-Rican artist who got initiated into Candomblé at the Gantois in Bahia once told me that Candomblé is like Jazz. You can’t represent the Orixás in pictures: they are dynamic entities, pure energy, and their image is the moving body of the initiate, incorporating the santo. Although the ritual of possession is structured and routinized, the dance of possession always includes unexpected turns, swirls and improvisations: unexpected things happen at a pace that is difficult to follow for those who are not acquainted with the Orixás. They can’t be reproduced; you have to see them live. According to the anthropologist Patricia de Aquino, the Orixás can’t be ‘‘freeze-framed’’ [2002]; that is, they are dynamic entities that cannot be reproduced in static images. What I found In Madalena’s house however contradicted all this subtle theorization about the impossibility of taking pictures of Orixás. Madalena had an extensive photo album and home videos of the ‘‘festas’’ she had celebrated to her Orixás. Expanding on what Castillo has already said [2008], perhaps the issue with photography is not so much the technique in itself but the control of its outcomes: the issue is partially about who takes the pictures, and for what purpose. The reason to restrict the visual reproduction has less to do with the aesthetics of Candomblé and more with notions of personhood. Photos are obviously indexes. Our image is a part of ourselves, and of course we don’t feel comfortable giving it away to outsiders. The fact is that in the house I was doing my fieldwork in with Madalena, pictures and videos of the rituals were regularly taken and were used Downloaded by [ROGER SANSI] at 09:11 14 December 2012 Images in Candomblé 27 privately in family photo albums and home videos. Of course visitors, outsiders or tourists taking pictures of rituals will not always be welcome. That is true for a Candomblé ritual as it is true for any ritual in our private life—a wedding, baptism, etc. What would we do if some unknown people start taking pictures at our wedding, because they think it’s exotic? I don’t want to deny theories of the impossibility of ‘‘freeze framing’’ the gods, but in the rejection of picture-taking, there is also a component of privacy that is very important. In more restricted terms this involves the issue of sorcery, or feitiço. The concept of a ‘‘volt sorcery’’ exercised through objects that stem from the person is common in Candomblé circles. Photos can fall under this category, as indexes of the person. It would take much longer to argue if this notion of sorcery is exclusive to Candomblé or is a wider belief in Brazilian society, with multiple historical roots [Pares and Sansi 2011]. On the other hand, people in Candomblé are quite cautious about discussing sorcery, precisely because of the wider prejudice that has been addressed specifically against them as sorcerers (feiticeiros). In general terms however I think we can argue that the concerns with intimacy and sorcery are very often coextensive, not just in Candomblé or in Brazilian society: the question is, who will use the image, what for, and to what extent can these images be considered a part of the distributed self? Madalena herself would never allow a picture of her to be in the hands of certain people who could harm her. In this sense, the reproduction of images does not imply that these images lose their power, but quite the opposite [Meyer 2008]. But if the reproduction and distribution of images is controlled by the house, the ‘‘danger’’ of these reproductions seems to be less urgent. Madalena’s record of the house’s rituals was basically a family photo album, and further than that, as I will explain later, a documentation of the activities of the house. It included the initiation of her filhas de santo, the annual rituals to the different gods, etc. The changes and the progress in the house were made evident in the pictures; and the photo album was a witness to the fact that the house was performing the rituals correctly. The videos, for Madalena, had a more shocking power. The first time she saw herself incorporating the Caboclo (Indian) spirit, she was stunned. At her house viewing the videos has become, more than an amusement, a formative experience. The iawos (initiates) regularly see the videos of past rituals. Seeing the videos they can identify details they had not perceived before. And more importantly they can get to grips with the fact that their own bodies are temporarily occupied by other people, by becoming familiar with these other people—seeing them on video or pictures. They are used, in other terms, as ‘‘personal memory banks’’ [Morris 2002; Van de Port 2005]. But which kind of memories are we talking about, if the person who is watching herself before has no recollection of the event? If she is seeing herself (her body) being somebody else? Commenting on the production of home videos in Candomblé houses, Van de Port [2005] has remarked that the introduction of these new media has not substantially affected the ritual structures of Candomblé so far: the videos reproduce the public side of Candomblé, the festas, concentrating on the performances and secondly also on the foods given to those who attend the performance; what is already the public image of Candomblé anyway. But Van de Port also argues that the form in which these videos are edited demonstrates that there are different Downloaded by [ROGER SANSI] at 09:11 14 December 2012 28 R. Sansi sources of ‘‘authentication’’ (or legitimacy), by adopting an editing style that is akin to TV, modes that are a part of the everyday life of most people in Brazil—essentially soap operas. This makes perfect sense with the general history of appropriation in Afro-Brazilian religions, which have been consistently borrowing models and styles from popular culture since their origins: in a similar way that Catholic images were borrowed in Candomblé not just because of religious syncretism but because Catholicism was hegemonic in Colonial and Imperial Brazil. When the public imagination started to be populated with other materials, like movies and TV, new characters, objects and images appeared also inside Candomblé houses. Just to give an example: Po Branca, Erê (child spirit) of Helena’s Caboclo, and allegedly the Caboclo son, is a menino da rua from Rio, dressed up in the clothes of a TV cartoon character. Van de Port is right in pointing out the clear televisual influences of these videos [2005]. In some of the more heavily edited videos that one can find in the markets in Brazil.4 What is more striking is the juxtaposition of different planes (up to three, but mainly two) bringing together the ritual plane with images of ‘‘Nature.’’ What we have is a clearly surrealistIc style, using the juxtaposition of human and natural events, in what André Breton defined as objective chance, as an image of the sacred. In fact this style is in perfect coherence with the juxtaposition that characterizes altars in Candomblé. Curiously enough, if there is a style that these videos seem to be based on, more than telenovelas, as Van der Port mentions, it would be the style of evangelical TV programs, which often use effects of juxtaposition, between holy or hellish landscapes and the Mass. As in the Candomblé videos, this juxtaposition can be described as a transposition of a miracle into filmic images. In any case many of these home videos, for example in Madalena’s house, are actually not edited at all, in a televisual style or any other. What is important for her is that these images keep a record of the past. The question is what past, which events are being recorded, and to what extent do these records manage to encapsulate the miraculous condition of some of these events? More than the ways into which Candomblé is represented, what is important for them is the presence of the sacred. Through photographs and videos people can see an image of their santo they had not seen before: in their own bodies. DOCUMENTS AND EVENTS For Madalena, pictures and video are documents of her house that can testify to her activities: they can testify that these things really happened there; that she was a real mãe de santo and that she was celebrating luscious festivals where the gods were descending on their initiates’ bodies. The pictures and videos are proof of the truth of the rituals. When I asked Madalena for images of her house, that was not a problem for her, and she allowed me to make copies of her pictures and videos. For she had understood that, in a way, that would also help legitimize her house in Cachoeira, or in other terms I would project her fame, or distributed person. I also wanted to take some pictures of the house myself, and from the very beginning offered to give her copies. Madalena described then Images in Candomblé 29 Downloaded by [ROGER SANSI] at 09:11 14 December 2012 the outcome of my research as a Livro de Actas, a Register. Before opening the Candomblé house, she had worked for a long time at the local archive, so was familiar with Registers. She thought about a book where I would explain the history of her house, with descriptions of the initiations and festas that took place there, together with pictures. I would be like a notary who would testify that all these things really did happen. In part that is what I have ended up doing, in my dissertation and then my book, at least in the first chapters. But what I could give her on the spot were the pictures. One day she decided to make a photo-shoot. She dressed up in her Ketu initiation clothes and ordered Dona Dete, her mãe pequena (second-in-command) and Mama, her daughter and ekede, to do the same, dress up in their full Ketu attire. Ketu is the ritual tradition of the main houses of Salvador, and also Dona Figure 2 Madalena fully attired as a mãe de santo. (Photo # Roger Sansi; color figure available online) Downloaded by [ROGER SANSI] at 09:11 14 December 2012 30 R. Sansi Baratinha’s, next to hers. Shortly before she had gone through the ritual of Deca, which entitled her to initiate in the Ketu tradition. Ketu is the dominant tradition of Candomblé in Bahia, led by houses like Casa Branca or Opô Afonjá. In recent years the federation of Afro-Bahian cults has put pressure on Candomblé houses to follow the Ketu tradition, presented by them as the right and legitimate form of Candomblé. She hung the picture next to the certificate of the Federation of Afro-Bahian Cults that allowed her to exercise as a Candomblé priest. Her original ‘‘nation’’ was Angola, although she is very much a self-made mãe de santo. She had initiated in Ketu because of the increasing pressure of the local Candomblé elites, but was still practising most of her festas in the Angolan ‘‘line.’’ I took pictures of them in the main public room, the hall. She liked in particular one picture with her implements for initiation, and decided to hang it on the wall of the main room. According to Dona Dete, one could see Lansã through that picture; only in that one, not in the other pictures, she said. Lansã is Madalena’s main Orixá; something had happened in the picture—an image of Lansã had emerged [Figure 2]. Some years later I went with Madalena to visit a pai de santo, a priest in a neighboring town. This pai de santo, like Madalena, had not been initiated; he was a rezador,5 and he incorporated a Caboclo spirit, a Native American spirit. He had asked Madalena to initiate him in Ketu Candomblé, which she did. Madalena became then his mãe de santo, his priestess. On this occasion, when we were leaving the house, Madalena asked him for a picture of him, to hang in her barracao. He showed her a couple of pictures. Madalena looked amused at one of them, saying that one could see his Oxum in the picture. Oxum is one of the Orixás of this father, but not the main one; that is Ogum. But in that picture Madalena recognized a shade of Oxum, as it were—her literal words were da pra ver Oxum encostada, ‘‘one can see Oxum encostada’’; an encosto being the presence of a dead person in one’s body. It is a term more widely used in the context of sorcery; but in this case it meant something like that there was a presence of Oxum in the picture, without being fully incorporated into the pai de santo’s body. CONCLUSION: IMAGES AND PERSONS To conclude, I think that the rejection of images, either in the form of Catholic images in Candomblé shrines or the reproduction of video and photography in rituals, stems from a similar, if perhaps inverted, concern with the question of representation. The movement of ‘‘re-africanization’’ in Candomblé has not just been a movement of ‘‘purification’’ of practices but one of ‘‘purification’’ of semiotic ideologies, as Webb Keane would say [2007], in which an intellectualized elite of Candomblé practitioners has incorporated certain theories of representation, and of the relation of objects and images with people. On the one hand, Catholic images are rejected because they are seen as ‘‘representations’’ of Catholicism, and therefore contradictory to a purely African religion. What this perception of Catholic images as representations of Catholicism seems to ignore is the fact that in religious practice these images are re-appropriated, and transformed into indexes of everyday practice in Candomblé, in ways that may be totally subversive from the official Catholic representation. On the other Downloaded by [ROGER SANSI] at 09:11 14 December 2012 Images in Candomblé 31 hand, the prohibition on the reproduction of rituals through the use of video and photography seems to revert to a notion of ‘‘presence,’’ more than representation: the reproduction of this presence through mechanical means would be perceived as implicitly contradictory with its ephemeral nature in rituals of possession. And yet what is interesting is that often perhaps the opposite is true: it is precisely in and through certain reproduced images, like photos, that certain unprecedented images of the santo appear. People recognize in such photos images that they had not recognized while a picture was being taken. It is clear that the issue is, more than the total prohibition of the mechanical reproduction of images, the control over distribution. This control is a control over the ‘‘aura’’ of images, to follow the Benjaminian terminology, or over their axé, to follow its Candomblé equivalent. In any case it is clear that the problem is not the technology itself, the possibility of reproducing images of the ephemeral, but its uses. In this sense, images can be ‘‘representations,’’ yes, but in a wider sense than just as texts that encode a certain message—as ‘‘ambassadors,’’ as Gell [1998] has mentioned, extensions or tokens of a certain power. Although in theory the Orixás do not like to be ‘‘freeze-framed,’’ in practice Candomblé practitioners are using the reproduction of their images as a means of documentation, education and legitimacy, often to legitimize or testify for one or another form of truth. In this sense, the mystique of the image as a document is not exclusive or particular to these Candomblé practitioners but takes a very particular shape, I think, in these contexts; as when Dona Dete recognized in that picture—and in only that picture—the sight of Lansã, as an elusive presence. The problem in general is that images can become not only representations but persons. The war of images is not only a war of or for representations but for distributed persons, distributed bodies. In this sense their dynamism is of a more complex kind than just ‘‘freeze-framing.’’ What is more interesting in these images is the dynamic ambiguity, not only between presence and representation but between object and subject: who is represented in these images? the spirit or the person? Madalena or Lansã?—both, at the same time. That ambiguity, that unstable superposition, is what makes these images more powerful and challenging for the people of Candomblé themselves, because it embodies their greatest secret, their greatest truth also: that eventually, the spirit and the person are to become one and the same, in the santo. ACKNOWLEDGMENT Mãe Madalena, the main character of this story, died (August 4, 2012) while this article was being edited. The author dedicates the article to her memory: hopefully these pages will be yet another powerful refraction of her distributed person, beyond death. NOTES 1. This definition of the text as a fixation of representations is perhaps unfair, but it’s the central idea that the structuralist and post-structuralist discourse has used in looking at texts as objects essentially different from subjects. In the last few years linguistic 32 Downloaded by [ROGER SANSI] at 09:11 14 December 2012 2. 3. 4. 5. R. Sansi anthropology has taken a radically different point of view, looking at the ‘‘text’’ as an active and participative social presence (co-text), or in other terms, a (possible) subject. This notion of the text is close to our notion of the image as person [Silverstein and Urban 1996]. In this sense W. J. T. Mitchell prefers to replace the term ‘‘body’’ by ‘‘object’’: that would define the material support of the image, while the medium would be the set of material practices that brings together object and image. The assemblage of object and image through the medium is what Mitchell defines as the picture, ‘‘the entire situation in which an image has made its appearance’’ [Mitchell 2005: XIII]. Although his distinction between image and picture is interesting and pertinent, as he himself recognizes, the picture in this model is an ‘‘interruption’’ in the process of the becoming of images; an important interruption—the moment of objectification of the process—but an interruption after all, that will be overcome by the process at some point. In this article I have preferred not to introduce this distinction because it would make the argument too difficult to follow and somewhat unnecessary for my argument, one that emphasizes the dynamism of images; and also because in fact the distinction between image and picture does not have a direct translation into Portuguese. The notion that altars are the result of processes of gift exchange has been also underscored in the case of Candomblé’s twin religion, Santeria, by Brown [1996]. As for example the videos of Pia Luiz de Lansã. Actually they can also be seen on YouTube: http://www.youtube.com/watch?v=0oM4G6RpKGs (accessed October 31, 2012). A Catholic layman with a gift for prayer, often also a specialist in the use of plants. REFERENCES Aquino, Patricia de 2002 No freeze-frame on Gods. In Iconoclash. Beyond the Image Wars in Science, Religion, and Art. Bruno Latour and Peter Weibel, eds. Pp. 234–239. Karlsruhe: 2 km=Center for Art and Media; Cambridge, MA: MIT Press. Barthes, Roland 1981 Camera Lucida. New York: Hill and Wang. Belting, Hans 1997 Likeness and Presence: A History of the Image before the Era of Art. Chicago: University of Chicago Press. 2005 Image, Medium, Body: A New Approach to Iconology. Critical Inquiry, 31(Winter): 302–319. Brown, David 1996 Altar Making and Gift exchange. In Santerı´a Esthetics in Contemporary Latin American Art. Arthur Lindsay, ed. Pp. 77–146. Washington: Smithsonian Institution Press. Capone, Stefania 1998 La quête de l’Afrique dans le Candomblé. Paris: Karthala. Castillo, Lisa Earl 2008 Entre a oralidade e a escrita: A etnografia nos Candomblés da Bahia. Salvador: Edufba. Gell, Alfred 1998 Art and Agency. London: Clarendon Press. 2007 Christian Moderns. Berkeley: University of California Press. Keane, Webb 2008 On the Materiality of Religion. Material Religion, 4(2): 230–231. Latour, Bruno 2002 What is Iconoclash? Or is there a World beyond the Image Wars? In Iconoclash. Beyond the Image Wars in Science, Religion, and Art. Bruno Latour and Peter Weibel, eds. Pp. 14–18. Karlsruhe: 2 km=Center for Art and Media; Cambridge, MA: MIT Press. Downloaded by [ROGER SANSI] at 09:11 14 December 2012 Images in Candomblé 33 Meyer, Birgit 2008 Powerful Pictures: Popular Christian Aesthetics in Southern Ghana. Journal of the American Academy of Religion, 76(1): 82–110. Mitchell W. J. T. 2005 What Do Pictures really Want? The Lives and Loves of Images. Chicago: University of Chicago Press. Morgan, David 2005 The Sacred Gaze: Religious Visual Culture in Theory and Practice. Berkeley: University of California Press. Morris, Rosalind C. 2002 A Room with a Voice: Mediation and Mediumship in Thailand’s Information Age. In Media Worlds: Anthropology on New Terrain. Faye D. Ginsburg, Lila Abu-Lughod and Brian Larkin, eds. Pp. 383–399. Berkeley: University of California Press. Parés, Luis Nicolau, and Roger Sansi (eds.) 2011 Sorcery in the Black Atlantic. Chicago: University of Chicago Press. Peirce, Charles Sanders 1984 Writings of Charles S. Peirce: A Chronological Edition. 1867–1871. Cambridge, MA: Harvard University Press. Rancière, Jacques 2007 The Future of the Image. London: Verso. Sansi, Roger 2007 Fetishes and Monuments. Afro-Brazilian Art and Culture in the 20th Century. New York: Berghahn Books. Santos, Deoscredes Maximiliano dos (Mestre Didi) 1967 West African Sacred Art and Ritual in Brazil, a Comparative Study. Ibadan, Nigeria; manuscript. Silva, Vagner Gonçalves da 1995 Orixás da Metropole. Petropolis: Vozes Silverstein, Michael, and Greg Urban 1996 Natural Histories of Discourse. Chicago: University of Chicago Press. Thompson, Robert Farris 1993 Face of the Gods: Arts and Altars from Africa and the Americas. New York: Museum for African Art. Van de Port, Mattijs 2005 Visualizing the Sacred: Video Technology, ‘‘Televisual’’ Style, and the Religious Imagination in Bahian Candomblé. American Ethnologist, 33(3): 444–461.