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

Aboriginal truth and white media: Eric Michaels meets the spirit of Aboriginalism

Robert Hodge

Eric Michaels irrupted into the still if murky pool of Australian anthropology like a number of stones hurled at random by a demonic child. The energy was shown as much in the seeming ephemerality of the publications in the last three years of his life as in their sheer number. The most significant of his works,The Aboriginal Invention of Television, was written in three months he claims, and has all the marks of cheap and hasty production, designed to make its intervention in time come what may. This is cowboy anthropology at its best: the dramatic ride over the hill with six-guns blazing, the mobilisation of the homesteaders (the Indians) to their own rescue, and finally the lonely ride back into the sunset.

But his well-aimed stones did hit some of their marks, and the waters remain gratifyingly troubled. His assault was sometimes contradictory and compromised but always vigorous and provocative, and its object was nothing less than Australian anthropology and the ideological complex that it serves and has served in Australian society. What is at issue is the politics of truth, on a terrain where old epistemologies were played out against new technologies and emerging politics. And Eric Michaels happened to drift into town in time to see it and try to make it happen.

The story so far.... As is now no longer in question, the English invasion of Aboriginal Australia consisted of a direct assault on all the material and cultural conditions of Aboriginal life, including both political oppression and cultural genocide. This assault was also accompanied from the start by what seemed like its opposite, a strategy of recuperation that expressed regret for the physical injustice and attempted to collect and preserve instances of the brutalised language and culture (along with material remains like skulls and skeletons). The discipline of anthropology in Australia rose out of this second ideological enterprise, whose oppositional role from the outset was framed by its overall complicity in processes of domination.

Because this enterprise is based on a contradiction its agents in the past (missionaries, other persons of good will towards Aboriginal people) as in the present (anthropologists, many others working in the modern welfare apparatus) have been a mixed group, with varying effects on and relations to Aboriginal people and their interests. Without their efforts the policies of cultural genocide would have been even more devastating, even if their enterprise also legitimated the dominant apparatus and attitudes. With this kind of proviso it is still possible to isolate a core set of assumptions that gave this enterprise an overall unity across a range of manifestations and functions. I will use the term 'Aboriginalism' to refer to this ideological core, by analogy with Edward Said's term 'Orientalism',1 which he developed to describe a characteristic ideological pattern within European colonialism.

The foundation premise of Aboriginalism is the construction of Aboriginals as 'primitive', in a binary opposition to 'civilised'. As primitives they become an endlessly fascinating object of the White gaze, able to generate unlimited discourse but never able to participate in it on any terms. Because the terms of the opposition are absolute, Aboriginals remain forever encapsulated in a self-contained universe, unable to speak or even understand their own meanings. This closed universe guarantees their authenticity and identity as Aboriginals, as worthy of Aboriginalist reverence, but any departure from its terms condemns them to angry denunciation for having betrayed their essential identity, as inscribed in their culture. This culture is incommunicable and incomprehensible for ordinary non-Aboriginals as well. Only the privileged class of interpreters (anthropologists in the present and missionaries or others who have entered into that sacred space in the past) can understand and mediate these meanings between the two incommensurable universes of discourse.

Clearly this complex is ideally constituted to act as an ambiguous instrument for ideological control. But at the same time it stakes out a space for Aboriginal survival and autonomy, in theory if not always in practice. It does recognise and give a high value to Aboriginal difference and to the intrinsic complexity of Aboriginal language and culture. Thus it has provided a base within the dominant society from which various alliances can be constructed by Aboriginal people. So even though the alliances of the past have been compromised in many ways by self-interested paternalism on the one hand and by deep resentment and suspicion on the other, nonetheless some kind of alliance is a better starting point for Aboriginal survival than overt hostility and brutal oppression. Aboriginalism has taken different forms at different periods in the history of Aboriginal - white relations in Australia, and though it has never been as purely benign as its proponents have fondly believed, by the same token it has been one of the catalysts for the regeneration of Aboriginal people that is currently gathering an irresistable momentum.2 At one time Aboriginal activists angrily denounced the paternalist pretensions and hypocrisy of Aboriginalism, when Aboriginalists had too much power. Now the balance of power has shifted, and it is time to see strategic opportunities in the sheer contradictoriness of the Aboriginalist enterprise. The issues raised by Eric Michaels' work are an excellent place to begin such a reexamination.

There are two forms of Aboriginalism which I will distinguish, since they have some distinctive ideological consequences that are relevant for the present article. In one form, Aboriginality is a single global construct, participated in in some mystical way by all Aboriginals from everywhere in Australia, at its extremes becoming a universal 'Savage Mind' which is a substratum of a universal history stored in an inaccessible but potent collective unconscious. In the other form, Aboriginality is partitioned off into innumerable entities each as self-contained as Aboriginality itself. The first form elides the many differences that are important to Aboriginal people. The second rules out on a priori grounds the possibility that Aboriginals could ever constitute an authentic group with their own common political and social culture, their own distinct interests.

In either form, Aboriginalism operates through two sets of premises about the dimensions of the Aboriginal mind which distinguish it absolutely from Western forms of thought. One concerns the 'Aboriginal' view of history and time. In this view 'our' own familiar linear view of time as a passage from past to present to future is merely a construct of Western (European) rationalism. For Aboriginals, it is said, this view is incomprehensible and utterly alien. Instead, for them, past and present intermingle in a multiplicity of patterns. As a result, both history and change or progress are equally unthinkable within Aboriginal thought (or truly Aboriginal thought - only contaminated Aboriginals will reveal the scale of their contamination by entertaining such ideas, which they can only have caught, like an infection, from interaction with Europeans). Along with this difference goes a different sense of reality. These primitive Aboriginals, the Aboriginalist version has it, occupy a mystical otherworld, full of significance though empty of material goods. This nonmaterial otherworld is their most precious, perhaps their only possession. So all other things may be freely taken from them, so long as that is left intact.

This version of Aboriginal thought is encapsulated in a single potent term, 'the Dream time' (or the variant 'Dreaming(s)' popularised by Stanner3). The 'Dreamtime' is a universe of meanings constructed purely according to Aboriginalist principles. It is offered as the official (and incomprehensible) English equivalent of 'the' untranslatable Aboriginal concept, even though the set of words used in Aboriginal languages normally does not include any allusion to or cognate of 'dream'.4 It has achieved wide currency, not only among nonanthropologists, who repeat the term with a warm sense of inwardness with Aboriginal culture, but also by many Aboriginals themselves, who have been told that 'Dreamtime/ing' is the official translation of this concept in their language, and have obligingly incorporated it into Aboriginal English. This naturalisation makes it especially potent as a carrier of Aboriginalism into the dominant culture, especially for sections of the Australian community that are most sympathetic to Aboriginal interests.

Eric Michaels' project was funded by the Australian Institute of Aboriginal Studies as a latter day contribution to Aboriginalism. He explains that his brief was to 'assess the impact of television on remote Aboriginal communities'5 where the single word 'impact' invokes the whole Aboriginalist complex, constructing traditional (remote) Aboriginals as reeling under yet another death-blow, this one perhaps at last fatal, from another wave of the genocidal assault. Michaels' title,The Aboriginal Invention of Television, polemically refuses this construction by inverting its terms. He seems to promise an Aboriginal invention of television, allowing the paradox to sit unexplained in his title. In the text that follows he pursues a version of this promise, attempting to project a complete media policy from the ingredients of traditional Aboriginal life. In a way it's like the strategy of the Aboriginal TV programme Barbekuieria, which projected an Australia in which Whites were the oppressed minority, under the domination of Aboriginals. The paradox in both cases opens up a space for radically rethinking dominant relations of power as they position Aboriginal and non Aboriginal people.

In terms of this paradox, the Aboriginalist commitment to cultural maintenance and preservation of traditional languages becomes problematised. Transformation, change, participation in the contemporary world all come onto the agenda. It is precisely here, however, that Aboriginalism shows its protean resilience. Michaels sets out 3 models of Media in Aboriginal Development: cultural maintenance, pan-Aboriginalisation, and assimilation.6 Although he withholds labels from his own work, his bias is clearly towards the 'cultural maintenance' model, even though in a form that is highly flexible and syncretic. So although he trenchantly attacks the convention of the 'ethnographic present', which constructs traditional society in a timeless abstract past, and he notes that Aboriginal life today is very different, nonetheless he acknowledges the "persistence of the cultural forms, admittedly in transformation"7 and uses this as the reason to give a rapid summary of 'traditional' society, as the source of premises for what will constitute an 'Aboriginal' use of contemporary media.

This is where the speed and mobility of the cowboy anthropologist is perhaps a source of vulnerability. Even (especially?) an intellectual nomad like Michaels needs a notion of authenticity, to be found in a belief in some unexamined and unavailable substratum of Aboriginality (his wry critique of Bruce Chatwin's more extreme form of 'paraethnography' distances himself from Chatwin's practices, but not from the impulses that drive it8). In Invention he devotes perhaps 8 pages scattered through the book to an account of 'traditional' society, as described by traditional anthropologists. His bibliographical saddle-bags are as light as is decent (only 8 references to anthropological classics), but there is a price for this easy freedom. That price is the ghostly survival of Aboriginalism in Michaels' own core premises, in a confidently attenuated form that protects it from inspection or critique.

Language and Truth

Language was a key item in his brief, as it is in the Aboriginalist complex. In this scheme, language and culture are so close as to be virtually inseparable, and cultural maintenance and preservation of the indigenous languages are seen as almost equally threatened by the advent of television. Michaels more or less accepts this position, which corresponds to a concern for the survival of their language that was expressed by the people he was working with. He can only distance himself from the Aboriginalist implications of the position by relying on untheorised common sense and perhaps a sturdy indifference to the whole topic. This still leaves him with a 'cultural maintenance' model that is uneasy about any form of mass communication, and unable to produce an indigenous model that is valid for more than one language area, so that the task of devising a policy to fill a satellite footprint is beyond him, for all his enthusiasm for the possibilities of aboriginalising modern media.

The blockage here comes I think from his residual deference to the traditional constitution of this intellectual field. In this case what he is recognising is the authority of structuralist linguistics, which has virtually monopolised the study of Aboriginal languages in Australia, just as it has done with Amerindian languages in America, and with languages of other 'primitive' peoples wherever forms of 'Orientalism' have held sway. Structural linguistics in many of its practices seems entirely mechanistic, and quite different from the metaphysics of Aboriginalism, but the two are complementary, as the analogous work of the Boas school in America shows. Out of this tradition came structuralist descriptions of many 'threatened' indigenous languages and also the 'Sapir-Whorf' hypothesis, which links forms of language, culture and thought into a single complex which is consistent within itself but incommensurable with any other such complex. Thus the language-culture complex of 'primitive' peoples is elaborate and incomprehensible to us moderns, as ours is to them. Whorf9 added the nice twist that the 'primitive' Hopi invented Einstein sometime before Columbus, just as Michaels' Aboriginals invented television. For both, the paradox helps to conceal the uncomfortable resemblance of their move to a founding premise of Aboriginalism.

The problem with the two main forms of structuralism, the mechanistic and the metaphysical, is that they both establish linguistic boundaries as absolute, using this to establish the basis for a theory of essentially static and unchanging cultures and societies. Michaels, however, does not stay long within the structuralist paradigm. On the contrary, his strength as a theorist is that he largely ignores structuralism, in favour of an opposing approach that is so rare within Australian linguistics that it would not even be called linguistics by most practitioners of that discipline. (Significant works in this vein include Sansom's classic study of Darwin fringe-dwellers, and Muecke's use of discourse analysis,10 both of them cited by Michaels). At the core of this approach is a concern with process, not structure. Elsewhere (outside the Australian linguistic establishment) this kind of approach is labelled variously 'post-structuralism', 'discourse analysis', 'sociology of language', or 'ethnomethodology'. Each of these practices has a relativistic form, but all of them describe phenomena that are concerned with power and its maintenance or negotiation, which therefore pass beyond the boundaries laid down by linguistic specificity, if that is how power and interaction flow.

An approach like this is already close to a political economy of information that can be applied to advanced communications systems such as satellite technology. Michaels works easily with this model, which is much more compatible than the other with claims for an "Aboriginal invention of television". In fact it would easily underpin a 'pan-Aboriginalisation' model, of the kind that Michaels specifically rejects in favour of his preferred model that insists on the absolute primacy of the local language group and the local community.

The problem of the role of Aboriginal languages in the mass media is a contentious and difficult one, and his position on it - essentially defering to Aboriginal communities to resolve it - is impeccable, as far as it goes. But the structuralist linguistic tradition can be invoked to construct a broader and more usefully contradictory space within which to debate the problem. It isn't the case that this tradition makes pan-Aboriginalism unthinkable or unAboriginal. Only monolingualism on the English model is unAboriginal. Adjacent languages in precontact Australia normally had a substantial amount of shared vocabulary. They were often closely related either genetically or by diffusion, and had many common features of grammar and sound. They did not function as incommensurable universes. On the contrary they were designed to have very leaky boundaries. Their construction of difference was complemented by strategies for constructing provisional alliances, primarily through kinds of exchange along major communication networks, what Michaels prefers to call 'dreaming tracks'.11

These networks did not construct a single panAboriginal community across Australia, but they meant that individual Aboriginals were not locked into self-contained, culturally unique communities either. And while it is true that a single minority Aboriginal nation within the Australian community could not preexist the European invasion, there were many continuities across the land in precontact Australia, as Willmot12 has argued, and as Michaels agrees. The languages are commonly divided into two broad groups, Pama-Nyungan and nonPama-Nyungan, and these two are speculatively linked to a common ancestor, ProtoAustralian.13 This ur language guarantees no more unity than Indo-European does to the nations of Europe, Australia, the Americas and the Indian subcontinent, but it provides the basis for the similar patternings found in many Aboriginal languages. Perhaps more to the point, it gives a substratum of commonality to 'Aboriginal Englishes', the set of forms that map Aboriginal grammatical forms and some of their patterns of sense-making onto an English base. This language, treated with contempt by many English speakers as a debased form of English, is an 'authentic' creation of Aboriginal people. It is comprehensible by Aboriginal groups throughout Australia but it also marks differences of region and level of discourse precisely, as Aboriginal languages do. The role in public broadcasting of this Kriol, as it is termed by linguists, is yet to be determined, and it could only be determined by Aboriginal people. But its existence still points to one possibility, a form of panAboriginalisation, that should remain on the agenda, in spite of Michaels' tendency to label all such moves as irretrievably unAboriginal.

Behind the issue of linguistic difference is the issue of the politics of truth, occluded in different ways and to different degrees by the two forms of structuralism. Since the politics of truth is concerned with processes of negotiation and struggle, out of which 'truth' emerges as a social construct, as an effect of discourse, it is to process-forms of linguistic theory that we need to turn first in order to develop an understanding of Aboriginal forms of truth. Michaels draws on this tradition to construct a convincing framework for outlining an Aboriginal theory of truth and its material conditions. One strand of his thinking emphasises the role of secret knowledge within the Aboriginal textual economy, the strict controls on who can know and speak or hear various kinds and levels of knowledge. This is received wisdom in traditional anthropology, where it is commonly represented, in the static fashion of structuralism, as timeless patterns of knowledge that can only degenerate from their pristine form, without any recognition that they were always only constituted through discourse and can always be reconstituted in the same way.

Michaels uses this tradition in a flexible way, noting the instability of these classifications as knowledge passes from group to group or is reprocessed under the pressure of new needs in new situations. In one instance whose implications he doesn't fully explore, the taboo on showing representations of the dead was put aside, when an old video of a Fire ceremony was shown, to the distress of some women who were not part of the discursive process that lead to the reclassification. Even closer to home, his note to For a Cultural Future states that the Yuendumu council was not entirely happy with the form of that book, which he had submitted to them for their judgment. He scrupulously published this disauthentication, but also proceeded with the book, signalling his own ambiguous sense of his relationship to collective processes of decision-making. The classic study of Aboriginal methods of constructing truth is Sansom's The Camp at Wallaby Cross, which examined how 'the word', the official form of truth for the community, is carefully and meticulously constructed and established, then equally forcibly policed, so that alternative versions are removed from circulation, becoming unspeakable. Sansom's work is especially persuasive because it meticulously demonstrates these processes operating in a fringe-dwellers' camp near Darwin, not in some reconstructed pre-contact Aboriginal society. From these materials he creates the picture of a monolithic and overwhelmingly successful truth-machine, working through principles that are fully recognisable to English speakers but with a perfection that is normally beyond us. In this respect his work still operates under the shadow of Aboriginalism, though he has put those principles at risk by his choice of his object for research. Michaels' version of Aboriginal truth-processes is marked by the same traces and similar contradictions.

But traditional languages as studied by structuralist linguistics14 contain other information that is relevant to Michaels' project. Their grammars encode precisely that taken-for-granted set of assumptions about reality that could function as Aboriginal premises in the new form of cultural policy that Michaels tried to envisage. Warlpiri, the language of the Yuendumu people, is a typical Western Desert language. It incorporates its theory of truth and history most elaborately in the rich set of elements in the verbal complex, elements which correspond to English auxiliaries of tense and modality which also modify verbs and classify and situate the event that the verb refers to.

'Tense' and 'modality' in English are often said to refer to time and truth respectively.15 In Western desert languages as in English the two dimensions are inseperable. They have 5 primary forms of the verb: imperative, future, potential, past and present. In this scheme, it is interesting to note that in spite of what Aboriginalism and the concept of 'Dreamtime' would lead us to expect, it is the future that is more precisely elaborated than the past. There are forms that further distinguish the immediate future from the remote future, and the immediate past from the remote past. Far from linear time being unthinkable within this language, Warlpiri organises time in a linear way with greater precision than English. Future forms can be used for events in the past, where they represent a future event in relation to a fixed point in the past, located within the speech event.

Within this temporal grid, there are further distinctions that require a sensitive knowledge of truth-conditions for proper use, since they rely on fine grades of modality. For instance, the imperative, the future, and the potential tenses or moods can all refer to future events, but with differing degrees of certainty, that certainty being a function of the authority and conviction of the speaker and the predictability of the event: both functions being dependent on particular social and cultural knowledge. And running across the various tenses there is a recurring distinction between definite events (located in a specific space-time) and indefinite events (performed habitually, customary actions), corresponding broadly to the distinction between empirical and general truths in the epistemology of English. Generally speaking, time is organised in the same way as space, using the same set of adverbs. Laughren16 has shown that spatial and directional terms in Warlpiri make up an elaborate geometrical system. For both space and time the picture is the same, as encoded in the grammar itself. In neither dimension is there a single kind of map, a single version of reality. Both employ at least two kinds of truth. One corresponds to objective truth in western epistemologies, and is at least as fully elaborated as in English. The other corresponds to forms of subjective truth, reflecting either the specific location of the event or its valuation.

What the grammar encodes is a set of multiple grids. One of these constructs a version of space and time that is entirely compatible with so-called Western linear thought. But the same grammar also encodes other forms of truth. What we have, then, is neither an inherent incomprehension of western truth (including the modalities of technological thought) built into the very language itself, nor any limitation to a single scheme, a single form of truth. This picture of a potential multiplicity of truths is also one aspect of the construction of 'the word', as reported by Sansom, where discursive heterogeneity had to be ordered and controlled and reduced to unity not by the prior unity of Aboriginal thought, but by specific discursive practices, maintained and policed by the whole group. Language does not override politics. On the contrary, language should be seen as a set of discursive resources whose deployment and use are determined by social and political considerations, which then leave traces in the language itself.

Michaels' Aboriginalist premises leave him unable to see the complete and complex trajectory of Aboriginal video as fully thinkable within the scope of a fully Aboriginal (Warlpiri) mode of thought. We can see this clearly in the tangled peroration of For a Cultural Future where he anguishes over the criticism that he is after all romantic and nostalgic. In practice he is simply failing to believe in the Warlpiri's right to inhabit the cultural space (including the use of video) that they have already effectively claimed. 'In the confrontation between Dreamtime and Ourtime, what future is possible?' he asks,17 as though the two strictly should never meet, as though the Warlpiri people he had worked with for 3 years hadn't lived in 'his' (our) time for all that time. And near his climax he writes:

A cultural future can only result from political resistance. It will not be founded on any appeal to nostalgia: not nostalgia for a past whose existence will always be obscure and unknown, nor a nostalgia we project into a future conceived only in terms of the convoluted temporalities of our own present. The tenses are difficult to follow here - but in a sense, that is precisely the critical responsibility now before us. Francis Jupurrula Kelly makes, is making, television at Yuendumu. He intends to continue, and so assure a cultural future for Warlpiri people. His tapes and broadcasts reach forwards and backwards through various temporal orders, and attempt somehow to bridge the Dreaming and the historical. This, too, is a struggle which generates Jupurrula's art.18

It is sad to note how much of his anguish here is generated by his conviction that what Jupurrulu is doing is very odd and almost threatens Aboriginality itself, and that to describe its aberration he, Michaels, has to do violence to English on a similar scale. But Jupurrula's Warlpiri could allow him to play even more complex games with tenses than Michaels, if that was the only problem, and the movement between 'the Dreaming' and the historical is a normal and central preoccupation of Aboriginal culture, not an almost impossible attempt by the two Culture heroes and Marginal Men (Michaels and Jupurrula) to bridge two incompatible cultural universes.

The Truth of Art

During Michaels' field work at Yuendumu, the community was as excited by its involvement in contemporary painting in the 'Western Desert' style as in video production. This was seemingly fortuitous, unrelated to the developments in video, so it is only noted in passing in Invention, as a topic of conversation that helped to establish him in the community. But Michaels became interested in this development in its own right, and theorised it in similar terms to the way he analysed video. The most important event in this sphere was the major enterprise of painting the doors of the primary school. This move was initiated by local elders and encouraged by the headmaster, but Michaels was involved in a number of ways, especially in the later stages, where the enterprise connected with wider systems of circulation, provoking the problem of authenticity in a new terrain.

Michaels encouraged the making of a video of the 'Yuendumu Doors', as part of the video project. He also participated in the production of a book, which was ultimately published by the Institute of Aboriginal Studies in 1987, with an 'Afterword' written by him.19 This article describes something of his role in the production, though with a measure of understatement. From this account it appears that the elders of the community at first wanted to have photographs published as a book. Michaels acted as their agent for this enterprise, and tried out the idea with the Institute. He reported back that there was little chance of large profits, on the scale of their rivals, the Papunya artists, (at least if they were published by the Institute - not the most commercial of publishers, incidentally, if they had wanted a lucrative coffee-table publication). So as he explains it he supervised the production of a parallel set of canvases for sale on the art market, with himself as 'patron', in addition to the book itself.

He was quite conscious of some major contradictions arising out of his position in the enterprise.

In the exchanges with the men I became a student of painting, but also of the exchange system. No longer an expert, in media, anthropology or anything, I was a pre-initiate. An 'unitiated boy's' Dreaming design was made and was the first to be hung. My problem was to fulfill this role, but at the same time, to exercise a critical European eye and occasionally let the painters know what would probably be saleable and what wouldn't. This could be a painful process...20

This statement economically locates a number of fissures. One is between 'painting' (the sphere of traditional expertise, including both processes and 'content') and 'exchange', the sphere of economics, where the Aboriginal system may be different but is commensurable to the European one, and can be overridden without detriment, so long as Aboriginals do not lose in the deal. But this fissure is the same as exists between the hustler and the artist in the system of circulation within the contemporary art world. Michaels poses a dilemma for himself, between extreme humility (in his role as 'preinitiate') and decisive arbitration (deploying his practised European eye, and gambling on his ability to read current fashions and tastes). But he doesn't seem at all inhibited by his extreme humility, or even by his awareness of the huge contradiction. On the contrary, both contribute to his authority, in relation to both communities: to Europeans, who lack his authenticating humility, and to Aboriginals, who don't know the international market. This contradiction, mediated in some unexplained way within himself, is also the reason for the epoch-making stature of the work itself, as a major event in the history of world art, whose radical innovations are able to render post-modernism obsolete, along with all its ancillary 'isms', while remaining rooted in the unreachable depths of the very ancient. This is a similar claim as he made on behalf of Yuendumu videos in For a Cultural Future, but he is able to make it much more strongly with greater conviction about the art.

Only the 'Afterword' is signed by Michaels, but I see traces of his work in many other places, unacknowledged as in the best Aboriginalist tradition, though perhaps in this case the community approved of his self-effacement. For instance, the Introduction is not signed, but it has all the marks of Michaels' brand of Aboriginalist discourse. It describes the 'keys' attached to each painting in the following terms:

The keys 'identify', at one level, some of the features in the paintings. These keys are not intended to 'explain' the paintings - full understanding of the designs comes only to full members of Warlpiri society - but rather to indicate to the reader that these are paintings loaded with meaning. The translations serve a similar function. Both translations and keys provide only an imperfect, infinitesimal glimpse into the complexities of Warlpiri religion and society.21

The notion that the keys are intended to signify meaningfulness itself is pure Michaelese. Equally Aboriginalist (and unAboriginal) is the polysyllabic 'infinitesimal' and the idea it contains, that the Warlpiri normal perception encompasses a single object so vast that it corresponds to a normal object glimpsed with abnormal brevity by Europeans.

We may be unsure as to who exactly 'intends' this function of the keys and translations, but in the book that follows we have to become aware of a devious communication strategy. The book consists of 30 colour plates each devoted to one of the doors. There is a text in Warlpiri that precedes the colour plate, then a translation in English, plus a diagram of the painting with a few 'keys'. The effect of this combination is certainly not to 'explain' in any helpful way. The Warlpiri text is usually longer than the English translation, leaving non-Warlpiri speakers (the intended readers, since Warlpiri people can see the actual doors much more easily than they can buy the book) supposing that something is being said from which they are excluded. The key on its own is normally close to useless in decoding the text, since it normally gives only a partial cast-list in a narrative, itself to be glossed by another discourse, both of which remain outside the text. So the verbal text is essential to supplement the 'key'. But on most occasions, the verbal text barely tells the story, and never glosses its significance.

At every point, then, the 'aids' to interpretation collapse, and much of the book is devoted to different ways of excluding English-speaking readers from the core texts (the set of paintings). This signifies not 'meaningfulness' in general, but the Warlpiri's deliberate withholding of meaning from Europeans, in a cool and seductive form of cultural hostility: a specific, politically motivated suppression of meaning, not an incommunicable plenitude of meanings. This may seem similar to the Aboriginalist claim of the introduction, but there is an important difference. What we have here is not the inherent incomprehensibility of this vast universe of meanings, as far as Europeans are concerned, but specific strategies for mystifying them. These strategies are similar in effect to the primary discursive strategies of Aboriginal societies, where full knowledge is carefully controlled, and withheld from most members even of the core society in the interests of power and territoriality. The book as a whole has just as many ways of excluding most Warlpiri (the written text would make demands on normal Warlpiri levels of literacy, even with the Warlpiri text, and the Warlpiri versions still withhold most of what an interpreter would really need to make much of the paintings) though of course it wasn't primarily intended for such readers, in spite of nearly half its pages seeming intended exclusively for them. But for Warlpiri children, the Warlpiri text would have done its job sufficiently if it merely excluded their European teachers from entry into the text, signifying not so much a plenitude of Aboriginal meanings as a poverty of European capacities to make meaning (especially enjoyable if they scanned it with awed incomprehension).

The translations are homogeneous and smooth, with no traces of Aboriginal English or oral forms, and though no acknowledgement is made to any literate European, one suspects an invisible hand - Michaels again? Whoever is the translator uses the occasion to reinforce the authenticity of the key term of Aboriginalism, 'Dreaming'. So frequent is it that this text provides a useful instance for a specific analysis of its effects. There are two Warlpiri words that are translated as 'Dreaming' (always with a capital). The most common is 'jukurrpa' (normally not with a capital). But sometimes it translates 'kuruwarri'. 'Kuruwarri' in the Warlpiri consistently refers to representations - paintings, designs, body-paintings etc. 'Jukurrpa' equally consistently refers to what is represented by the painting, the central being, usually an animal or plant, behaving in the narrative as a person. The two words together make the basic semiotic distinction between signifier and signified, and neither refers to the process of signification itself. These are the primary distinctions that are overridden in the Aboriginalist term 'Dreaming', which systematically conflates myth and character and the semiosic process itself.

This process is not mystified in Warlpiri as a kind of 'dream'. Munn claimed, in her fascinating and influential Aboriginalist study of Warlbiri iconography, that many narratives were said to come in dreams, especially those by women, and she glosses djugurba (as she transliterates it) as 'ancestral times; dream; story' (and not ancestor figure).22 But 'jukurrpa' never refers to a dream or to a mode of time in this set of texts. When it is translated as 'Dreamtime' it is grammatically a noun in apposition to the figure it describes (eg 'In the Dreamtime Witchetty Grub came from Yarlukarri'23 translates 'Jukurrpa, jukurrpa yanurnu Ngarlkirdi, kanunjumparra Yarlukarrijangka', where 'jukurrpa' describes Ngarlkirdi the ancestral or totemic Witchetty grub, not the time of the action, 'yanurnu', which is in the normal narrative past.) Although Aboriginalists could defend a translation of 'jukurrpa' as dreamtime on the grounds that these events clearly didn't happen in the present, this is not how the term functions grammatically or semantically.

But the worst crime against both Warlpiri thought and the English language is when 'Dreaming' translates the agent of a verb (eg 'The Dreaming travelled', a sentence that is unspeakable and incomprehensible in English). The translation normally though not invariably follows the convention of translating 'jukurrpa' as 'Dreaming' as subject of a verb when it refers to a plant or animal behaving in the narrative like a human. For example this is said of Yarla the yam, subject of door 8: 'From here the Dreaming, who was originally a person, spread out like yam roots and became them'.24 This sentence (translating Yapawiyi, jukurrpalpa wapaja) deals with the problematic transformation of jukurrpa into material yam. This is clearly a complex concept which departs from everyday concepts of 'reality', but communication is not helped by the added difficulties of the English term 'Dreaming', defying English grammar by using a gerund to refer to the individual agent of a physical action. In For a Cultural Future Michaels does discuss this term, and distances himself explicitly from its typical uses: '"The Dreaming" has become appropriated to describe a cultural product (and to sign a commodity) which seems to me very much an invention of Western imagination.'25 This is well said, so it is surprising that he continues to use the term without further apology through the rest of this book, as he did even in his last published article.26

The Warlpiri who helped to construct the book successfully tricked the Whitefellas by giving them nothing except a (perfectly correct) belief that they aren't being told anything. We cannot now learn whether Michaels knew more than he let on, and was part of the trick. Nor do we know whether the Warlpiri went along with the mystificatory use of 'dreaming', happy to be complicit in the Aboriginalist nonsense for their own purposes. But it does seem likely that Michaels was convinced that this kind of semiotic trickery is a guarantee of a new kind of truth. He concludes his article by quoting Baudrillard's celebration of post-modernism's transcendance of authenticity, but sees Aboriginal art as able to provoke in a new form the possibility of a 'semiotics of authenticity'.27 The word 'authentic' and its derivatives appear 9 times in this article. Along with inexplicable meaningfulness, this is the primary signified that he sees in this art: a guarantee of absolute truth and inexhaustible meaning, as the inaccessible deep structure behind a 'look' that is as up-to-date as the latest fashion from New York or Paris, comprehensively demolishing Post-modernism's twin spectres of relativism and the emptying out of meaning.

In a sense he is right, insofar as the Warlpiri's deviousness unselfconsciously and persuasively signifies their taken-for-granted conviction that the Post-modernist dilemma is irrelevant to their problems. In a situation of social and cultural struggle they do not need to worry about the emptying out of meaning from a common language or basis of truth. Instead they need to resist the systematic appropriation of their culture along with their land. But since the problems are so different, there is the risk that would-be allies like Michaels are liable to be underinformed, unable to decode the complex meanings because they do not suppose that this is a proper task. There is a cost here, one which affects Michaels' central enterprise of constructing an Aboriginal media and culture policy (though it may be that this enterprise is precisely what the Warlpiri wanted to disrupt).

The relation of these texts to the country constituted the primary basis of their authenticity according to Michaels. These texts were situational, the actual (irreplaceable, unique and hence authentic) doors in the actual community (Michaels carefully distinguishes these authentic texts from the canvasses that also circulate as official copies, and from the photographs in the book). The book contains a number of maps. One gives the location of the places in the landscape referred to in the paintings. Another gives the location of the doors in the school. It turns out that the order of the paintings in the book does not follow any linear route around the paintings (door 1 is next to door 28, door 2 is some way off, between doors 5 and 20, and so on). The location of the doors shows an interesting relationship to the localities they refer to. In general, the compass points are reversed. So door 1 faces north, but refers to Kirrirdi, which is south, and so on. When a door is on the outer side of a building, this means that a spectator will be looking towards the relevant part of the country, but this does not apply when the door is on an inner wall.

The strategy here is rather like the practice adopted in war time of changing the direction of street signs to confuse invading armies. The doors themselves seem systematically to invert the real directions (though the full rationale for their order remains obscure, and would need further study to elucidate it). The book then scrambles even this order, so that someone at Yuendumu has a chance to work out the correct orientation, but readers have no idea of where anything is to be found - unless they attend closely to the map and notice the pattern, which nothing in the book encourages them to do. What the move shows unmistakeably is yet again the spatial dexterity of Aboriginal people. Not only do they have an extraordinary grasp of the landscape over a great distance (aided by and recorded in precisely such mnemonics as these paintings) but they can then scramble these maps, relying on their ability as readers to unscramble them again, to restore them to their correct order. The first skill has been recognised since the days of the early settlement, by countless Europeans who got hopelessly lost in the landscapes that the Aboriginals knew so well. The second, however, attributes to them both a transformational facility and a deviousness that Aboriginalism has not wanted to acknowledge.

Michaels did not keep the two media of painting and video separate. Part of his project was to draw on traditional forms and their transformations as the basis for an Aboriginal media policy. So an acryllic painting by Andrew Japaljarri Spencer, 'Satellite dreaming', which serves as the logo for Yuendumu broadcasts is also used by Michaels as the basis for an extrapolation to communication networks in a mass broadcasting system. Its title and use by Yuendumu broadcasters are sufficient to give it exceptional status as a policy document representing Aboriginal views. Michaels' procedure here is entirely proper in taking it very seriously indeed. This is quite a different matter from the classic Aboriginalists' tactic of extrapolating a whole world view from their own understanding of the meaning of the culture.

However there is a danger that Michaels' wish to give due attention to this important document can lead to his assigning it an absolute status, because of its overwhelming authenticity and truth to Aboriginal values. This painting shows a set of concentric circles at the centre, representing Yuendumu, with 16 tracks radiating out to 16 other sites, of different sizes, all consisting of concentric circles. This uses classic Warlpiri iconography28 to represent Yuendumu as the complex centre of the social community and the media universe. Its credentials in the tradition seem impeccable. But it is carefully emptied out of specific meanings as encoded in accompanying narratives, and thus is an instance of a 'post-modern' Western Desert acrylic, employing 'weak nominative networks' to use Lyotard's phrase.29 The paradox is that this painting signifies the supremacy of Yuendumu in its content, but its form signifies Aboriginality itself. Its ideological claims are transparent, and cannot be ignored if it is to be taken seriously as a political statement. Michaels briefly notes how 'ethnocentric' it is, but we need to do more than note this and pass on. What is at stake is the fundamental policy issue, for Aboriginal media and for Aboriginal people in general, between the localism that Michaels espouses and the relatively panAboriginal policy that he attacks in Willmot's policy proposals.

In practice, as I have suggested earlier, traditional Aboriginal language and culture provide models and mandate for both constructions of Aboriginality, the localist and the translocalist. They transmit a contradictory set of models because this was as much a survival issue for precontact Aboriginals as it is for their contemporary descendants. In art the copresence of two contradictory principles is not only recognised by Michaels, it is precisely the quality that in his view gave the Yuendumu doors their importance as a 'definitive series of western desert designs', as a 'complete text, a primer'.30 Earlier paintings on single topics gave a misleading impression of the overall scope and intricacy of the epic map in which individual artists situate each work. This larger map functions like Metz's 'grande syntagmatique'31 as an overriding principle of organisation and interpetation. Michaels observes, of some of the early Yuendumu canvases, that they 'seemed uninteresting because they were actually small corners of far larger ones.'32 It is the smaller texts, within this tradition, that tend to be organised in circular forms, radiating from a single centre. Works on a larger scale normally use a rectangular grid, still not comprehensive but infinitely extendable outside the arbitrary rectangular frame.

To illustrate this different principle of organisation I will use door 29, Yiwarrakurlu, about the Milky Way. This painting is divided in the middle by the stars of the milky way, depicted as a band not a circle. Above and below, the key tells us, are witi-poles, posts draped with vines used in initiation ceremonies. The accompanying text would give great comfort to Van Daniken and his tales of space travellers ("This story that I am telling is about my fathers in the Dreamtime who made the stars travel across the sky"33). What I want to emphasise is the implied scope of this text, covering a large stretch of territory, more than can be legitimately represented in a single painting or narrative ('I am only telling of that part of the Dreaming story which came from the north. This same Dreaming belongs to others. We all join up - Yarripilangu and Yarrungkanyi. It is a big Dreamtime story about the millions of stars which shine above us as we sleep'34). Underlying this implied network of sites of ownership of aspects of a single narrative is a map that probably does not coincide with the footprint of a satellite, but it has the right scale, and represents a useful and indigenous balance between localism and the needs of a larger Aboriginal community. It is interesting to note, incidentally, that there is a place in this text for a different version of a 'Satellite dreaming' - one that traced the route of a satellite through the milky way, locating individual sites where earth station-jukurrpas paused and left their solidified remains.

A conception on this scale explicitly underpins Willmot's proposal, which Michaels took issue with.35 Michaels' criticism that Willmot didn't consult adequately with individual communities is fair comment, but that is a different matter to his assumption that Aboriginal society is intrinsically and exclusively organised at the level of the local group. The lesson from art confirms the lesson from language. Traditional Aboriginal society had a set of contradictory models available for making sense of time and place, a repertoire to be used for different purposes and in different contexts, not a single model that prescribed and determined all possibilities of the thinkable. In saying this I do not wish to fall into the Aboriginalist fault of using an analysis of Aboriginal language and culture as a basis for telling Aboriginals what they should think, about media policy or anything else. I only wish to disrupt the Aboriginalist strategy that removes whole classes of option from debate by attaching to them the potent label 'un-Aboriginal'.

The Threat of the Real

Perhaps the most powerfully disabling premise of Aboriginalism is that Aboriginals are definitionally unable to grasp 'reality', immersed as they are in the womb of 'dreamtime'. Michaels replicates something of this assumption in his discussion in Invention of the likely invisible impact of western media on traditional Aboriginals. To set this up he repeats some anecdotes (which he labels as such) which illustrate the well-worn theme of Aboriginal naive and ridiculous misreadings of realist texts. He adds his own contribution, an amusing tale of the response of Aboriginal school children to a 'Dracula' video. He observes, in his common sense mode, that the whole episode could equally well have occurred in an 'eastern suburbs primary school'.36 He is probably right, as my own research indicates.37 What is disappointing, then, is that he proceeds as though he hadn't noticed this disclaimer, and takes for granted that there is a serious problem here. He accepts the word of Aboriginalists that fiction is an alien form for Aboriginal culture. This is the necessary correlative of the Aboriginalist belief that Aboriginals' grasp of reality is constrained by the dreamtime, a fictional world so totally conflated with reality that the separate categories of realism and fiction cannot exist.

I don't have independent research evidence that Michaels is wrong about the Yuendumu people, and in fact the Aboriginal construction of reality and their strategies for assigning reality values to Western media forms are complex and important topics that would repay further study (it's one of the areas that the Institute might have expected Michaels to study as part of his initial brief). My point is that Michaels himself has no evidence for this acute realism-defect among Aboriginal people, and in fact provides incidental evidence to the contrary. Yet he is still able to say categorically:

The conclusion must be that the important fact is not the introduction of video or television to traditional Aboriginal society, but the introduction of western fictional genres.38


In the slightly later For a Cultural Future he put this even more strongly: "I did not yet realise the centrality of the prohibition against fiction for the Warlpiri oral tradition".39

It might be said in Michaels' defence that he wasn't really interested in Aboriginal reception of western forms, and this is true, though there is still a danger that his pronouncements may be taken to be more reliable than they are, coming as they do from the foremost authority on Aboriginals and television. His real area of concern is with Aboriginal strategies of production and the complex reception strategies that complete the meaning of such texts. This is where his most exciting and important work has been done. It is still marked to some extent by his Aboriginalist assumptions, but it more than repays further attention, to disentangle it from that tradition and to follow up some of its implications.

The first topic I will consider is the set of genres that Michaels identifies, from a few hundred hours of tape. He lists 11 genres, but virtually all are marked by what he noted with some disapproval as excessive realism, their reliance on documentary forms, in a 'direct cinema' style.40 In this form 'the events that were filmed were happening irrespective of the camera's presence'. This characteristic confirmed his belief that these people were incapable of the concept of fiction, though it is equally likely that it is only the expensive and cumbrous strategies of western illusionism that they had neither the time nor resources to aspire to.

He does not seem to take into account the modality value that seems, with good reason, to have been assigned to video as against painting in a total set of generic forms, reconstituted to include video. The literality of video causes acute and possibly insoluble problems for texts whose discursive strategy requires layered entry. Michaels lists these problems, all of them concerned in one way or another with the ambiguities and elisions that protect secret or restricted knowledge. This is where the acrylic art tradition comes into its own, able to communicate and conceal this whole important class of meanings without rupture. The problems that would exist if video were the only form of communication do not present difficulties if video coexists with art as a major popular form for the community. This is precisely what is happening at Yuendumu now, in his account. Problems only arise if video is given the kind of priority and scope for such communities as it has for urban European communities. The test case for Michaels, which is the climactic example in For a Cultural Future, was a performance of a Fire ceremony, where problems arose both with an ethnographic film recorded 20 years earlier and with a contemporary Aboriginal attempt to record it on video. Most of the problems that arose in this case were the result of the imperialist premises of ethnographic film clashing with core systems of power / knowledge of Aboriginal society. They show the consequences of a misguided Aboriginalist attempt to overload video genres with inappropriate content, owing to its assumption that if a mass media system cannot carry this content, then the whole goal of 'cultural maintenance' is jeopardised.

But precisely because he was opportunistic, Michaels was responsive to what Aboriginal people wanted to present on video, and for this reason he was able to provide some important insights into Warlpiri discursive regimes, the basic categories that organise production and reception. One case study that he reports concerns the genre of 'Community meetings'. He calls this 'one of the most politically important genres to develop'.41 In the case that he studies, the community pursued a dispute with the Education Department in Alice Springs. A group made the four-hour journey to Alice Springs, and the Yuedumu Video unit taped the whole meeting, as did representatives of CAAMA the Aboriginal media unit, and the ABC. Dave Japanangka articulated the rationale for this procedure: 'We want this to be videotaped so that we can prove we really did come here, and so the Education department can't lie about what was said.'42

As Michaels reports it, the community edited down 12 hours of meeting to a single 3-hour tape, whose technical quality was mediocre but whose reach was impressive (approximately 4500 people saw it within two weeks). This tape "galvanised audiences, who sat riveted for three hours, in almost every community I observed".43 A key reason he gives, which picks up Japanangka's comment, is "authenticity": "the tapes were unquestionably authentic because of the images and because it was known that they were taken by Aboriginal camerapeople". But this is a very different kind of authenticity from what he valued in the Yuendumu doors. It makes use of the evidential value of video images in accordance with familiar Western regimes of truth, using this product of European civilisation to confront another endemic quality of that civilisation, namely the propensity of its official representatives to lie on a grand scale. It would be difficult in all of this to sustain the argument that these simple Aboriginals did not really understand how complex 'reality' is, and that their use of video was inappropriate or inept.

Even more interesting is the description and analysis of the Coniston tape. This was an important story for Yuendumu people, the story of a massacre that was a decisive moment in their history. It is discussed at length in both Invention and For a Cultural Future, and I will not attempt to do justice to his nuanced and convincing analysis. I will only draw attention to a few points that are relevant to my argument. First there is his awareness of the carefully constructed nature of the 'truth' of this text, arising in a context of negotiation and struggle. He notes that the text had to be composed against a background of competing versions within the Aboriginal community, as well as against 'White' history. 'Whose version finally emerges as 'true' may be of some consequence to the community,' he observes, with a wry realism in the understatement. He describes in some detail the strategies for giving this text its Aboriginal authenticity, including the adaptation of the traditional dual authority of the division between kirda and kurdungurlu, teller or owner and guardian or corrector, and the use of 23 non-participants as witnesses to the truth of the narrative. The elaborate apparatus for ensuring truth signals an obsession which recognises how difficult truth is to sustain, and how important it is. Michaels' narrative of this process is an important contribution to our understanding of the kinds of 'truth' that may be possessed by the texts that oral history has to work with. His discussion is no less interesting because it raises issues of truth, history and the production of memory that have been analysed with contemporary western genres and media.44

Another quality of the video that he comments interestingly on is its use of what seem by Western conventions to be excessively long landscape pans. He considered the possibility that these are the products of naive film-making. But Jupurrulu the Aboriginal camera person had a convincing alternative explanation for this feature:

In interview, Jupurrula provided a purposive explanation for every motion of his camera. 'This is where those policemen came over that hill', or 'that is where dreamtime figures are in that tree', or 'this is the track old Japanangka came round'. Explained in this fashion, it becomes clear that for the Warlpiri cameraman (and presumably his Warlpiri viewers) the camera is tracking inhabitants of the landscape, historical and mythical figures who reside there, but are not apparent to normal vision. This attention to landscape is so persistent in all Aboriginal video-making that it seems unlikely that it can be analysed as naive or unintentional.45

This technique invites comparisons with Charles Chauvel's 'locationism', as deployed in his definitive filmic representation of Aboriginals, Jedda.46 Michaels' commentary also allows us to see the incorporation into a seemingly realist text of certain discursive strategies and forms of truth from other Aboriginal genres. The visual image, specifically through the redundant camera movement, signifies the site of intertextuality, where other discourses intersect (or perhaps a hierarchically ordered set of discourses, differentially distributed amongst the audience in the way that is typical of Aboriginal discursive forms). These texts have different kinds of modality and truth, including the historical and the 'mythological'. The result is a single, stratified text that manages distinct orders of meaning and different specifications of the real, while concealing that that is what it is doing. In this respect it functions very like the book of the Yuendumu doors, and the doors themselves. It produces an effect of the real which is far from simple or naive, though at the same time it is culturally specific.

Michaels had every reason to be delighted with the emergence of this text and this event, out of a serendipity that owed not a little to his sometimes crassly interventionist style of research, and to the trust and good will that he generated among the Warlpiri people of Yuendumu. In spite of the risk of failing to respect the virtues of sound scholarship and thereby tending to corrupt the young, I want to point out how useful Michaels' intellectual imperfections often were. In his last published article, his review of Bruce Chatwyn's Songlines, Michaels notes that Chatwyn had been sneakily reading rather a lot of traditional anthropology, and that in this unexamined and unacknowledged form it had done bad things to his intellectual formation. He proposes the label of 'para-ethnography' to apply to Chatwyn, so that anthropology can return the compliment, inspecting the likes of Chatwyn as aberrant specimens of the discipline itself. But Michaels himself occupied a terrain between ethnography and para-ethnography, scuttling between the pre-modern and the postmodern, at his most illuminating when the contradictions were most acute. In all this his idiosyncratic Aboriginalism played a crucial role. I have perhaps overemphasised the extent to which its comforting mystifications constantly intruded into his project, annulling its radical potential, shadowing the discourse of Aboriginals with the discourse of Aboriginalism. It's equally the case that without the Aboriginalist tradition Michaels would never have got to Yuendumu, and if he had no one would have read him as attentively as now they must.


1. See Edward Said Orientalism, (Harmondsworth: Penguin, 1985). The connection to Aboriginals is made by Vijay Mishra.

2. Among many commentators tracing and driving this change of attitudes is Henry Reynolds , as in The Law of the Land (Ringwood: Penguin, 1987.)

3. W Stanner, After the Dreaming (Sydney: Angus and Robertson, 1969) was an influential formulation of this concept. The term was prominent in the work of two of the most influential Aboriginalist studies of the Warlpiri, M J Meggit, The Desert People (Sydney: Angus and Robertson, 1962) and Nancy Munn Walbiri Iconography (Ithaca: Cornell University Press, 1973).

4. See Peter Sutton, Dreamings: the art of Aboriginal Australia (Ringwood: Viking, 1988), p.15.

5. Eric Michaels, The Aboriginal Invention of Television (Canberra: Australian Institute of Aboriginal Studies, 1986), p. xiv.

6. Ibid, p. xvii.

7. Ibid, p. 2.

8. Eric Michaels, "Para-ethnography", Art and Text, n. 37 (1988), p. 49.

9. B L Whorf, Language, Thought and Reality, (Cambridge, Mass: MIT Press, 1956).

10. Basil Sansom , The Camp at Wallaby Cross (Canberra: AIAS, 1980) and Benterrak, K, Muecke S, and Roe, P. Reading the Country (Fremantle: Fremantle Arts Centre Press, 1985).

11. Michaels, "Para-ethnography'", p. 48.

12. Eric Willmot, No Silent Land, (Canberra: Government Printers, 1985).

13. R M W Dixon, The Languages of Australia (Cambridge: Cambridge University Press, 1980).

14. Examples that I have drawn on for this article include studies of the Western Desert group of languages: Laurie Reece, Grammar of the Wailbri Language of Central Australia (Sydney: Oceania Linguistics Monographs, 1970); Wilf Douglas, An Introduction to the Western Desert Language (Sydney: Oceania linguistics monographs, 1964) and Aimee Glass and Dorothy Hackett, Pitjitjantjara Grammar (Canberra: AIAS, 1970). Michaels' own fullest published account of his limited and problematic encounter with the Warlbiri language is contained in "Ask a foolish question: On the methodologies of Cross-Cultural Media Research", Australian Journal of Cultural Studies, v. 3, n. :2 (1985), where he describes his heavy reliance on both professional (structuralist) linguists and interpreters, with sometimes comic results.

15. But for the argument that tense and modality are inseparable in English, see Gunther Kress and Bob Hodge, Language as Ideology (London: Routledge and Kegan Paul, 1979) and Bob Hodge and Gunther Kress, Social Semiotics (Cambridge: Polity Press, 1988).

16. Quoted in Dixon, p. 108.

17. Eric Michaels For a Cultural Future, (Sydney: Artspace, 1987) p. 71.

18. Ibid, p. 78.

19. Eric Michaels, "Afterword'", in Warlukurlungu Artists association Kuruwarri Yuendumu Doors, (Canberra: AIAS, 1987).

20. Ibid, p. 139.

21. Warlukurlungu Artists, p.2.

22. Munn, op cit., p. 28. The missionary-linguist preferred to emphasise a religious meaning as primary, translating it as "the Eternal" (his capital), op cit., p. 188.

23. Warlukurlungu Artists, p. 55.

24. Ibid, p. 47.

25. Michaels, Cultural Future, p. 29. Stephen Muecke is another important 'postmodern' theorist of Aboriginal discourse who is strangely in thrall to the category of 'the Dreaming'. In 'Aboriginal oral narrative in Ideological contexts' (in I Reid and S Gunew, Not the Whole Story, Deakin: Deakin University Press, 1983) he argues that 'the dreaming' functions as 'the constant supplementary signified' of traditional Aboriginal narratives, and then sees this as part of their ideological meaning.

26. Michaels, "Para-ethnography".

27. Michaels, "Afterword", p. 142.

28. Munn, op cit.

29. J-F Lyotard The Differend, (Manchester: Manchester University Press, 1988).

30. Michaels, "Afterword", p.135.

31. Christian Metz, Film Language: a Semiotics of the Cinema (London: Oxford University Press, 1974.)

32. Michaels, "Afterword" p. 137

33. Ibid, p. 127.

34. Ibid, p. 127.

35. Michaels, Invention.

36. Ibid, p. 47.

37. See Bob Hodge and David Tripp, Children and Television (Cambridge: Polity Press, 1986).

38. Michaels, Invention, p. 49.

39. Michaels, Cultural Future, pp. 43-44.

40. Michaels, Invention, p.60.

41. Ibid, p. 69.

42. Ibid.

43. Ibid, p.71.

44. See eg K Tribe, "History and the Production of Memories", Screen, v. 18., n. 4 (1977/78), as one example of the substantial body of relevant work on the problematic nature of 'realism' within contemporary discursive regimes.

45. Michaels, Invention, p. 63.

46. See eg Stuart Cunningham, "Chauvel, the last decade", Continuum , v. 1, n. 1 (1987).

New: 22 November, 1995 | Now: 9 May, 2015