Continuum: The Australian Journal of Media & Culture
Vol. 3, No. 2, 1990
Communication and Tradition: Essays after Eric Michaels
Edited by Tom O'Regan

Enlisting the Warlpiri

Tim Rowse

Yuendumu may be one of the best documented Aboriginal communities, and this is one of the reasons it was chosen as a site for this study. But it also means that research has itself become institutionalised here, and must be accounted for as much as the church, the school or the council. 1
Eric Michaels 1986

Who asked what of whom, in what language, under what circumstances? 2
Eric Michaels 1985

The topic of this paper is the historical and social conditions of non-Aboriginal research acts among the Warlpiri people of Yuendumu. Yuendumu has been a research site, not only for Eric Michaels, but for scores of researchers of many disciplines and interests, since its foundation as a government ration depot in 1946. My aim in this essay is to cease taking that 'place' as given, not only for the reasons expounded in Paul Carter's critique of 'Imperial history', 3 but also for two other reasons. First, I wish to think about the intellectual consequences of one of the 'spatial' aspects of the colonisation of Central Australian Aborigines: settlements have made Aboriginal people available for study by intellectual workers such as Eric Michaels and myself. Second, I think we should not assume that the Yuendumu 'community' can be taken for granted, a unitary object for representation.

In what ways, if any, was the fact of Yuendumu a matter for Eric Michaels' analysis? I will try to answer this question in the first section of this paper. In the second I will briefly compare two accounts of the relationship between power and knowledge, that which can be read in some of the work of Michel Foucault and that of Bruno Latour. I will state my preference for the latter. In the third section, I will discuss some of the historical conditions which gave rise to a research site called Yuendumu. After that I will introduce an essay about doing research at Yuendumu which I believe deserves to be better known.

Eric's Yuendumu - what kind of community?

Section Two of The Aboriginal invention of television in Central Australia, 1982-6 4 is entitled "Yuendumu, Northern Territory: a case study". If, as Eric observed, 'research' is one of Yuendumu's fifteen institutions, 5 what kind of institution is it? Eric's main point was that Yuendumu residents have become accustomed to hosting research according to their own interests. Each researcher (nowadays requiring community permission to reside there, under the Aboriginal Land Rights (NT) Act 1977) presents to Yuendumu residents a fresh problem of assessment: What does he/she want? What can I/we gain by cooperating with this person? Typically, Eric formulates this interest in terms of a community's need to communicate. Researchers are

an opportunity for effective outward communications of the community's conditions through publication, media and direct access to government. The community attempts to appraise these opportunities, and its chances to get their own perspective across, and may entertain certain visitors for these reasons. 6

It is interesting that Eric referred to a 'community's' interest and appraisal, rather than to the interests and appraisals of individuals or families. This is somewhat at odds with his observation elsewhere that: "the diversity of opinions and perspectives characteristic of Aboriginal life means that a researcher can usually find at least some informants to support any given proposition." 7

Francoise Dussart, another ethnographer of Yuendumu whose research was concurrent with Eric's, has stressed not so much the diversity of opinion within that 'community' but rather the competition among residents for the ethnographer's attention, a rivalry that arose, soon after her arrival, between two groups of women "each of whom saw me as a resource, and as an additional way in which to validate their traditional knowledge." 8 It was no simple matter for either group to conscript Dussart to their purposes. A prominent woman in one group did not want Dussart to camp with her, for instance, as "she felt my presence would put too much pressure on her. She feared that many of her relatives would step up their demands for food, money, petrol and so on, which they would assume I would provide and which she would have to distribute." 9 Deciding to live in a house which was within the domain of neither group of women, Dussart writes that it took "months of diplomacy" to share her time among different groups whose ritual life she wished to study.

Dussart's account suggests two deficiencies in Eric's answer to his very appropriate question about the institutionalisation of research at Yuendumu. First, the 'community' as such may not exist as a unified entity which makes appraisals according to 'its' interests. Rather, Yuendumu is a congeries of rival interests, defined by lineage, age and sex. This is not to say that Yuendumu residents are incapable of a sense of overriding collective interest (I will give some examples below), but only that such collectivities are problematic achievements, not given states of affairs guaranteed by the mere fact of co-residence, with a Council and other symbols of corporate unity, at a place called Yuendumu. Second, if researchers who stay become institutionalised, then we had better refrain from assuming that they do so in one way or for one reason. Better to assume that each research program has its own Yuendumu history.

Here is an example. One of the most prolific (in terms of numbers of papers published) researchers ever to have worked at Yuendumu, M.J.Barrett, was a student of the physical forms of the Aboriginal body, particularly its skeletal structure and dentition. 10 Barrett's earliest observations amounted to a prolific production of data: 127 Warlpiri mouths examined and annotated in August 1951, 131 in May 1952, a total of 224 persons examined, including 34 children seen twice. 11 He worked from a three room hospital with two full-time sisters, and also had the support of teachers at the new school. 12 The machinery of dental research was institutionalised early in the history of Yuendumu. That hundreds of Warlpiri allowed Barrett and co-researchers to measure their craniums and peer into their mouths over a short time requires telling a particular story of the institutionalization of research, a story different to that told by Dussart or modelled in Eric's generalisation.

If we admit these two criticisms then Eric's sketch of 'research' as a Yuendumu institution becomes less a guide to the sociology of research and more a clue to some characteristic assumptions and arguments of Eric Michaels. Was Eric committed to the idea of a 'communicating community'?

It is a crucial (and laudable) feature of the politics of his research that he was so committed. But the 'community' he postulated is an idealized locus of Walpiri-ness. Let me give an example. Michaels and Kelly 13 recount the production, and speculate about the reception, of a video which they helped to make about the 1927 Coniston massacre. The authors make no reference to a 'community' when describing the social organisation of that process of production but rather refer to the traditional relations among members of linked patrilines and matrilines in order to account for who appropriately did what. 'The community' is mentioned only in relation to the likely reception of the finished video. Referring to the social organization of the video's production, they suggest:

We might see this as an experiment by Jupurrula to embed the video-making process in his own cultural forms and thereby create a tape which the community would agree was proper and authentic. 14

Here 'community' refers to a postulated audience of 'traditional' people, that is, an audience whose likely criteria as viewers can be inferred from ethnographic reports of their social relations of ritual production: to the extent that the relations of video production recapitulate the relations of ritual production, the video narrative will be judged 'proper and authentic' by the 'community'.

However, elsewhere in Eric's reporting, it is clear that the 'community' was not necessarily able to come to such unifying judgements about a screened artefact's adherence to custom. "In reviewing an old film of the Fire Ceremony," he relates, "which people wished to use as a mnemonic to revive the ritual twenty years later, an audience of senior owners of the ceremony agreed that the deceased were all 'in the background', although my eyes thought otherwise (while the women disagreed and refused to watch)." 15

Michaels and Kelly also betray some uneasiness about the assumption that 'the community' is a repository of Warlpiri tradition, when they remark on the uniqueness of the Coniston video. A staged narrative, produced at the request of the researcher, its assembly involved its producers in deliberation about the conditions and contents of its artifice, unlike other Yuendumu tapes which "record daily life and already assembled events."

How much easier it is to restrict tapings to events that emerge from the community itself, where one is assured the proper relationships are already established, or at least, the cameraperson is not responsible if they are not. 16

That is, the 'community' is imagined as a scene of "proper relationships...already established", except when it is not.

Eric tended not to make the possibilities of that 'not' his concern as an ethnographer. Rather he made it his task to expound the vitality and integrity of the culture at the service of which he put his catalytic energy; he therefore tended to assume the community's propensity for 'properness' in the application of Warlpiri rules. Because Eric looked to video and television to play a part in the orderly reproduction of that culture, his Yuendumu is usually evoked as an imagined community of custom uncompromised.

Such fancy can be theoretically and politically productive. It allowed Eric to sketch magisterially a new paradigm within which to describe Aboriginal traditions and their changes. His "Constraints on knowledge in an economy of oral information" 17 bids us rethink Aboriginal social structure - as an ensemble of pathways of information withheld and released according to rules. It was in learning these rules (partly through inadvertently violating them) that Eric established the acceptability of his research to the 'community'. 18 To describe Warlpiri culture becomes an essay in the ethnography of its orality, in this article a series of illustrative stories about the handling of information. The political import of this focus is then spelled out.

Materialistic anthropology probably envisages precontact Australia as a continent pulsing with demographic patternings of resource foragers. Because mining, pastoralism, massacre, and disease have altered the demography and dispersed or despoiled the material resources, it must regard the contemporary situation as archaic fragments, remnants of a strategy no longer adaptive. But an information perspective reveals a continent of information foragers and sees the coming information revolution as an extraordinary opportunity for the re-creation of Aboriginal society. 19

One of Eric's strengths as a politically committed ethnographer is made plain in this passage: one's pessimism about (and on behalf of?) the Aboriginal enclave is arguably rooted in an impoverished theoretical imagination. Renovate theory - by shifting attention from despoiled materiality to intact orality - and new political possibilities become evident, opportunities to harness recent changes in communications technology and to shape government policies so as to give substance to self-determination - in short, to produce a 'cultural future'.

I will not discuss Eric's policy recommendations or their vicissitudes other than to highlight the importance of the term 'community' in his Current Anthropology paper. It seems to me that his political project may here have set limits to his theoretical imagination.

His concept 'community' subtends his exposition of information pathways. Traditionally the information flows characteristic of Aboriginal culture marked out "Dreaming tracks" and

... the sum of these lines is a network, a vast grid of tracks and attendant special sites that enmeshes the entire continent, locating every Aborigine within it in relationship to every other Aborigine and to the landscape. 20

Though 'Aborigines' are here evoked as individual nodes of the network, Eric acknowledges that is is not just individuals which interact. Speech, he has already argued, is "capital of a particular sort that individuals, and communities in ceremonial exchanges, can manipulate to social and economic advantage." 21 But were there always 'communities' as nodes of the continent-wide network, or are they a recent social form? His answer to this question is not confident.

The isolation of remote communities may be a post-contact effect, one which can be offset by new communication technologies, appropriately designed, once the challenge is admitted. 22

The 'challenge' seems to include the reconnection of people previously linked. Without sketching the sociology or history of the social form 'community', Eric instanced ceremonial exchanges between communities as one of the relinkings to be encouraged.

This perspective on the past and future raises several questions. At what point in the history of Aborigines' contact with Europeans did 'communities' become the collective agents of ceremonial exchange? If there is a problem in the reproduction of Aboriginal culture (implicit in his mention of the 'opportunity' of its 're-creation') 23 then how are the entities known as 'communities' - with their alternate potentials of consolidation or dissolution - implicated in the Warlpiri's 'cultural futures'? These seem appropriate questions to put to an ethnography of orality, but Eric neither theorised nor historicised (apart from retelling a chronology of Yuendumu) his 'communities'. 24

The one break in this theoretical silence is unhelpful. In an aside he remarked

I choose [communities] over 'tribes' or other terms which have proved problematic in application to Australian social structure. 25

Problematic indeed. It is curious that Eric made no reference to Mervyn Meggitt's Desert people 26 whose exposition of the Walpiri sense of 'community' is almost single-handedly responsible for that terms's retention of any respectable place in the lexicon of academic studies of traditional Aboriginal society. According to Meggitt, among the Walpiri there were four divisions or 'communities': Yalpari, Waneiga, Walmalla and Ngalia. Four hundred of the 500 Warlpiri living at Yuendumu and Mt. Doreen when Meggitt wrote were Ngalia, 27 a fact which one could cite in arguments for or against the existence of a Yuendumu 'community' at at that time.

Other ethnographers of other Aboriginal peoples have not used the term 'community' to refer to a level of traditional social organisation because they have not found it to exist as 'the maximal political entity'. 28 Indeed so shapeless have proved the observed and inferred collectivities of Aboriginal society that Burridge, after summarising the ethnographic evidence against the existence of 'tribes', recommended that

because it is convenient to recognize and label a collection of people who have more in common among themselves than they have with others, tribe is adequate enough. It identifies, gives placement on a map. [Paul Carter take note.] But it is still an imposition on the Australian situation. The coherence it might suggest has little correspondence with the realities of Australian life over the past fifty years or so, and whether or not there ever were groups of Australians organized into specific and particular tribes one or two centuries ago we do not know. Either we can use tribe as a simple convenience, roughly but not exclusively fulfilling the conditions rehearsed above, not think of tribe at all, or resort to something like Unspecified Larger Groups: ULGs. 29

Burridge's argument for ULGs applies to ethnographies which attempt descriptively to reconstruct Aboriginal society as it was before Europeans affected it. Ethnography which takes note of history is quite capable of tracing the emergence of new senses of community. Meggitt argued that government settlements (three of which - Yuendumu, Hooker Creek and Phillip Creek - serviced the Warlpiri he studied in the 1950s) were becoming new foci of sentiment as a new generation born on the settlements grew up with a sense of each settlement as the country of his/her conception and birth. However, he warned, "such a re-arrangement may not become effective until after the death of the men who still remember clearly the boundaries, dreaming-sites, and other features of the old countries." 30 That those same old men might strive to transmit those older senses of affiliation to place is a possibility envisaged more in Eric Michaels' than in Mervyn Meggitt's writing, yet it poses more problems for the former. Might not the reproduction of some traditional knowledges create a tension within Warlpiri life between identification with country of birth (i.e. Yuendumu) and a sense of belonging to the countries and kin of one's father's and mother's lineages, countries and kin which may be far from Yuendumu? Not a destructive tension, perhaps even an enriching complication and diversification of one's heritage, this uncertainty about what loyalties are to be perpetuated gives the reproduction of Aboriginal culture an internal (amongst Aborigines) as well as an external (Aborigines/non-Aborigines) politics, a politics lost sight of when 'community' is used simplistically. Some exchanges between communities must work to strengthen the solidarity of lineages between places of residence, others work to forge unities of interest among co-resident people. 31

Perhaps it was because Eric's eye never left that part of his audience which works in and around the bureaucracies of Aboriginal welfare, an audience for whom 'community' is an unimpeached descriptive term, that he did not, in his more policy-oriented writing, make 'community' a problem for inquiry. 32 The term 'community' has become essential to the contemporary representation of Aborigines' interests: the formation of Councils, the assessment of their revenue bases and the authorisation of their expenditures all rest on a notion of 'community'. The term postulates that which counts, in administered Aboriginal society, as the 'public' sphere, the crucial forum of their 'common wealth' as Aboriginal people begin to engage in the tasks which Australian democracy has recently devolved to those it once treated as 'wards'. One of those tasks, whose devolution Eric helped to promote, is the administration of local video production and television reception facilities. Partly as a result of Eric's advocacy, 'remote communities' have recently each been given such goods. The first person to inquire into their use will have to go beyond Eric's framework and unpack the term 'community'.

I do not wish to say that 'community' is an absolutely unwarranted term, only that it requires our persistent attention to its political and theoretical utilities and disutilities. I agree with Patrick Sullivan's suggestion 33 that the task now facing the anthropology of Aborigines is to help devise organs of representation which reflect their traditions of collectivity while enabling them to exercise self-determination among non- Aboriginal state and economic structures.

2 Research as political action

Michel Foucault has done more than any other recent writer to link the exercise of power with the generation and deployment of knowledges. Domains of administrative action, he has argued, are necessarily also domains of knowledge. Knowledge and power are mutually constitutive. In working out this argument, Foucault had some fascinating instances to consider: the hospital, the prison and later the confessional and the psychiatric or psychoanalytic interview. He and other writers have also mentioned the asylum and the school-room. It would be foolish to predict the limits of the list of cases illustrating the political and epistemological conjuncture starkly exemplified in the 'carceral archipelago': administered populations are fields of knowledge. Indeed, why not add the Western Desert settlements?

But to make an institution such as Yuendumu an instance of the Panopticon - the starkest form of interpenetration of knowledge and power -is to begin to see the limitations of the Panopticon as exemplar. Yuendumu has been a place of incarceration and surveillance in only the very weakest sense of these terms. Welfare Branch records are replete with public servants' worries that Western Desert people were incorrigibly nomadic. True, the pattern of their nomadism has been transformed by contact with non-Aboriginal institutions; the material motives and means and the temporalities of Aboriginal movement are different now. But settling Central Australian Aborigines down so that they could be inducted into the 'normal' Australian way of life has proved beyond any Administration's resources. People have moved between congregations of their relations on missions, settlements and cattle stations, getting rations and, now, welfare cheques, where they could, in measured, cautious and incomplete submission to non-Aboriginal tutelage. 34 A Panopticon so undisciplined is something else altogether.

Nonetheless, settlements such as Yuendumu have been privileged research sites; as Eric noted, Yuendumu is a relatively well documented place of Aboriginal habitation. However, what makes Yuendumu conducive to researchers is not the captivity of its inmates but the possibilities it presents for what Bruno Latour calls 'alliances'.

I cannot confidently summarise Latour's recent, idiosynchratic theoretical essay "Irreductions". 35 However two of his themes I find valuable in thinking about research at Yuendumu. The first is that science must be understood as an arrangement of forces which acts to produce effects. Such effects are usually inscriptions 36 known as the data of the study. The second proposition is that these assemblages of forces imply internal political relations. Latour is quite prepared, for instance, to speak of 'alliances' between a biologist and his/her microbes, alliances which persist only under certain specific conditions. Without those conditions the microbes will not behave in accordance with the politics of the alliance and the scientist's endeavour will fail. Knowledge is thus the forceful rearrangement of matter, and matter refers to all things - human and non-human, 'society' and 'nature'. What distinguishes Latour's work is his lively sense of the contingently political nature of all arrangements of matter, and his scorn for epistemology or for any enterprise which seems to grant to knowledge a foundation in a transcendent order of rules. "We cannot distinguish between those moments when we have might and those when we are right." 37 Methods of investigation could only ever be historically situated actions, and transcendent notions of knowledge or science only disguise the contingent conditions of existence of those arrangements of matter and forces which we call knowledge and science. Thus

It is not a matter of science. If arguments were sovereign, they would have all the potency of a gouty monarch immured in a crumbling castle. If science grows, this is because it manages to convince dozens of actants of doubtful breeding to lend it their strength: rats, bacteria, industrialists, myths, gas, worms, special steels, passions, handbooks, workshops...a crowd of fools whose help is denied even while it is used. The superior school of facts is too often one of arrogance. Enlightenment leads to the crassest form of obscurantism. 38

Latour's aphorism that "nothing is known - only realized" seems to me to expand our imagination of the possible variety of circumstances in which knowledge and power are mutually constitutive. The Western Desert settlements were not Panopticons, enslaving subject Aborigines to Foucault's 'eye of power'. Rather they were relatively durable assemblages of matter in which the Warlpiri were predisposed to recrutiment to certain alliances with researchers, and to which researchers were attracted - ripe for the formation of alliances with Warlpiri of intiative. No alliance was guaranteed or pre-ordained. The contingencies of research, of mutual solicitation and manipulation, remain endless. It is within such a framework, so open to history, that an understanding of Yuendumu research acts is possible.

3 Marshalling Walpiri Testimony (1)

Research into the Warlpiri precedes the formation of any settlement for them. Indeed, one of the very first efforts to mobilise the Warlpiri as allies in the production of a certain knowledge was itself part of the process of choosing a settlement site. 39

Organised welfare initiatives towards the Warlpiri date from the early 1930s, when the Lutheran Missionaries of the Finke River Mission, Hermannsburg, became aware that Ngalia Warlpiri were beginning to migrate southwards to an area known as Haasts Bluff, then the most northerly extension of Lutheran patrols. Pastor F.W. Albrecht was one of several observers of the pastoral and mining frontier who wished to save Aborigines of the desert from the fate that seemed to have befallen the Arrernte around Alice Springs: physical destruction and cultural and spiritual corruption. Their submissions about Warlpiri distress, to the Federal Minister of the Interior, the Administrator of the Northern Territory and the Chief Protector of Aboriginals (whose office was renamed Director of Native Affairs in 1939), helped eventually to persuade the federal government to open the Yuendumu ration depot and mission in 1946.

Two research processes may be distinguished in the ten years leading up to that action. In the first, the Warlpiri were objects to be protected by declaring reserves and 'prohibited areas' - where they were to be issued with rations. In the second, the Warlpiri, in order to be protected from one particular pastoralist, had to be induced to speak, to articulate a certain experience of oppression, as evidence in court.

In 1936, Adelaide University's Professor of Pathology, J.B.Cleland told the Minister for the Interior that the Granites area (Walmalla Warlpiri country to which gold-miners had been drawn in a 1933 rush) had become a focus of 'detribalizing' contacts with Whites. According to chief Protector Cook 40 the Granites was already 'prohibited' to Aborigines. Cleland went further, recommending a reserve (in effect, an area 'prohibited' to Europeans) from the northern border of Mt.Doreen (a pastoral lease and wolfram mining camp settled in the late 1920s by Bill and Doreen Braitling) to the Granites and west to the Western Australian border. His proposal included discontinuing Braitling's access to the spring 'Chilla Well' to the north west of his lease. 41 Cook, who, in 1935, had heard the same suggestion from anthropologist Olive Pink and from the General Secretary of ANZAAS, warned his superiors in the NT Administration that those in favour of the proposed Tanami to Alice Springs stock route opposed the suggested reserve. 42

The Administration's first Patrol Officer, T.G.H. Strehlow, also suggested a sanctuary for the Warlpiri: the great reserve in the south western corner of the Territory should be extended east to the 132nd parallel and north to include the Ngalia homelands, the Davenport Ranges. But Cook was again sceptical, arguing that only about 200 Aborigines would initially benefit, and that the Minister must weigh this small gain against existing pastoral and future mineral investments. 43 The Braitling interests, and future European interests in the area, were in this way well 'protected'.

Notwithstanding the failure of the reserve proposal, certain ad hoc welfare measures were extended to the Warlpiri in this period. In northern Warlpiri country miners issued rations; further south, Mt. Doreen became a centre for gathering Warlpiri who also received rations.

The miners' treatment of those they rationed was not strictly supervised. Strehlow's 1937 patrol to the Granites (and his search of records) found no evidence that the Granites had ever been a prohibited area, and noted that licences to employ Aborigines had been issued to miners there in 1933. He did not think the field big enough nor the newcomers sufficiently disreputable to warrant prohibition of their contact with Aborigines. The principal miner, though employing Aborigines in conditions that violated the Aboriginals Ordinance, was let off with a warning that he had better comply in future. 44 In June 1944, Patrol Officer Sweeney found the same miner at fault in his failure to issue blankets and clothes, but made no other criticisms. 45 In a separate report he estimated that there were about 100 Walbiri in the Granites/Tanami area. Though he did not comment explicitly on the suggestion of a reserve for these people he remarked that, generally, young Walpiri who had left their desert homelands "have no desire to return - they speak of it as 'hungry country'; the older people are still drawn to their old tribal hunting grounds when seasons are good". 46

Sweeney therefore saw a need for government-subsidised rations to continue to be issued to 42 Walbiri by a white family (the Harris's) then living at Tanami. He noted as well that a few aged people at the Granites could do with rations. When the Harris's left, in 1945, the Native Affairs Branch sent Catholic lay missionary Frank McGarry to ration them. In May, McGarry found 60 at Tanami and 30 at the Granites. "Never have I seen people living in such appalling conditions." He asked for a settlement to be established at Police Well, eight miles north west of the Tanami field, and forecast that any one issuing rations would attract a clientele of 250, none of whom would be able to live off the nearby country. 47 Meanwhile, McGarry began to ration according to his own policy. "Two laws made are: No wash, no breakfast; No work, no tucker. They both work excellently." But to enforce that rule, McGarry had to control food distribution among the Warlpiri themselves; by issuing food that he cooked himself he foreshadowed one of the most important post war institutionalisations of rationing - 'communal feeding'. 48

By January 1946, there were 160 Warlpiri at Tanami Settlement. A medical report criticised the decision to ration there, saying that the water supply was insufficient and health consequently very poor. Everybody should be moved to the Granites without delay, advised the report. 49 But water was a precarious matter there too, and, later in 1946, the Warlpiri rationed at the Granites were moved south to the settlement which had just been set up east of Mt. Doreen (Yuendumu).

It was in choosing Yuendumu's site that an unprecedented politics of research was briefly attempted.

When Strehlow inspected Mt. Doreen in 1940, he was appalled by the incidence of venereal disease there, and made the rather empty threat to Braitling that Aboriginal labour would be taken away from the lease and its diggings. Strehlow's disquiet was apparently relayed via Albrecht to the South Australian Baptist Union which, in March 1944, raised with the government the possibility of a mission among the Warlpiri. Warlpiri were then known to be at Tanami, the Granites and Mt. Doreen, and 100 at Haasts Bluff. 50 With help from Albrecht, Baptist pastor Laurie Reece inspected southern Walpiri country by camel in June and July 1944.

Reporting this trip, Reece was scathing about Braitling. Puzzled at first by many Warlpiri's apparent avoidance of his party, he was told by some friendly Warlpiri that many people had had unpleasant experiences of Whites, including the lessee. One described Braitling as "a proper greedy fellow", another as "a larrikin" who, brushing aside the men, enjoyed the sexual favours of young girls and kept two of them as his concubines at Vaughan Springs. Reece's informants said they were prepared to talk to the police about it. He concluded that Vaughan Springs and the Davenport Ranges were actually Ngalia homelands, but that fear of Braitling now alienated those parts from them. Vaughan Springs, by far the best water for hundreds of miles, should therefore be the site of any Ngalia mission and most of Mt. Doreen lease would have to be a reserve, Reece argued. Braitling could keep the country near Mt. Doreen itself.

On receipt of Reece's report in August 1944, the Administrator began cautiously to involve himself in the issue of a mission to the Warlpiri. He told Reece in December 1944 that if the Baptist Union wanted Vaughan Springs it would have to negotiate with Braitling. Meanwhile, he sent police to Mt. Doreen to investigate the Braitlings. Mr. Braitling was not available for interview, but Inspector Littlejohn became convinced from other inquiries that Braitling's association with two girls was unsavoury. He doubted, however, that a conviction was possible. The Administrator recommended that Patrol Officer Sweeney be equipped with car to visit the area more frequently. 51

Over the next year, Sweeney confirmed that the Warlpiri were being prevented from using Vaughan Springs, a violation of Section 24 of the Crown Lands Ordinance which gave Aborigines rights to gather natural foods and water on pastoral leases. He also noted a number of Aborigines' allegations that Braitling had severely beaten a man, Jimitja.

The government decided to prosecute and sought more statements. These Sweeney collected from men and women who seemed fearful of being caught talking. Eight incidents of flogging were documented from Aboriginal testimony, some of it from stockmen (Aborigines classified as 'half-caste') who had assisted Braitling in his punishments. The same unease among witnesses was evident to Sweeney when the evidence was presented to a hearing. 52 Nonetheless, Braitling was committed for trial.

But when the allegations came to trial they were not sustained by the witnesses recruited by the Native Affairs Branch. In August 1945, the Supreme Court acquitted Braitling of flogging Jimitja. Even Jimitja denied before the Court any knowledge of the incident. Other witnesses, Warlpiri with little experience of the White world, did not perform convincingly under cross examination. Though the allegations against Braitling remain unproven, others to whom I have spoken do not doubt their veracity. Baptist Pastor Tom Fleming, resident at Yuendumu from 1950 to 1975, told me in April 1989 that he had no doubt Braitling had flogged Jimitja and others (though he had never asked Jimitja); it was a matter of normal practice in that time and region, he said, to give "troublesome" Aborigines a "hiding".

It would seem, then, that Warlpiri testimony became the object of a power struggle between Native Affairs Branch officers and the Braitling family. As the source of rations, discipline and patronage, the Braitlings held a monopoly of European hegemony on Ngalia Warlpiri territory, powers to which the state then had no material answer. So confident was Braitling, that he argued in the trial that the Native Affairs Branch, desiring his land for a mission, had framed him by coaxing falsehoods from witnesses.

Justice Wells took this charge sufficiently seriously to order Justice Simpson's enquiry. Simpson's report not only vindicated Sweeney's honesty as a reporter of Warlpiri statements, but revealed that, by the time the alleged flogging went to trial, Sweeney's senior officers had already rejected his recommendation to resume about 700 square miles of Mt.Doreen, including Vaughan Springs, as a Warlpiri reserve.

In short, competing claims to Vaughan Springs gave rise to competing claims to Warlpiri testimony. The failure of the NAB to 'realise' in Court the Warlpiri knowledge of Braitling may not have decided the issue of where the mission/settlement for Ngalia Warlpiri was to be located, but it does serve to illustrate the political character of the conditions in which such realisations are attempted. It was Braitling, not Administration officers, who most effectively mobilised Warlpiri testimony. 53

Rationing began at Yuendumu in 1946, and the Baptist Mission in 1947. After twenty-three years of their influence, during which time a number of researchers made Yuendumu their site for observing the Warlpiri, an ambitious project was inaugurated, Middleton and Francis' study of Warlpiri child-rearing.

Marshalling Warlpiri testimony 2

The formation of settlements for the western desert people enabled, for the first time, the collection of their vital statistics. Though the Northern Territory Administration was not keen to publicise it, epidemics of gastro-enteritis which swept through these settlements were reported in the regional Press from 1960-1. When Yuendumu's Superintendent Ted Egan asked Territories Minister Paul Hasluck for permission to decentralise Yuendumu's population in 1960-1, in order to cope with failures of the settlement's water supplies, he was able to appeal also to Hasluck's dawning sense of the health dangers of the western desert settlements. 54 In 1963 a demographer, F.L. Jones, used Welfare Branch and Bathurst Island Mission statistics to make one of the first scientificaly reputable estimates of the infant mortality rates of Northern Territory Aborigines, 142 deaths per thousand children under one year of age in the period 1958-60, almost seven times worse than the rate for Australians as a whole. 55 A September 1964 report in the Sydney Sun 56 criticised the unhealthiness of the Territory's government settlements, alleging an infant death rate of 208 for Central Australian settlements. Public commentary on Aboriginal welfare policies was therefore given a new numerical index of government and mission performance in the early 1960s. To this day, the rate of infant mortality (considerably improved by the late 1980s) has been a key statistic, something tangible in a field of public policy otherwise full of imponderable ethical dilemmas and obscure cultural considerations. Proponents both of the 'welfare as genocide' thesis and of the 'we've done our best under difficult conditions' reply have had cause to highlight this index.

It was in this political context that, from 1969 to 1971, a small team of psychologists investigated the "physical, social and psychological factors" in the ill-health of young Aboriginal children at Yuendumu. 57 The resulting book Yuendumu and its children reported on: the physical environment; the 'organisational roles in the Aboriginal settlement family'; the 'immediate environment and care' of Yuendumu children; nutrition; and mothers' attitudes and practices. Thematically, their study amounted to a cool, respectful but determined critique of Warlpiri mothering. Western desert women's practices towards their offspring were (and are) problematic from the point of view of educational and clinical staff whose notions of good mothering (practices complementary to the services such staff are trained to offer) derive from non- Aboriginal Australia.

A long and intelligent essay appends Middleton and Francis' book: "Reflections on the conduct of research on Aboriginal settlements". For the insights it yields into the social bases of research practice among Australian Aborigines, I know of only one other paper with which to compare it: Eric's "Ask a Foolish Question". 58

How did the Warlpiri construe the psychologists' intentions and methods? According to Middleton and Francis Warlpiri "accepted our explanation that we were there to learn how they 'grow up' their children". 59 Two words are crucial in this sentence: "learn" and "grow up" (a transitive verb in Aboriginal English).

A number of commentators have pointed to a distinctly Aboriginal style of learning. Coombs, Brandl and Snowdon 60 have usefully summarised what is known of Aboriginal teaching/learning practices. They include: a respect for certain kinds of knowledge as the property of lineages and persons and a consequent orientation to older persons' and their patrimonies rather than to disembodied information; an emphasis on observation and imitation by the learners; learning by doing and doing enjoyably; the reinforcement of correct behaviour (and the spirited but kindly mockery of error).

Only some of what the researchers did with the Warlpiri could be described as 'learning' in this sense. For example, children and elderly people were pleased and amused when the researchers showed "interest in Aboriginal life by attending ceremonies and by attempting to learn the Warlpiri language". 61 But that is not essentially what the researchers meant by 'learning'. To generate and inscribe data (about the environmental and social context of child development) was their over-riding purpose, activities formalised into certain legitimate protocols of academic research: observing domestic situations, undertaking interviews, administering tests and questionnaires, transcribing clinical data. Middleton and Francis, at several points in this text, evaluate Yuendumu situations according to whether they were conducive to such data generation. The terms in which they report their openness to Warlpiri priorities typically include the retrospective calculation of the costs of 'rapport'.

The establishment and maintenance of rapport took much time, and was sometimes achieved only at the expense of efficiency and ease of data collection which thus took much longer than had been anticipated. 62

If it is likely that the researchers and their subjects differed in their ideas of 'learning', were they any closer in what they understood by "growing up" children? Was not the meaning to be given that term at the centre of the differences to which the researchers referred when they wrote that

.... the background against which the research was carried out was one which emphasised the differences between the Aborigine and the European, and highlighted the superiority and dominance of the latter. 63

Features of child socialisation regarded as essential by Europeans - a nutritious diet, diligent attendance to schooling given in English, personal and domestic hygiene - seem to have been peripheral from the point of view of the Warlpiri. Their list of the essentials of 'growing up' probably included: learning the rules of avoidance and address; becoming familiar with the country's physical and spiritual resources (including far away country to which Yuendumu adults could only rarely take their children); getting to know the persons with whom one would eventually have to deal, in 'marriage' and in ritual, as complementary owners of that country. Learning to live with Whites - avoiding as much trouble and getting as many benefits as possible - also had now become what a Warlpiri learned.

In sum, for the two parties to these research acts, different sets of behaviours were central both to 'growing up' someone and to the investigation of (learning about) that process. Perhaps this difference can further be glimpsed in the way the psychologists understood Yuendumu's fundamental divisions of time and space.

It was Welfare Branch policy to require work of able-bodied adults in return for rations. Though the required working days were variable and negotiable in their length and number 64 officialdom wished to prepare desert nomads for the temporal mould into which 'normal' Australian lives are poured: the five day working week followed by the weekend. It was within this frame that the researchers worked. On weekends, not wishing to bother the Warlpiri in their 'free time', that is, the times when no Europeans were seeking to interact with them, the researchers eschewed the rigours of data generation and

only carried out observations or collected information incidental to activities such as hunting trips that were suggested to us by the Aborigines. 65

These weekend diversions, with their low intensity of data generation, may well have been the moments when the Warlpiri understood themselves to be offering the researchers the best chance to 'learn' about Warlpiri 'growing up' of children. Hunting and gathering trips

reversed the usual European-Aboriginal relationship and in such situations, where they rather than the Europeans initiated activities, more information was spontaneously volunteered than in meetings with the same people on the settlement. 66

The hunting trips were away from the settlement, not only in the obvious spatial sense, but also in that the Warlpiri hosts were in family groups, not part of the settlement mass. 67 However useful and personally gratifying for all was the spontaneity of such trips, on weekends the researchers resolutely kept to the structured generation of data as their primary research mode, the raison d'etre of their acceptance of whatever hospitality the Warlpiri offered. Accordingly, "it was...impossible to avoid entirely the aura of official authority". 68 If a child was to be weighed, its mother must bring it to the clinic. The requirements of the study thus converged embarrassingly with the settlement's authoritative routines. European staff sometimes ordered the Warlpiri to cooperate with research procedures, so jeopardizing the rapport and the feeling of friendly reciprocity built up on weekend bush trips.

But the atmospherics and etiquette of data generation, the varying faces of coercion, consent, avoidance and cooperation which the Europeans and Warlpiri presented to one other amount to only one strand in Middleton and Francis' account of the cultural politics of their research act. Just as fascinating and important are their reports of the difficulties of the interviews and tests themselves.

The interviewers' problems exemplified some of the "constraints on knowledge in an economy of oral information" noted by Eric but, importantly, they also show the adaptation of such an economy to the colonial situation. Some of those to be tested and interviewed did not come to the hall where space had been set aside for that work and so were interviewed "in the village" (crude government-built houses among which Warlpiri camped in family groups). "The only drawback to this venue was the lack of privacy it involved; interested neighbours tended to gather to watch and assist in the proceedings." 69 Another problem, with which the researchers were able to deal, was to select interpreters who were not in avoidance relationships with those to be interviewed. Some of the ambiguities created when colonial governments program social change were evident when one young interpreter advised that such taboos were now weakened, and then attempted unsuccessfully to interview women who still adhered to those very same rules.

Interpreter brokerage of a different kind troubled the researchers when some interviewees seemed to become bored.

Because the interpreters were not very quick at translating replies, long periods sometimes elapsed between questions. One way of avoiding this drawback would be to allow the interpreter to conduct the entire interview and translate it from the tape later when he could take his time. However this would leave control of the procedure in the hands of the interpreter and the interviewer would be unable to follow up interesting or obscure points. 70

Fifteen years later, Eric investigated another matter which the researchers encountered as a practical difficulty: the culturally-determined variability in the meanings of graphic images. To ask mothers about their responses to child illnesses, the researchers used drawings of children with various common afflictions. But, among some respondents,

The small boy with the infected hand was seen as being blind because his eyes were closed, the undernourished child as being unable to walk because she was seated, and the child sweating with a feverish cold as having chicken pox. 71

In using pictures as cues, the researchers also discovered that photographs were to be avoided in favour of drawings, if 'generic' situations were the subject to be probed. Photos were not read in terms of the generality of the situation depicted. "If [Warlpiri] did not recognise the person pictured, they wanted to know who he was and where he came from..." 72

Cognitive and cultural differences, however fascinating or practically problematic, are scarcely a surprising discovery, neither for those then researching the Warlpiri, nor for us now. But 'difference' was not quite the issue in 1969-71 at Yuendumu, and perhaps it never is. 'Power' was and remains inextricably associated with difference. What could never be eliminated from the vicissitudes of testing and interviewing across such cognitive gaps as the above was the historical fact that the Warlpiri became a researchable population only to the extent that they allowed Europeans to take charge of them.

That the field of knowledge was a also field of power was both a benefit and a hindrance to Middleton and Francis' efficient generation of data both inscribable and valid. They worried that Warlpiri answers to questions about their beliefs and practice were a revelation of "what the Aborigines know of European views rather than what they themselves believe". 73 Some of their interviewers' problems, they conjecture, derived from the embarrassment, to a cultural broker, of having to reveal to Europeans Warlpiri answers which the brokers thought to be incorrect from the point of view of the Administration's sanctioned settlement ways. The researchers also gave a most interesting example of the politics of such revelations of the Warlpiri way. Though "few mothers referred to the medicine men", the researchers knew that "most parents took their sick children to them." 74 How did they know? They were told while out bush with small family groups and even with 'medicine men' themselves, 75 being the learners in a Warlpiri setting. So 'rapport', for all its inefficiencies, came to validity's rescue.

Years after the end of 'assimilation' as official government policy, the aims and functions of its programs remain open to debate. Middleton and Francis reveal one function that hitherto I had not realised: by living at Yuendumu, the Warlpiri became more testable.

The Aborigines appeared to have a shorter attention span in test situations than have Europeans of comparable ages and tasks had to be suitably brief. The school children, however, appeared to cope well with the long interview about family responsibilities; although their interest seemed to flag towards the end, they continued to respond with care. The same cannot be said for the women who, when their attention wandered, appeared to answer haphazardly. The difference between the school children's and the women's responses was undoubtedly the result of the children's greater familiarity with test-like situations, and with being asked to do tasks at school that they were expected to complete even if they found them uninteresting. 76

There could be no clearer example of the powerful constitution of a domain of knowledge and data generating procedures, a domain in which Yuendumu residents became variously susceptible to being allies of the research act.

Television and Motherhood

Let me conclude by comparing two of the Yuendumu-based research programs mentioned in this paper. Both Eric's work and that of Middleton and Francis were prompted by policy considerations. In the 1960s, the physical reproduction of western desert people in government settlements was jeopardised by a high rate of infant mortality; in the early 1980s, the cultural reproduction of the same people looked as though it would be compromised by video and broadcast television. Matters of policy were at stake in each project, and both sets of researchers wished to help their subjects avert the threats.

In both projects it was necessary to elucidate Warlpiri practices in circumstances where new technologies were available. Middleton and Francis tend not to focus on that technology as a problem for their analysis: houses, cash canteens, the school and the clinic were the taken for granted facilities of the modern life to which the Warlpiri were being invited to adjust. Middleton and Francis considered only the communal dining room as a possibly negative welfare technology. 77 Eric's work, by contrast, is informed by a literature which studies the cultural differences in the use to which film and video equipment are put by different peoples. His interest in furthering a discussion inspired by Marshall McLuhan 78 leads him, indeed, to query the lack of any previous questioning of the effects of the introduction of writing to the Warlpiri. 79

Middleton and Francis' description of Yuendumu attitudes and practices amounted to a critique of the persistence of traditional Warlpiri mothering. Eric, however, was delighted to find in the production and consumption of video/television the imprint of ancient custom. "I want to emphasise the continuity of modes of cultural production across media", he wrote (emphasis in original). 80 Where Middleton and Francis investigated the distressing slowness with which Warlpiri women adopted non-Aboriginal notions of parenting, Eric celebrated tradition and difference in a policy climate in which lip service, at least, was paid to cultural pluralism in broadcast media.

The question 'do people of different cultures see, and therefore represent the world differently through visual media' produces answers which can be offered as evidence to argue Aboriginals' rights to access the new media systems proposed for remote Australia. 81

The methodological contrast is consistent with the political - cultural. Middleton and Francis, as I have noted, mobilized data generating apparatuses devised by non-Aborigines: clinical measurement, interviewing in groups whose composition, for the sake of validity and efficiency, was to be determined by the researchers, and a number of 'tests', most successfully administered, as we have seen, on the schooled section of the Yuendumu population. Eric's field techniques were a combination of the classic anthropological method of 'hanging in there', talking, listening and observing, and 'participant observation', after placing in Warlpiri hands a technology which made new activities possible. "Aboriginal people that I work with now", he wrote

express their interests in media by their activities and their plans. These findings would not be captured by survey research, or individual opinions, as they are socially emergent facts, expressed not as individual observations but by collective community action. 82

Yet, I do not want to hand all the prizes to Eric. Middleton and Francis also wrote sensitively about the points of receptivity and resistance which they found among the people of Yuendumu. Their instructively worried appendix, with its acknowledgement of the problems and opportunities of testing - of the element of brokerage in interpreting, of the enigmatic reluctance of young men and women (in contrast to old people and children) to help the researchers speak Warlpiri - is to be considered alongside Eric's work as a fine basis for a sociology of the research act.

Middleton and Francis's essay is born out of a much stronger sense of being at odds with the Warlpiri. That Eric's project stood in a relationship of expression of Warlpiri practice, rather than of interrogation and critique, and that he used a methodology permissive of learning rather than bent on the generation of 'data', accounts for this difference. This relation of expression also, however, determined a certain reticence about the implication of his research in the politics of Yuendumu. Despite his raising the question of the institutionalisation of research at Yuendumu, we learn far too little of how that happened in his case. When he lists the genres of video found at Yuendumu, 83 I ask: were some genres produced more by some types of people than by others? or did the same 'crew' tend to turn up to make all videos? who were they? and why them? We have only fragments to seize upon. "On some days there were two or more production teams operating..." 84 A "Warlpiri network" made up of video-workers from Yuendumu and Lajamanu

never quite developed. I believe this has much to do with the personalities and social identities of the video-workers in the different communities, as well as a certain competition that is evident between them. 85

This sentence is no less tantalising, and only a little less informative, than the story of Nampijinpa's failure to become a 'video-worker'.

As there were only men to provide the training, the situation was shy and awkward. The Nampijinpa was also more Europeanised than many Warlpiri women, and her relations with traditional life in the community were not well developed. Her son remained uninitiated so that her status in ceremonial life was ambiguous, meaning that she could not provide much assistance at the Women's Museum. Her video work was mostly restricted to tapes of school activities. 86

My primary source [about how fictional videos are read] comes from the videoworkers (mostly men of my own age group: 28-40)... 87

By neither describing nor explaining the intra-communal politics of the differential participation of Yuendumu people in his project, other than in these fragments, Eric's ethnography of that township remains unfinished. His commitment to a wholistic notion of 'communities communicating' held him back from that analytic step in which representations of 'community' are themselves questioned, studied as the outcome of somebody's practice. Whether that self-limiting framework would have lasted any longer than Eric's pursuit of a particular policy objective is, sadly, an issue that can never be resolved.


I would like to thank Mary Laughren and David Nsh for their comments on an earlier draft.

1. Eric Michaels, The Aboriginal Invention of Television in Central Australia, 1982-6 (Canberra: Australian Institute of Aboriginal Studies, 1986), p. 14.

2. Eric Michaels, "Ask a Foolish Question: on the Methodologies of Cross Cultural Media Research" Australian Journal of Cultural Studies, v. 3, n. 2 (1985), p. 56.

3. Paul Carter, The Road to Botany Bay (London: Faber and Faber, 1987).

4. Michaels, The Aboriginal Invention of Television.

5. The other fourteen listed by Eric were: School, Baptist church, Police, Community and Tribal Council, clinic, Housing Association, Yuendumu Mining Company, Yuendumu social club, Sports Teams, Ngarliyikirlangu Cattle Company, Outstation Council, Men's and Women's Museums.

6. Michaels, The Aboriginal Invention, p. 27.

7. Michaels, "Ask a Foolish Question", p. 55.

8. Francoise Dussart, Warlpiri Women's Yawulyu Ceremonies: a Forum for Socialization and Innovation PhD thesis, Australian National University, Canberra, 1989, p. 1.

9. Dussart, p. 1.

10. Two bibliographies on Aboriginal Health index their entries by region and author. See P.M. Moodie and E.B. Pedersen, The Health of Australian Aborigines: an Annotated Bibliography (Canberra: Australian Government Publication Service, 1971) and Thomson and Merrifield (Canberra: Australian Institute of Aboriginal Studies, 1988). The entry for Yuendumu in Moodie et al (p. 237) is particularly long.

11. M.J. Barrett, "Dental Observations on Australian Aborigines: Yuendumu, Central Australia 1951-2" Australian Journal of Dentistry, 57 (1953), pp. 127-37.

12. T.D. Campbell and M.J. Barrett, "Dental Observations on Australian Aborigines: a Changing Environment and Food Pattern", Australian Journal of Dentistry, 57 (1953), pp. 1-6.

13. Eric Michaels and Francis Kelly, "The Social Organisation of an Aboriginal Video Workplace," Australian Aboriginal studies, n.1 (1984), pp. 26-34.

14. Ibid, p. 32.

15. Eric Michaels, For a Cultural Future (Sydney: Artspace, 1987), pp. 36-7.

16. Ibid, pp. 32-3.

17. Eric Michaels, "Constraints on Knowledge in an Economy of Oral Information", Current Anthropology, v. 26, n. 4 (1985), pp. 505-10.

18. Ibid, p. 506.

19. Ibid, p. 509.

20. Ibid, p. 508.

21. Ibid, p. 506.

22. Ibid, p. 509.

23. One of Eric's rare vignettes of the disruption of cultural continuity is his observation that: "The senior men at Yuendumu had been unable to convince the boys there that they should learn traditional dancing." See Ibid, p. 506.

24. It would be possible further to trace the theme of 'community' further in Eric's work. In his "The Indigenous Languages of Video and Television in Central Australia", he refers to diverse 'linguistic communities' to be linked by satellite until they become "a community of people sharing symbols, though, as I have argued, not necessarily interpreting these in the same way." See Eric Michaels, "The Indigenous Languages of Video and Television in Central Australia" in Jay Ruby and Martin Taureg eds., Visual Explorations of the World (Achen: Edition Herodot im Rader Verlag, 1987), p. 279. One of the last pieces he wrote, a joyous description of Sydney's Gay Mardi Gras, evoked its creation of 'communitas'. See Eric Michaels, "Carnivale in Oxford Street", New Theatre Australia, n. 5 (1988), pp. 4-8.

25. Michaels, "Constraints on Knowledge", p. 507.

26. Mervyn Meggitt, Desert People (Sydney: Angus and Robertson, 1962).

27. Ibid, p. 48.

28. Ibid, p. 51.

29. K. Burridge, Encountering Aborigines: Anthropology and the Australian Aboriginal (New York: Pergamon,1973), pp. 128-9.

30. Meggitt, p. 72.

31. Dussart describes both inter- and intra-communal competition in the performance of ritual. According to her there were six basic kin groups making up the Yuendumu population at the time of her fieldwork (the mid 1980s) and, as her problems with 'diplomacy' show, there could be rivalry between them to display excellence in ceremonial knowledge and performance. However, "when Warlpiri people define themselves to other linguistic groups, they present themselves as one large extended kin group which lives at Yuendumu." (Dussart, p. 73) At an intercommunal ceremonial gathering that she observed, "young men came running from the bush carrying a wooden stick on which the word 'Yuendumu' had been painted....they told me that Yuendumu had 'won' because the men gave the best performances during men's secret ceremonies." (Ibid, p. 130) The prize? The right to lead performances when not on one's own ceremonial grounds. Mary Laughren (pers. comm.) who has worked at Yuendumu for many years, studying Warlpiri language, argues that when Warlpiri found themselves living together in large aggregations sponsored by the colonists, they saw to it that lineages intermarried. By such exchanges, they did not eliminate competition, but they ensured that there were bases of cooperation to limit such rivalries. By the exchange of marriage partners some sinews of 'community' formed.

32. Eric once commented on the way the colonizers must fantasize tractable structures of indigenous social life.

So much of the European approach to unfamiliar peoples can be glossed as 'Take me to your leader'. If the indigenous political structure is not hierarchical and does not identify general purpose representatives, the usual alternative seems to be to attempt to establish such structures and designate leaders. While this point may seem essentially an administrative problem, it is also a methodological issue. (Michaels, "Ask a Foolish Question", p.48).

33. Patrick Sullivan, "The Generation of Cultural Trauma: What are Anthropologists For," Australian Aboriginal Studies, no.1 (1986), pp. 13-23.

34. Tim Rowse, White Flour, White Power? Rationing, the Family and Colonial Authority in Central Australia, PhD thesis, Sydney University, 1989.

35. Bruno Latour, "Irreductions" in The Pasteurization of France (trans. A. Sheridan and J. Law) (Cambridge: Harvard University Press, 1988).

36. Bruno Latour, "Give me a Laboratory and I will Raise the World" in K.Knorr-Cetina and M.Mulkay eds., Science Observed: Perspectives on the Social Study of Science (London: Sage, 1983).

37. "Irreductions", p. 183.

38. Ibid, pp. 204-205.

39. Probably the first gathering of Warlpiri for research purposes was arranged in August 1931 for medical scientists from the University of Adelaide by a Mr. Kramer, a missionary, at Cockatoo Creek, about thirty kilometres north east of what is now Yuendumu. (J.B. Cleland, "Anthropological expedition to Central Australia", Medical Journal of Australia, December 19, 1931, p. 793.

40. F1 36/577 "University of Adelaide Expedition to the Granites", 6/10/36. Files are to be found in the Australian Archives, Northern Territory Branch, Darwin. Hereafter Australian Archives (NT).

41. F3 8/30, "Miss Olive Pink Secular Sanctuary: Granites, Tanami District", 7/9/36. Australian Archives (NT).

42. F1 36/577 "University of Adelaide Expedition", 6/10/36.

43. F1 38/418, "Aboriginal Reserve South West Corner Northern Territory", 9/3/37, Australian Archives (NT).

44. F3 8/30, "Miss Olive Pink Secular Sanctuary", 10/5/37.

45. F1 43/65, "Patrol Officer Sweeney Reports on Patrols", 13/7/44, Australian Archives (NT).

46. F1 43/65, "Patrol Officer Sweeney Reports on Patrols", 5/8/44.

47. F. O'Grady, Francis of Central Australia (Sydney: Wentworth Books, 1977), pp. 142-3.

48. Ibid, p. 145.

49. Ibid, p. 146-7.

50. A452 54/552(1) "S.A. Baptist Union: Mission to Natives at Granites (Yuendumu), Chinnery 18/4/44, Australian Archives, ACT.

51. F1 48/58, "Proposed Baptist Mission, (Yuendumu) NT", 11/8/44, Australian Archives (NT).

52. F1 48/58, "Proposed Baptist Mission, NT", 13/8/45.

53. According to Meggitt the choice of site did not displease its later residents (whom Meggitt met within seven years of the choice being made):

Predominantly Ngalia in composition, the settlement was in Ngalia country. To the west, Mount Doreen station, with its large Warlpiri population, was within easy walking distance, and the ritually-important Vaughan Springs area was accessible from there. Lander [Yalpari] Warlpiri to the east could be reached in a series of short journeys via the camps at Mount Denison and Coniston stations. There was also a comparatively large volume of motor traffic to and from Alice Springs, where "outside" Warlpiri could often be contacted. In the opinion of its residents, Yuendumu had much to commend it as the nucleus of a new community. (Meggitt, p. 73)

54. The decentralisation of Yuendumu and its implications for Commonwealth policy on outstations is discussed on Australian Archives (NT) file F1 69/1864 "Outstations on Haasts Bluff".

55. P.M. Moodie, Aboriginal Health (Canberra: Australian National University Press, 1973), Table 4.

56. Beazeley, "Fate of displaced Aborigines" (Sydney) Sun, 9 September, 1964.

57. M.R. Middleton and S.H. Francis, Yuendumu and its Children: Life and Health on an Aboriginal Settlement (Canberra: Australian Government Publishing Service, 1975), p. v.

58. Michaels, "Ask a Foolish Question".

59. Middleton and Francis, p. 198.

60. H.C. Coombs, M.M. Brandl and W.E. Snowdon, A Certain Heritage: Programs for and by Aboriginal Families in Australia (Canberra: Centre for Resource and Environmental Studies, 1983), chapter 3.

61. Middleton and Francis, p. 195.

62. Ibid, p. 194.

63. Ibid, p. 194.

64. Rowse, pp. 324-5.

65. Middleton and Francis, p. 196.

66. Ibid, p. 195.

67. Ibid, p. 195.

68. Ibid, p. 195.

69. Ibid, p. 196.

70. Ibid, p. 200.

71. Ibid, p. 203.

72. Ibid, p. 203.

73. Ibid, p. 205.

74. Ibid, p. 205.

75. Ibid, p. 195.

76. Ibid, p. 205-6.

77. Ibid, pp. 86-7.

78. See Michaels, "Constraints on Knowledge".

79. Ibid, p. 505.

80. Michaels, For a Cultural Future, p. 70.

81. Michaels and Kelly, "The Social Organization", p. 34.

82. Michaels, "Ask a Foolish Question", p. 51.

83. Michaels, Aboriginal Invention of Television, pp. 54-9.

84. Ibid, p. 65.

85. Ibid, p. 59.

86. Ibid, p. 58.

87. Michaels, "The Indigenous Languages", p. 10. This passage appears in a typescript version of this article but not in the published version in Ruby and Taureg's book. The typsecript version was part of a collection of essays Eric assembled in July 1986 under the title "Aborigines, Information, and the Media: The Aboriginal Encounter with Introduced Communications Technology" for consideration by book publishers. None of the publishers Eric contacted were interested in the collection (pers. comm. with Eric - the editor).

New: 22 November, 1995 | Now: 15 March, 2015