Festivals and other public events that feature indigenous dance performances are a burgeoning phenomenon, both in Australia and in the international arena. My aim here is to trace the field of power relations in which such festivals are embedded and within which they are constituted. Festivals make for fascinating study because they present spatially and temporally contained domains in which the performative aspects of human social relations and sensual embodied expressions of social practice can be directly observed and experienced. Yet festivals are only apparently contained events. Although participants are meant to experience a festival for the term of its duration as a whole, self-contained world, it is only a partial world. Boundaries between a particular festival event and the social order are highly permeable. Festival performances reach out beyond the festivals into the everyday world, and they can be fully understood only with reference to wider social situations and the political, economic, and social interests and state processes and practices that, in fact, produce them. In addition, the state deceptively asserts its presence within the festivals. Indeed, agents and agencies of the state colonize the festivals, so that the festivals become prime sites for recognition of the effects of the state (Trouillot 2001: 126).

     This paper focuses on two contemporary festivals in northern Australia, one from Cape York and the other from northeast Arnhem Land. I explore the nature of these festivals as public events and investigate their significance as a form of political practice in relation to the presence of the Australian state in Indigenous lives. More broadly, my research concerns the role of festivals in articulating Indigenous rights discourse within Australia and how Indigenous groups deal with the state through harnessing the power of performance.

     Dance and song are significant forms of symbolic capital. Live performances provide people not only with an avenue for presenting culture as spectacle but also with a means of political engagement, or performative dialogue, with others. Public festivals in Australia are complex sites of national identity formation (J. Kapferer 1996). Thus, festival performances must be read not as static representations of culture but as dynamic strategies of power. Similarly, the Indigenous festival phenomenon should not be dismissed as a mere expression of identity culturalism or as a traditionalist revival of a long-dead past. Festivals provide opportunities for Indigenous people to retrieve tradition so as to put it to work to make sense of history and to negotiate the processes and practices of state power as it expresses itself in Australia today.

Colonization and Folklorization

It might be argued that contemporary, state-sponsored festival performances are just another expression of the folklorization of Indigenous peoples that goes hand in hand with their continuing colonization (Balme 1998, 2007; Rogers 1998). During former decades of European settlement in Australia (as elsewhere), colonial agents, missionaries, and others perceived Indigenous ritual and ceremonial activity as both a political and a moral threat. Such practices became acceptable to these newcomers only if they were rendered innocuous through theatricalization and revalued in terms of European desires and fantasies. Yet the idea of performing for colonial audiences appears to have been readily accepted by Aboriginal people as a means of communicating with the newcomers and having some control over their relationships with them. According to Parsons (1997: 46), public performances of dances and music by Aboriginal people emerged as a "cultural product" in the nineteenth century. He identifies four major kinds of performance of this type: 1. the "peace corroboree," staged to mark a new state of cooperative relations between Aboriginal people and the Crown; 2. the "command performance," for official state visitors; 3. the "gala corroboree," to mark social occasions significant to the settler populace, such as charity sporting events and annual agricultural shows; and 4. the "touristic corroboree" (ibid.).

     Indigenous involvement in these kinds of performance could be read as merely the response of a powerless people to Western demands for representations of the 'primitive other.' Yet, although they evidence asymmetry in power relations, I suggest that such performances allow Indigenous people to engage with settler Australians, if not entirely on their own terms, at least with some sense of agency and control (cf. B. Kapferer 1995a, 1995b). This has been the case since the early period of European contact. For example, James Morrill, one of a group of individuals who survived the wreck of the ship Peruvian in 1846 on the mid-north coast of Queensland, describes a "grand corroboree" in which his party was induced to participate. It was staged "by the original group of Aborigines who had encountered the survivors for the benefit of a gathering of various regional clans" (Hayward 2001: 4). Morrill writes:

The first thing they did was lay us down and cover us over with dried grass, to prevent our being seen till an appointed time. They then collected from all quarters to the number of about fifty or sixty—men, women and children—and sat down in a circle; those who discovered us stepped into the centre, dressed up in our clothes, with a little extra paint, danced one of their dances, at the same time haranguing all present, recounting how they discovered us . . . from whence they had brought us, and all they knew about us . . . and then as a finale we were uncovered, and led forth into the centre in triumph. (Cited in Haywood ibid.)

     Performances such as that documented by Morrill in reaction to first encounters with Europeans were the responses of reflexive, adaptive, creative people dealing performatively with powerful new forces in their lives. The agency of Aboriginal people is similarly evident in the present context. Indigenous performances should not be read top down as merely complicit responses to colonial agendas (cf. Magowan 2000: 310). For example, perhaps one of the most enduring and powerfully symbolic images of the Indigenous fight for rights and recognition within Australia today is that of a Wik woman dancing her victory dance outside the Australian High Court in 1996 after the landmark case in which the court ruled that pastoral leases did not necessarily extinguish native title. As Magowan argues, through her dance, Typingoompa substantiated her "claim to indigenous rights and her authority to dance for land" (ibid.: 312).

     Indigenous Australians have on numerous occasions attempted to harness the power of performance by drawing politicians and government bureaucrats into the performance events as participants. For example, in 1997 on Elcho Island, Yolngu clan leaders led the prime minister of Australia, John Howard, through a secret part of a ritual, and Wik women danced on the lawns of the Parliament House with Senator Harradine, who at the time held the balance in the Senate over the Wik native title legislation. According to Magowan, "The effect of leading the prime minister through the ritual was to place him in a particular performative dialogue, one that was bound by Yolngu models of political control and authority embedded in the corporeal dispositions of the dances" (ibid.: 318).

     Similarly, on December 17, 2004, Djabugay elder Enid Boyle danced at the official event organized to mark state recognition of Djabugay native title over the Barron Gorge National Park. In each of these cases, the Aboriginal participants actively sought to harness the poetic power of performance as an expression of their own agency and that of their people, to mark their presence in relation to the state, and to secure recognition of continuing connection and entitlement to land in the face of a history of dispossession.

     Anthropologists and other scholars have long debated conceptual oppositions between tradition and history and, in relation to Australian Aboriginal people, "the issue of the supposed absence of historical consciousness among traditionally oriented Aborigines" (Beckett 1994: 97; see also Kolig 1995, 2000; Merlan 1994; D. Rose 1984; Rumsey 1994; Swain, 1994; Urry 1980). It has been argued that "traditionally oriented" Aboriginal people frame the past in terms of mythic thought rather than history. However, as Beckett (1994: 99) writes, "We should not scorn to look at humbler, less structured forms such as jokes, songs, genealogies, stories" as ways in which Aboriginal people articulate memory and engage with history. Similarly, I argue that festival performances can constitute a form of strategic communication of historical consciousness.

Disinheritance and Cultural Revival

Many Indigenous Australians today express a great sense of loss over disinheritance, not only of their land but also of the cultural practices and performances that linked them to land and to one another. They mourn the cultural dispossession they see as resulting from their forcible removal, dispersal, and confinement to reserves and missions after European colonial expansion into northern Australia. As an Aboriginal woman noted about her life on the Mona Mona mission,

When I was growing up we had . . . most of our corroborees . . . but the missionaries when they heard the clapstick, you know, they didn't like to hear that . . . They sort of cut it out and it gradually died out then until now this generation is trying to revive it. (Florence Williams, interview July 1994)

Cultural revival" became a catchphrase during the 1970s and 1980s among Indigenous peoples in northern Australia and elsewhere. During the early 1970s, the Aboriginal Theatre Foundation was established, with encouragement from the Australian Council for the Arts, "to preserve, to restore and to sponsor Aboriginal dancing and singing" (von Sturmer 1973: 2). The foundation was officially incorporated in 1970, with a National Executive Committee (mainly composed of non-Indigenous members) that met three times a year. The foundation provided an avenue for Indigenous peoples to address an increasing sense of cultural disinheritance and a very real fear that they were indeed 'doomed' to extinction—that is, cultural extinction, not racial extinction as the 'doomed race' theorists had assumed (McGregor 1997). One of the two Indigenous members of the original committee, from the Northern Territory, had this to say about the foundation:

I am deeply worried that our young people could forget our culture. It is our sacred song and dance that expresses, and is the root of, our law and our discipline. . . . I believe that the Aboriginal Theatre Foundation can help us to make our culture live forever. I also believe it is just as important that the Foundation help non-Aboriginal people to understand how deep and important is the meaning of this culture, through dance and song. (Aboriginal Theatre Foundation 1972: 15)

Thus, Indigenous people sought new means to express and transmit memories of cultural beliefs and practices that they feared would soon be forgotten and that they believed to have been in existence at "the threshold of European colonisation" (Keen 2003).

Authenticity and the Invention of Tradition

Cultural revival in Australia, as elsewhere, has been evaluated as being less about revival than about the fabrication or invention of culture. These ideas can be linked to works focusing on the politics of identity, as well as to debates over the idea of the past as 'constructed' in the present and of tradition and/or heritage as mere invention (see Beckett 1988; Friedman 1992; Handler and Linnekin 1984; Herzfeld 1982; Hobsbawm and Ranger 1983; Linnekin 1991; Merlan 2000; Tonkinson 1997).

     In accord with constructivist concepts of tradition as something that has been invented are assertions that dances staged for tourists are mere concoctions aimed at the tourist dollar. It has been argued that Indigenous Australian dance festivals, the hula in Hawaii, and other such performances are not authentic expressions of traditional culture but are, in fact, products of colonialism. Yet, as Sahlins argues with regard to the hula,

[T]he hula as a sign of Hawaiianess, of the indigenous, was not born yesterday nor merely as the construction of the Hawaiian Visitors Bureau and prurient Haole interests. The hula has been functioning as a mode of cultural co-optation for more than 150 years—a significance, moreover, that was already inscribed in the meanings of hula performances before the first White men set foot in the islands (Sahlins 1993: 8).

Sahlins's point is that, although the hula, as part of the current Hawaiian renaissance, may be an 'invention,' it is not a colonial invention. It is not "a Western fabrication of Hawaiianess," nor is it "an Hawaiian fabrication in response to the West" (ibid., 11). As Sahlins (ibid.: 13) writes, "We are not dealing with people who have nothing and are nothing." According to Sahlins, innovations follow logically from the people's own principles of existence. They are not simply imposed from outside by the colonial order or by the commercial forces of a global economy. Like hula, contemporary Aboriginal dance festivals are not an invention divorced from cultural forms and historical contexts.

     The Aboriginal cultural renaissance is clearly not just state sponsored and generated for the tourist industry. It is a political response by Indigenous people, an attempt to control their relationships with the state. An ethnographic study of festivals celebrating Indigenous cultural renewal reveals the role that dance performances play in a fraught political arena in which the criteria for indigeneity are continuously being contested and negotiated.

Cultural Festivals

I provide here a comparative account of two Indigenous Australian cultural festivals held in the north of the country: the Garma Festival, which is staged on an annual basis by Yolngu clans from northeast Arnhem Land, and the Laura Aboriginal Dance and Cultural Festival, which is held biennially in Cape York (cf. Henry 2000a, 2000b, 2002). Both festivals are relatively recent institutions. The Garma Festival was first held in 1999, while the roots of the Laura Festival in Cape York can be traced to 1972, when a regional festival was sponsored by the Aboriginal Theatre Foundation and hosted by Aboriginal people from Lockhart River, Cape York.

     Some festivals in Australia are staged by Indigenous people who are familiar with one another's practices and who have relationships in the lived-in-world beyond the event. This is the case with the Garma Festival. By contrast, the Laura Festival brings together many unrelated and culturally distinctive peoples from all over Cape York Peninsula. Although similar in substance, the festivals are different both in flavor and in terms of the "logics of their design" (Handelman 1990: 7). These differences reflect not only the cultural differences of Cape York and Arnhem Land peoples but also their different histories and experiences of European colonialism.

The Laura Festival, Cape York

The pragmatics of organization of the first Cape York Festival, at Lockhart River, necessitated some rapid adjustments in performance practices by the participants. The festival, as a newly forged intercultural space, brought together distant groups to perform in the presence of one another for the first time, and this led to some disquiet and debate concerning the ritual protocols that should be followed.

     For example, von Sturmer (1973: 4) writes that some of the dance leaders were concerned about "the intermingling of sacred (though public) and secular dances" at the festival and were disturbed by "the failure to carry out the proper ritual procedure which should follow the introduction of unfamiliar and powerful dances." According to von Sturmer (ibid.), it took many behind-the-scene consultations among various groups before correct protocols were thrashed out and the dancing went ahead.

     Because these early festivals were envisioned as a means of 'cultural revival' of 'traditional' song and dance, they generated lively debate and competition among groups concerning the relative traditionality of the performances. In this politics of knowledge, as Chase (1980: 419) observed, "whoever had the greatest range of 'old fashion' dance was thought to be the most successful."

     The Lockhart River Festival spawned the idea for an annual Cape York Festival, hosted by a different community each year and supported by state government funding. It was held in a number of different Cape York Aboriginal communities, which had been established during the mission and reserve era, before a permanent site was chosen. The festival is now held biennially near the tiny town of Laura. While allowing Cape York peoples to celebrate their social and cultural differences, the festival fosters connections that have contributed to an emerging regional Cape York identity (Chase 1980: 421).

     The festival was initially staged solely for Indigenous people of Cape York Peninsular, not for tourists. The organization of the festival was for many years under the control of the State of Queensland Department of Communities, through a committee of elected Indigenous representatives. It was organized much like a sports carnival, with a competition between dance groups representing their respective former mission and government reserve communities. However, during the 1980s, the festival began to attract an increasing number of domestic and international tourists. To accommodate the demands of the tourist audience for cultural authenticity, the organizing committee decided on a number of rules to be followed by groups wishing to participate. Included among these were that all dance performances "must be properly cultural and traditional," dancing costumes "must be traditional," and "no modern instruments (such as kerosene tins) are to be used" (as related in a document titled "History of the Laura Dance and Cultural Festival," released to the media by festival organizers in 2001).

     There was some discontent among the participants regarding the control and influence of the state ministry and the rules and regulations imposed by the committee. In particular, people objected to the fact that the dance festival was staged as a competition with prizes awarded to the winning dance teams. Some argued that competition was contrary to Aboriginal spirituality. There was disquiet concerning the increasing accommodation of the performers to the demands of the tourist audiences. Initially, 'revivals' of ritually performed dances of the past were showcased, with an accompanying attitude of solemnity and a sense that the participants were being drawn into the presence of the sacred ancestors (and vice versa). Yet with time, and in articulation with the demands of state agencies and domestic and international tourist audiences, the festival became increasingly like a sports festival.

     Among Indigenous groups, there is an apparent hierarchy in terms of which groups can demonstrate a stronger continuity of traditional knowledge and practice, with certain groups recognized as 'more traditional' and, therefore, culturally stronger than others. The criteria upon which this recognition is based include the participation of 'song men' who are able to accompany the dancers in their language and of elders with knowledge of 'remembered' dances that may have been performed in the past in a ritual context. If an unprecedented event occurs (for example, as when a member of a dance team died as a result of a heart attack), the advice of elders from these more traditional groups is often sought, as they are considered to be more knowledgeable about how to handle the contingencies of life in a culturally appropriate way.

     It is interesting to observe the different responses among the audience to the performance capabilities of the dancers. A person sitting next to me at the 1999 Laura Festival complained that he could not take good photos of some groups because the dancers had their backs to him (they had not adapted their performance for the tourist audience). He praised other groups for their virtuosity and obvious audience appeal. Yet, for Indigenous members of the audience, the less showy dances have a symbolic power that the dances of the more tourist-oriented teams lack. They are accompanied by song men singing in an Indigenous language, which signifies that they have closer links to traditional culture. In 1997, the dance team from Kuranda danced for the first time to songs newly composed in Djabugay. They were complimented by an elder from another group who said he was happy to see that they were getting their language back and were now able to produce their own song men. He had felt "sorry" for them at earlier festivals. Here tradition is 'rediscovered' in contemporary innovation. The words of the songs were composed with the help of a linguistic anthropologist, but their source is still believed to be the dreaming ancestors. The anthropologist is thought to be merely the conduit, the tool that enabled the revelation of what was/is always in existence.

	Figure 1. State and the Arts. The Mayi Wunba Group from Kuranda dancing at the Laura Aboriginal Dance and Cultural Festival, 2013. Photograph by Rosita Henry.

The Laura festival ground was handed back under the Aboriginal Land Act of 1992 (Qld) to the Western Kuku Yalanji people in a special ceremony at the 1997 festival. The traditional owners of the area now have control over the running of the festival. In 2003, the dance competition was reintroduced in an attempt to revitalize the festival by increasing audience satisfaction. Some performers expressed discomfort over this development, suggesting that it was culturally inappropriate that the performances should be staged in terms of a competition. Others disagreed, arguing that such competition is not alien to Aboriginal tradition and that it serves to enhance virtuosity.

     While the festival organization is in the hands of the traditional owners, the state's presence at the festival remains strong. Various bureaucratic welfare and other state agencies, including the National Native Title Tribunal and numbers of Aboriginal representative bodies, promote their products and services at the festival. The Laura Festival is as much a competition for Aboriginal clients by the welfare state and by state agents—referred to by Collmann (1988) as "boundary riders" or "brokers" (including lawyers and anthropologists)—as it is a dance competition. Thus, the state reproduces itself through the festival, sustaining itself by feeding upon the very Aboriginal life world that it claims to support.

The Garma Festival, Arnhem Land

The Garma Festival, at a place called Gulkula near Nhulunbuy, is an annual event hosted by the Yolngu, indigenous people of northeast Arnhem Land. The festival is organized through the Yothu Yindi Foundation (2003), a nonprofit charitable organization that was established in 1990 by representatives of five of the Yolngu clans—Gumatj, Rirratjingu, Djapu, Galpu, and Wanguri—to "support and further the maintenance, development, teaching and enterprise potential of Yolngu cultural life." The foundation is working collaboratively with a number of Australian universities to develop the Garma Cultural Studies Institute to facilitate sharing of Yolngu and Western knowledge. It has also established a Music Development Centre to support local songwriters and musicians and to facilitate the documentation of traditional song cycles. The Garma Festival is funded partly through various state and federal government bodies and programs, as well as through sponsorship of a number of nongovernmental organizations and industry bodies.

     Garma is a Yolngu name/concept. It condenses, or distills, many different but related meanings. We were told at the festival that it "implies balance" between Indigenous and non-Indigenous peoples (see Yunupingu 1993). In other words, it is a place of learning where Yolngu people of different clans, affiliated with that country, come together and where the first (public) stages of initiation are performed. Garma also refers to a public genre of ceremonies that include songs (manikay), accompanied by didjeridu (yidaki, the Yolngu dhuwa moiety name for didjeridu) and clapsticks, and associated dances, painted designs, sand sculptures, and objects (Keen 1994: 138). In other words, it is a concept of a kind of social relationship or a moral principle of how people should relate to one another (that is, in a balanced way), a place, a type of ceremony, and a level of knowledge. It symbolically condenses the link between people, knowledge, and place.

     Unlike the Laura Festival, the majority of people attending the festival are not tourists but invited politicians, government bureaucrats, academics (such as, anthropologists, ethnomusicologists, lawyers, linguists, and other scholars), media personnel, and people linked in various capacities with the Indigenous arts industry and Indigenous education, employment, health, and welfare (for example, community arts coordinators, community development workers, and Indigenous artists). In addition, at the 2003 festival, I observed participants from all over the world who were specifically at the festival for the purpose of attending the yidaki master class (didjeridu students from Germany, Japan, Iceland, and elsewhere). Very few of the participants at the festival could be classed simply as tourists. Yolngu people outnumber all other participants.

     The Garma Festival takes an interesting form. It is predominantly a set of educational workshops and a conference organized around a particular topic and embedded within a festival of Aboriginal dance, music, and other expressive cultural practices. Each year, a different forum or academic program with a specific theme has been organized as an inherent part of (or an umbrella for) the festival. The theme of the first festival was "'Bush University': Natural and Cultural Resource Management" (1999). This was followed by "Gathering of Indigenous Scholars" (2000). The themes of subsequent festivals have concerned a range of social and economic issues of significance to Indigenous Australians, including education and training, law and the criminal justice system, the environment, art and culture, livelihoods, leadership, and health. The theme of the 2013 festival was "Getting People Together," while the 2014 theme is "Responsibility, Reform and Recognition."

     The forums are structured very much like a regular academic conference, with invited keynote speakers and plenary sessions that break up into a number of different panels and/or workshops. It is mostly Yolngu people who lead the sessions and workshops, thereby sharing their concepts and ideas on the relevant theme. Concurrent with the forum sessions are cultural workshops, such as women's basket weaving, painting, woodcarving, spear making, yidaki playing, and tours for gathering bush foods and painting materials. It is thus possible, if one is not a panel member or a speaker in a session, to come and go, to move in and out of the forum proper, and to escape the talkfest, if one is so inclined.

     Celebrated as a means of marrying Yolngu and balanda (non-Indigenous) knowledge systems, the Garma Festival is very much 'entangled' with the state through its links with the tertiary education sector. In particular, the University of Melbourne and Charles Darwin University play a key role in aspects of the festival. At the 1999 Garma Festival, Yolngu leaders prepared a message stick that they sent to invite the vice-chancellors of Australian universities to travel to Gulkula (the Garma Festival site) to attend a Garma ceremony. At this meeting, the Australian Vice-Chancellors' Committee (AV-CC) formed a working party that facilitated a gathering of Indigenous scholars at the 2000 festival, during which the Garma Declaration, a statement about Indigenous higher education in Australia, was developed. The festival and its accompanying academic program are envisioned as being part of a 'bush university' and of an ongoing development of the Garma Cultural Studies Institute, established to "sustain and extend Yolngu intellectual traditions and knowledge systems" and to "develop partnerships and collaborative relationships with places of learning, other Indigenous peoples and the wider community" (Yothu Yindi Foundation 2003: 11).

     The logic of the organization of the Garma Festival is, thus, much like that of a conference or convention. It is a highly orchestrated event, entangled with the tertiary education sector and with high-level state and federal political institutions and industry bodies. A Yolngu participant in one of the forum sessions referred to the Garma Festival as a "window to the nation." It could also be interpreted as a space of governmentality. Trouillot (2001: 127) argues that the state has no "institutional fixity or geographical fixity" and that its materiality resides "in the reworking of processes and relations of power so as to create new spaces for the deployment of power" (cf. Aretxaga 2003; Das and Poole 2004; Foucault 1991; N. Rose 1999). I suggest that one such space is the cultural festival, where effects of the state—in particular, the identification, legibility, and spatialization effects, as described by Trouillot (2001)—are clearly recognizable. Festivals foster the construction of, desire for, and consumption of indigeneity (the identification effect). They are sites where knowledge of, and the means for governance of, indigenous people is produced (the legibility effect) and where social and political boundaries are generated (the spatialization effect).

The Poetic Politics of Dance: A Politics of Knowledge

Dancing is the main focus of the three-day Laura Festival, and different dance teams are scheduled to perform one after another, all day long. At the five-day Garma Festival, however, the dance performances are the climax of the academic/educational activities of each day. The bunggul, a Yolngu performance of dance and song, takes place just before sunset (bunggul is the Yolngu term for dancing but also the generic term for ceremony; see Tamisari 2005: 179). It restates what might have been lost during the day's talkfest, that is, that Yolngu people (of a number of related clans from northeast Arnhem Land) are in charge. It clearly reestablishes the identity of the hosts and the guests at the festival.

     The two festivals, Laura and Garma, bring non-Indigenous participants into an Indigenous space. The remoteness of the festival grounds, out in 'the bush,' means that the participants are captured for the time of the festival. Unlike a city festival, there are no hotels, motels, or lodgings available as a means of escape: visitors are expected to camp at the festival ground. The participants are granted autonomy in choosing the level and nature of their participation. At both festivals, a number of different activities goes on simultaneously, and participants can choose to connect to the festival through any one or more of these. Although the Laura Festival dance performances are more obviously geared to a tourist audience than the Garma bunggul, in both cases the performances are not aimed just at visitors. Constituting instead a performative exchange among Indigenous groups themselves, they evidence a dynamic continuity of social relations and a display of power politics. Although there is an emphasis on the cultural continuity of the performances, culture is not represented as something static or fixed in times past. Rather, culture is performed as a dynamic process. Elders can be observed instructing the junior dancers on the dance ground during their actual performances. The significant point here is that the dances are not necessarily presented as finished products. According to Smith (1997: 60), in her study on Wik ancestral dance, the performances by Wik peoples at festivals such as Laura are a public proclamation of their "continuing link with land, with Ancestors, and with the past." I suggest that such presentations are also a public statement of the continuity of cultural transmission. The performances are as much exhibitions of a process of teaching and learning as they are displays of song and dance routines. Festivals are an opportunity to evidence cultural vitality by publicly performing a process of transmission and acquisition of embodied knowledge. At Garma, the children dance with their elders on the bunggul ground. At the Laura Festival in 1995, for example, there were 170 child performers of a total of 478 dancers, while, in 1997, there were 202 of a total of 467. At more recent festivals, child performers have been in the majority.

     By placing an emphasis on the ritual, ancestral aspects of the performances, Indigenous people claim authenticity in terms of embodied knowledge of the past and of sacred connection with the ancestors and the land. However, virtuosity in dancing is valued as a powerful means of capturing not only the attention of the ancestors but also of the audience. However, if it is allowed to dominate, virtuosity leaves the dancers vulnerable. According to Franca Tamisari (2000: 283), who has written of Yolngu responses to, and understanding of, virtuosity in their dance performances: "As for virtuosity, the full extent of one's power makes one vulnerable. . . . [T]he capacity to act on and change others requires a disposition to be in turn acted upon and changed, or as the Yolngu would say seeing the other's feeling and inner desires is paralleled by being invaded by them." While dancing at public festivals opens up the possibility of acting on the sentiments of the participant audience, including agents of the state, the dancers are also in danger of being invaded and transformed by the inner desires and feelings of that audience.

     The dances are a means not only of continuing the links with ancestors and the past but also of creating connections in the present. Through the festival performances, the dynamics of the relationships between individuals and groups is publicly restated. For example, as the two dance groups were introduced at the 2003 Garma Festival, the announcer (Galarrwuy Yunupingu) noted that they had recently, in the lived-in-world beyond the festival, ritually exchanged colors, so that the group that was dancing under the red flag had at the last festival danced under the yellow flag. (In this way, it was also made clear to the visitors that Yolngu 'business' is alive and well outside the festival context and that the performances in the festival are not just a reconstruction of a past long gone.) Gumbula and De Largy Healy (2004) provide a fascinating account of the complexity of contemporary political relationships between different Yolngu clans attending the festival (cf. Preaud 2005).

     The dance performances can also be read as a commentary on the morality of the colonial encounter and the colonial practices of the European settlers. Examples at the Laura Festival include comic, lighthearted dances, such as one that depicts Aboriginal first encounters with the European bee (which has a sting, unlike the native bee). Another dance mimes the strange antics of European gold prospectors and miners. However, there are also more direct political statements through performance. At the 1997 festival, the Aurukun dancers marched onto the dance ground in T-shirts screen-printed with a map of Cape York identifying Wik territory, while a spokesperson explained the Wik High Court decision on native title (Wik Peoples v. Queensland [1996] 141 ALR 129) to the audience. In the "heterotopic space" (Foucault 1986) of festivals, Indigenous people reconstitute cultural symbols to address contemporary realities and confront their social situation.

     At the Garma Festival, Yolngu call attention to the inequitable nature of the relationship between Indigenous people and Europeans. For example, some of the performances at the 2003 festival concerned narratives associated with Indonesian fishermen from Macassar (Ujung Pandang), who had trading relations with Yolngu until just after the turn of the century. A beche-de-mer (trepang) industry continued from the early seventeenth century to 1907. The spiritual significance of the Yolngu-Macassan connection to particular Yolngu clans has been documented in the film Spirit of Anchor by Barker and Glowczewski (2002). The songs and dances at the Garma Festival acknowledge the moral application of the principle of reciprocity in exchange relations between the Macassans and Yolngu. According to McIntosh (2000: 144), "Despite episodes of violence and bloodshed, with the passing of time and the blurring of memories, the seafarers are remembered with great fondness, particularly when compared with the European missionaries . . . miners and bureaucrats who came in their wake."

     In highlighting the moral principles employed by Macassans in their relationship with Yolngu, the performances at Garma operate as a form of admonition, underscoring the amoral practices of the Europeans who, in contrast to the Macassans, invaded and took without fair return. The songs and dances concerning Yolngu relations with the Macassans can be interpreted much like Deborah Rose (1984) interpreted the Aboriginal narratives about Captain Cook in the Victoria River District of the Northern Territory. Unlike the Macassans, the Europeans did not employ the valued moral principles of "reciprocity, balance, symmetry and autonomy" in their relations with Yolngu people. The Yolngu-Macassan relationship is, thus, held up as a model of what constitutes moral social engagement between different peoples.

     At the Garma Festival, a six-meter representation of the ancestor figure Ganbulabula (the sugarbag hunter), carved by Yolngu leader Galarrwuy Yunupingu, is erected in the center of the bunggul ground for the period of the festival. What is Ganbulabula's role in relation to the festival? Perhaps he is, as Handelman (1990: 133) puts it (but in relation to the Madonna in the Palio Festival of Siena), a kind of "ideological underwriter," a means of legitimizing the festival according to principles, precepts, and regulations that are higher than (or beyond) those of the state. In other words, Ganbulabula embodies a higher moral law that is above and beyond state law and embraces all of the participants in the festival, Yolngu and visitors alike. The visitors are meant to leave the festival not radically transformed in terms of their nature but with recognition of the moral precepts of Yolngu law and an understanding of the immorality of balanda law (that is, European law or 'Captain Cook's law').

     At the festival, Yolngu present a way of being, a philosophy of respect, and a basis for mutual and equitable relationships between Yolngu and balanda (European). The festival is not independent of the lived-in everyday world, since it, in fact, springs from it and is in response to it (see Handelman 1990: 27). Nevertheless, the festival, albeit for just a short time, is experienced by Yolngu and their visitors as autonomous from everyday life, operating according to its own law—Yolngu law.

The Laura and Garma festivals are public events that re-present the lived-in-world. According to Handelman (1990: 49), "Events that re-present do the work of comparison and contrast in relation to social realities." In other words, an event that re-presents raises "possibilities, questions, perhaps doubts, about the legitimacy or validity of social forms, as these are constituted in the lived-in world" (ibid.). Such events comment on and call into question the world as it is by inverting it (but not necessarily transforming it), by neutralizing distinctions, and by proposing alternative possibilities of being.

     For example, at the Garma Festival, it is Yolngu who are in charge. They are the lecturers at the Garma bush university, while the academics and bureaucrats and politicians become their students, their 'initiates.' During the 2003 festival we were told by a number of Yolngu participants that garma is but the first stage of a ceremonial complex involving public or 'outside' knowledge and that there are other levels of knowledge or 'inside' knowledge that Yolngu initiates are taught after they have first participated in garma. Nevertheless, even in the context of garma ceremonies, there is secrecy, or what Keen (1994: 226) refers to as "secrecy in public." This secrecy in public is the key to the politics of knowledge. At the Garma Festival, it is made clear to the participants that there is much more to the story or performance or painting than is being revealed. We are repeatedly told in various ways that what is being conveyed to us at the festival is open knowledge (garma). Yolngu teachers stress that garma is only the first level of a system of knowledge consisting of a number of restricted deeper levels.

     Therefore, even though songs, dances, and designs are performed in public, they may have their secret interpretations. As Keen (1994: 226) notes, performances in public may, in fact, operate to emphasize "the ignorance of those without access to knowledge of this secret significance and the privilege of those with inside knowledge." Balanda visitors to the festival are thus placed in the position of powerless initiates under the authority of powerfully knowledgeable Yolngu elders.

     Festival visitors are advised not to ask too many questions but to learn by watching and listening, and also, where appropriate, by doing. Visitors are invited to try weaving, to participate in making a didjeridu or a spear, and to taste bush food. They are also taught where it is not appropriate to observe or do these things—that is, they are taught Yolngu intellectual property law. Through a complex dynamic interaction between processes of concealment and revelation, that knowledge is transmitted among Yolngu people (see Keen 1994; Morphy 1991; Tamisari 2000). Emphasized again and again is the idea that there are levels of knowledge that are not revealed but that are generative of what is revealed. The visitors are allowed access only to the public realm of knowledge (garma). Bureaucrats, government officials, politicians, academics, and others become neophytes, and hierarchies of the lived-in-world are inverted.

Conclusion

The phenomenon of the Indigenous cultural festival has burgeoned in Australia in tandem with state multicultural policies that foster the celebration of culture as a factor in the government of people. Festivals such as the Garma Festival and the Laura Aboriginal Cultural Festival are 'governmental technologies' in that they work to constitute the collective identities necessary for the task of "governing through community" (N. Rose 1999: 189-90). Yet through participation in these festivals, Indigenous people are able to confront creatively the contradictory social forces that affect their lives by engaging with state agencies and bureaucratic realities according to their own terms. What De Soto (1998) has argued for the carnivals of the German peasants of the Black Forest also holds true for festivals of Indigenous Australia. Through such festivals participants "express contemporary existential fears and economic insecurities arising from national and transnational political communities, markets, and bureaucratically enforced policies and regulations in which their lives are embedded" (De Soto 1998: 482).

     The festivals allow Aboriginal people to bring various parties into a spatial and temporal frame that they themselves regulate and direct. These events provide a means of creating, albeit only momentarily, a microworld in which the vagaries and uncertainties of their interactions with state agents can be controlled, or at least can be more easily controlled than in everyday life. In this way, Indigenous people are better able to confront the state and the asymmetrical and oppositional relations that they have with its agencies. Governmental and nongovernmental agents of the state, removed from their familiar social spaces and their bureaucratic agendas, are drawn, if only briefly, into a web of relationships and obligations not of their own making but of Aboriginal fabrication. These public events, and in particular the use of media coverage, enable Indigenous people to reach an extended audience and communicate their concerns as widely as possible, both nationally and internationally. Through the festivals, they attempt to show publicly not only how they view the world but also how their worldview might provide a moral discourse (or moral principles) through which their relationships with the state and with other Australians might be transformed.

     By using the performances as a mode of cultural transmission and acquisition of embodied knowledge, Indigenous people challenge the idea that their dances are a mere theatricalized presentation of fixed traditional forms. They question the dominance of a discourse that, by producing and celebrating a peculiar concept of 'traditional culture,' denies them the contemporary reality of their lived experiences and the agency to control this reality. At festivals, people do indeed celebrate tradition, but they do so to put it to work to make sense of history and to negotiate a way through the myriad government technologies that have invaded their life worlds. Festivals allow Aboriginal people to be not only custodians of the past but also agents of change. As Mary Douglas (1995: 23) has aptly reflected, "Time past is remembered, privately or publicly, when it can be used in time present to control the future."

Acknowledgments

This paper draws on a public lecture I gave in Mexico City at the invitation of the Comision Nacional para el Desarollo de los Pueblos Indigenas (CDI) and the Australian Embassy (July 8, 2004) and again at James Cook University (JCU) at the colloquium "Indigenous Strategies of Communication: Cultural Festivals and New Technologies" (July 18, 2004). Although, of course, the responsibility for the substance rests with me, I am grateful to the colloquium participants, particularly Barbara Glowczewski and Marcia Langton for their comments. I am also indebted to Franca Tamisari for generously taking the time to read the essay and provide valuable feedback, and to my colleagues at JCU, especially Michael Wood and the late Douglas Miles. The field research on which this chapter is based was conducted with the help of a Merit Research Grant from James Cook University. I especially acknowledge my friend, the late Maggie Wilson and my colleague, Christine Togo-Smallwood, for accompanying me to the field, for helping me to record the festivals, and for sharing their insights on the festivals. Finally, I thank Judith Kapferer for her invitation to submit this essay for publication and for her encouraging suggestions on how to strengthen my analysis, and Brenda Farnell for the opportunity to republish it in this issue of JASHM.