Australian Journal of Cultural Studies
Vol. 1 No. 1, May 1983

Discourse, history, fiction: language and Aboriginal history

Stephen Muecke

First presented at the Anthropological Society of Australia's annual conference. Canberra, August, 1981.

The main problem for Aboriginal History, as I see it, is to authenticate the appropriate discourse for its transmission. At the moment the 'authentic' accounts of Aboriginal history are firmly locked in academic standard English. Recently an Aboriginal working party for the Bicentennial History Project has challenged the assumptions of historians that history and the language in which history is presented are somehow independent of each other. [note 1] Prior to European intervention in this country, and well after that point, Aboriginal peoples have been recounting their histories in traditional languages, and, of late, in creoles and Aboriginal English.

Events in history exist only insofar as they exist in discourse. The discursive representation of events carries with it, furthermore, guidelines in the form of grammatical structures and tropes for the interpretation or reading of the events. Texts will be written or spoken not only when something worth telling about happens, but in response to broader institutional or cultural demands for the event to be recorded, not necessarily for posterity, but also for the achievement of current socio-political ends.

The 'rise' of Aboriginal History, as a discipline, seems quite naturally to coincide with the practice of talking about it and writing about it in academic English. But people are still talking about 'history' in the narratives of Mangala, Pintupi, Ngaanyatjarra, Wongai, Wik-Mukan and so on. If an event, once represented in one of these languages, finds its way into academic English, then it is no longer quite the same event. The status of the event, which includes its place in the chronology of events, and its 'importance' in relation to the narrative which history is already creating so that it will be found and inserted in the place made ready for it, lies in its discursive formation: the way that talk structures, values and authenticates the existence of the event. As Foucault says:

Discursive practices are characterized by the delimitation of a field of objects. the definition of a legitimate perspective for the agent of knowledge, and the fixing of norms for the elaboration of concepts and theories. Thus, each discursive practice implies a play of prescriptions that designate its exclusions and choices. (1977; p 199.)

Since no text can innocently or neutrally portray events, there is similarly no neutrality available for reading texts which do portray events. The reading I want to propose here, discourse analysis, aims at a neutrality in regard to history by looking at the way history or non-history is formed in language itself. It seeks

page 72

not to re-analyse the knowledge of history, but to analyse the specificity of historical ways of talking, to attempt to see how the discourse of history creates the object history. This is how Foucault sees the role of linguistic analysis today:

it is ... not a theoretical reworking of knowledge acquired elsewhere, the in terpretation of an already accomplished reading of phenomena, it does not offer a 'linguistic version' of the facts observed in the human sciences, it is rather the principle of a primary decipherment: to a gaze forearmed by linguistics, things attain to existence only in so far as they are able to form the elements of a signifying system. Linguistic analysis is more a perception than an explanation: that is, it is constitutive of its very object. (1970; pp 381-382.)

To this general approach to language must be added, if I am to proceed with my analysis, the notion of 'subject position'. Not only are texts spoken (or written) from a certain position, which can be socially or politically determined, but they speak to specific positions of the listener or reader. Different types of texts will specify different positions for the subjects (speakers, listeners) involved such that not only will language incarnate meaning in a system of signs, but also in the form of a series of positions it offers for the subject from which to grasp itself and its relations with the real. (Nowell-Smith, 1976, p 26.) The positions thus set up are by no means as numerous as the number of individuals in society. Language makes significant distinctions between speaker positions; the French linguist Emile Benveniste made the first distinction between what he called the sujet de l'énonciation (the speaking subject) and the sujet de l'énoncé (the subject of the utterance). In the first case the position of the speaking subject gives a statement its import (as in "I promise you"), while in the second it seems unimportant who is uttering the statement (as in "The king died in 1897.") This distinction forms one aspect of Benveniste's further distinction between two fundamental types of discourse - history and discourse.

It will also help us locate some of the problems associated with the writing of history in the cross-culture situation in which Aboriginal History finds itself. Here history often tends to be produced in transpositions from oral Aboriginal narratives to written academic English.


French presents its speakers with two forms of the past tense, one of which the aorist (passé simple) is used only in written texts; it has disappeared from use in spoken language. Benveniste explains this shift in terms of a differentiation of status of certain written texts in opposition to spoken language. These written texts become histoire or history, which carries connotations of truth, while spoken texts which use other forms of the past tense, as well as the present are discours or 'talk'. 'History' is related to events which are talked about without the intervention of the speaker of the narration. In 'talk', on the other hand, the intervention of the speaker is a characteristic.

This distinction carries through to the pronoun system of French (and this is also present in English) where 'history' is characterized by the use of the third

page 73

person (he, she, it) and 'talk' by the alternation of the first and second person (I/you). Colin MacCabe summarizes:

Language is the combination of two autonomous but intersecting systems: the world of the first and second person which define the subjective realm of discours, with a tense system related to the moment of speech and the world of the third person which defines an objective realm of histoire with a tense system related to the moment of the event. (1979; p 282.)

The grammatical differences between the two discourses thus serve to locate the subject in different positions in relation to the text. The reader of 'history', for instance, is not directly addressed by the writer, and is temporally removed by the tense from the sphere of action. The overall effect is for the text to become authoritative and the reader unquestioning. In the next section I shall describe other devices contributing to this effect.

'History' thus achieves a truth effect because it seems embedded in time and removed from the variable perceptions of the subjects of 'talk'. Fiction combines these two discourses. In classic fiction the talk of the protagonists (and sometimes the author-reader dialogue) is articulated in relation to the third-person past-tense narration of the plot, which constitutes histoire. The alternation between the two discourses constitutes the effect of fiction. To make an obvious example, the fictionalization of historical accounts involves at least the inclusion of dialogue, and the fictionalization of spoken discourse (e.g. tape-recordings) would involve the addition of narration.

These three types of functional discourse (history, fiction and 'talk') are by no means universal, at least in the uses to which they are put. Aboriginal societies, for instance, do not recognize a category 'fiction'. There are stories and talk among the spoken genres. Stories can be true or 'of the dreaming'. Stories which are not true are 'liar', but do not take narrative form. [note 2] It would seem then, that all Aboriginal oral narrative is true in their sense of the word if it does not fall into the 'dreaming' category. Such stories also inevitably contain dialogue ('talk'). European Australians listening to such stories find difficulty in interpreting them. Their training does not tell them whether to interpret them as history, 'just talk' or fiction. If the text is spoken and full of dialogue, then surely it is 'just talk'. But since the talk is organized around a plot, it must be fiction. How then can the Aboriginal narrator insist that the story is true? I shall try to answer some of these questions in the analysis to follow.


Sandawara or 'Pigeon', an Aboriginal resistance fighter from the Kimberleys, was killed in 1897. He was killed again, in a cultural sense, by Ion Idriess in his book Outlaws of the Leopolds in 1952. Idriess, no doubt unwittingly, effectively wrote Sandawara out of any significant place in Australian history through the cultural bias that he and his contemporaries brought to bear on the events of the time. A lot of work remains for researchers in Aboriginal history. They have to reassess many accounts of Aborigines written at a time when Aborigines were not considered a part of history as active shapers of history.

page 74

The analysis I am about to make is intended as a critical comparison of black and white accounts of an episode in Aboriginal history. It was prompted by the appearance of Colin Johnson's Long Live Sandawara in which Sandawara achieves slightly dubious immortality as a fictional hero. An examination of the language of these texts (Johnson's, an oral Aboriginal version and a written white version) will, I hope, reveal something of the way in which different languages (English, Aboriginal English) and different texts (novel, oral narrative, historical account) predispose writers and readers to certain interpretations.

In addition to this, the circumstances of the production of the texts, for instance, if they are published or 'just talk', will invest them with a power, or lack of power, quite unrelated to what the text is trying to say.

It is commonplace to say that historical accounts are selective. But not only are things left out or selected according to dominant ways of thinking; language generally leaves us with little choice about how to construct accounts. Genres are available to writers or speakers, and are chosen according to the purposes of the writer. So if one wants to sound authoritative on a subject one generally selects this style:

"... Pigeon was sighted only about 100 yards from the Cave entrance. Escaping by climbing the cliffs, Pigeon dropped the cartridges he was carrying. That night, wounded and weakened, with only three cartridges left, Pigeon again tried to escape but was overtaken and shot by Constables Buckland and Anderson. The day was 1st April, 1897. The news reached Derby on 6th April, 1897. According to Ion Idriess, the author of "Outlaws of the Leopolds", Pigeon's body was "put in a tree" by tribesmen and eventually laid in the Cave of the Bats. This was in accord with their immemorial burial customs." (ms. Public Library, Derby, c. 1975.)

This is part of a document which was available for tourists at Derby public library. The story it tells is one well-known to all Kimberley people, but, as we shall see, this story is told in very different ways by whites and Aborigines. A history, alternative to the one above, was given to me by Sam Umbagai of Mowanjum Community in 1977:

"then some, one bloke said "Ay" -
'is brother then you know? -
said er, "Better shoot 'is foot" -
and they shot 'im right in the foot, you know?
an' 'e fell down -
well 'e said "You blokes got me -
you're my own brother" -
an' they shot 'im again -
coupla shots, finish -
'e was finish then -
an' they cut 'is throat off -
picture 'im you know? -
'is body 'is clothes shirt an' forty-four—

page 75

boots, took 'em away to Perth -
'an ol' N--- was there, this ol' Joe -
- well, dat's al far as I know -
well, 'e's in museum in Perth now -
seen 'im?
(SM: No, I didn't know that)"

The differences between these two stories are extraordinary in view of the fact that they deal with the same event. In the first, Pigeon's confinement is assum ed - he tries to "escape". But all this time he is actually in his own tribal homelands. Furthermore, the police are given credit for his final defeat and his hopelessness is exaggerated; he continues to try to "escape" - like a wound ed animal. [note 3]

In the Aboriginal story, it is Pigeon's cousin (sometimes called 'cousin brother' - hence "you're my own brother") who defeats him by magical means. In the variations of the Pigeon story which can be found in the Kimberleys, Pigeon's weak spot is always his hand or foot, the reason being that he made himself immortal by hiding his 'soul' in a pool of water. After that bullets could pass through him without effect. The tracker who finally kills him could 'see' the soul palpitating in the water. [note 4] Having removed it and thrown it on the ground, he made Pigeon once more vulnerable to the bullets, especially if they hit his hand or his foot. The 'white' correlate of the immortality stories is connected with the trick that Pigeon used at least once. After having been wounded and having left traces of blood, he would get someone to announce that he was dead, thereby calling off the hunt.

In the Aboriginal story Pigeon dies like a hero with a 'last word': "you're my own brother". His gear is taken away so that it can be exhibited on an effigy in the museum in Perth. In contrast, the 'white' story chooses to emphasize the Aboriginal burial custom, both stories choosing detail which locate the listening or reading subjects as Aboriginal or non-Aboriginal. Both ceremonies might well have occurred, but these two versions of the one event illustrate a rule for the writing/telling of history in cross-culture encounters. This rule would be something like: select details which are outside of, or extend, the general knowledge of your projected reader or listener. This rule thus depends on cultural difference.

Both extracts name the authority on which they have the story. The 'white' version names Idriess who wrote about the events after coming through the region some years after they occurred. His account is based largely on police records. [note 5]

The 'black' version of the story names a man who was there, Joe N---, who told the story to Sam Umbagai. To the white listener the 'Achilles heel' motif (which is also a motif of traditional mythology) is implausible, such a listener is much more likely to put his or her trust in the 'official' printed word with its precise specification of distances, dates, numbers and names. Furthermore, the ideology behind the European version is committed to representing Pigeon as a defeated man even before his death. This anti-Aboriginal ideology, which has its roots in colonialist economics and racism, was extremely strong at the time the 'official'

page 76

records were made. For instance, an extreme view was taken by the editor of the North West Times on the 2nd October, 1894:

It would be a good time for the Western Australian government to shut its eyes for say three months and let the settlers up here have a little time to teach the nigger the difference between thine and mine ... it would only have to be done once and once done could easily be forgotten about. (Pederson, 1980; p 66.)

Later the police were given discretionary powers to deal with the problem of Aboriginal resistance to pioneering settlement, which largely took the form of organized gangs making guerilla-style raids on homesteads, ambushing parties of settlers, and systematically killing livestock. Howard Pederson, in his history of Pigeon and his gang, stresses the organized nature of the Aboriginal rebellion and the very real threat it posed for pastoral expansion in the region. The removal of this threat required large-scale efforts on the part of the police force - many Aborigines were shot. The first ('white') version of the death of Sandawara should be read without taking into account discourses on Aborigines as they have been historically formed.

The oral Aboriginal version of the story places its emphasis in locating the events in conjunction with the human relationships around which they occurred: the narrator knew Joe N---, who was there, and Pigeon is defeated by a man who was a close kinsman. Furthermore dialogue ('talk' or discourse) is used as a convention to represent the scene as relevant to the present time, as if the dialogue of the participants were still going on. A listener of Aboriginal stories is positioned in at least two ways. Firstly as a new person following a retelling of a 'true' story where the narrator retraces once more the words of the original participants, and secondly as a participant who must respond actively during the telling of the story (as I did in the last line of the first extract). The listener is thus linked, personally, via previous narrators like Joe N--- to the actual event. The reader of the 'white' history does not have this personal privilege, and is called upon to visualize (rather than listen): the reader's gaze is directed ("... Pigeon was sighted ..."), then the 'scene is set' for the main event ("Escaping by climbing the cliffs, Pigeon dropped the cartridges he was carrying."). The 'white' history relies on the word. Generally speaking, Aboriginal narratives do not construct a 'scene' with descriptive passages, but go straight to the dialogue. They are also often introduced with formulae like: "I'll give you the word on that one".

Colin Johnson's version of the death of Sandawara also relies very much on imagery, and in this, as well as in other respects, it is literary:

Minutes pass and then far ahead they see the hobbling Sandawara marching towards the sun. His long shadow reaches back to mock them, to menace them. They slow, but creep closer and closer to the man who is not even aware that he is being followed. The trackers shudder as the warrior stops and turns towards them. Upright and proud, he stands there. A smile lights his face, then he sinks to the earth. The pack seeks cover and opens fire. Bullets lash toward the man, but there is no sound or movement from him. At last the trackers gingerly approach the fallen figure and circle it. They edge in and

page 77

stand looking down. The white men are far off. The black men stare at their fallen brother and watch as he stirs and gets into a sitting position. 'Brothers, the white man can never take what I have', he gently murmurs, then falls back into freedom." (Long Live Sandawara, 1979: p.166)

In this passage Pigeon the historical figure disappears, to be replaced by a romantic hero. The position of the reader shifts once more. We (as readers) need no longer be white or Aboriginal, the distinction which affected the possible readings of the two previous texts. Now we need only relax in an understanding of the novelistic conventions such as syntactic inversion ("Upright and proud, he stands there"), metaphor ("A smile lights his face ...", etc.) and the imagery, which constructs a 'view' of the scene more from the "far off" position of the "white men" than from any Aboriginal vantage point. [note 6]

Furthermore, the reader is not in a position to know, from this text, if Pigeon existed at all. This would not be important if it were not for the fact that Johnson seems committed to representing Pigeon/Sandawara as a revolutionary (notions of black brother-hood are in the text). What has happened in the transformation of the story into fictional discourse is that the conventional devices of Johnson's literary style undermine what would appear to be his attempts to revive the Sandawara story as a story of black struggle in the early days of colonization.


Aboriginal writers of literature, and writers of Aboriginal history, are faced with a number of problems when they draw on accounts of events involving blacks and whites in Australia. There has been, and still is, a kind of pressure exerted by racism in Australia which will shift the significance of texts based on these accounts towards aesthetic fictionality and away from contentious politicality.

This pressure has different consequences for the two categories I isolated above, 'fiction' and 'history'. Taking fiction first, it is the existence of the Ned Kelly and Ben Hall bushranger stories as a minor genre which would permit the 'Pigeon' events to fall into the same category (a book like Idriess' Outlaws of the Leopolds). This writing/publishing strategy effectively locates the work outside of any field of interpretation which would include race relations as a conceptual category. The same strategy has oriented the production of many Aboriginal myths as children's stories, a non-serious genre.

It also allows for the construction of the 'hero' as a textual configuration; once a character is recognised as a hero, then certain narrative possibilities will eventually follow from this, such as the exemplary moral behaviour of the individual, or, on the contrary, the radical flaunting of convention. The hero construction thus allows any contradictions which may be set up in the course of the story to be resolved at the level of individual motivations, rather than, say, at the level of the depiction of social conditions.

Allegory or metaphor will tend to dominate the rhetoric, and this is true for Long Live Sandawara where the Pigeon episodes of the story represent the sovereignty of a revolutionary past standing as a projection and justification for

page 78

the contemporary actions of a group of young urban guerillas. The figure of metaphor thus signals the 'literary' text, as opposed to the 'historical' text; and it can be deployed as a broad structural feature as well as a linguistic feature at the level of the sentence.

Secondly, in the academic writing of 'history', the tendency is to produce an untiring positivity, a series of assertions as to 'what actually happened'. Once history discovers a new positivity, for instance that there really was Aboriginal resistance to invasion after all, then this assertion tends to be repeated without any tendency for the discourse to question its historical formation, that is, to perceive the set of relations which first produced an absence of Aboriginal resistance, then suspected its presence, then boldly asserted its importance for historical 'truth'. While this is not true for all writers of Aboriginal history, my present aim is not to castigate sorne of them, or even the discipline itself, but to try to point out some of the general features of the discourse, the main one here being that in the discourse of history the historian as speaking subject need not take responsibility for what is said. (Barthes, 1982; p 18.)

It could be said that the most significant Aboriginal material culture is now produced as texts (rather than paintings, boomerangs, etc.), and that the producers of these texts, if they are Aborigines whose desire to write was born of political struggle, are on the horns of a dilemma. Either they can write 'serious' history and grapple with the difficulty of inscribing politics of race relations into the text, or they can adopt fictional conventions and find that the aesthetic becomes the dominant critical category for the reception of their work.

To transcribe the oral tradition might seem a solution, but I think one would have to be careful not to do this in the spirit of a longing for a primordial and innocent speech-before-writing. The kind of textual production I have been discussing here is totally within the field of writing; speaking presents a different series of problems in relation to different institutions (e.g. radio). Transcribed speech is a kind of writing, a conventional representation of spoken language which elides the non-verbal gestures and the intonations which modulate it. To use the oral tradition in the construction of a different kind of writing would be another rhetorical device, but it might be one which functioned to renovate literary and historical orthodoxies. With the emphasis, in the oral tradition, on the process of telling, the visual or imaginary might be displaced as one dominant fictional category (and the 'absent authority' as an historical speaker position) in favour of the textual strategy of the implication of writer and reader positions in a process of specific deployment of discourses significant to the present.

Stephen Muecke teaches at the South Australian College of Advanced Education.


1. "When the cues, the repetitions, the language, the distinctively Aboriginal evocations

page 78

of our experience are removed from the recitals of our people, the truth is lost to us." "A Black view of history, culture" The Age, 13/2/81.

2. 'Liar story' was explained to me as, for instance, telling someone that there is a snake nearby: "Oh! You only bin tell me liar story!"

3. See also the following account in Marshall and Drysdale, whose interests in zoology and 'graphic' description lead them to classify the 'snarling' Pigeon along with the zebra finches; he is a native animal:

"His end came at Tunnel Creek, today as full of food - flying foxes, crocodile eggs, night herons and countless long-tailed and zebra finches - as it was in the days of Pigeon. Two patrols led by black trackers converged on the outlaw. He delayed his getaway into the tunnel (with its three outlets) in the hope of getting a close shot at yet another white man. But the patrol shot straighter than he; and in his wounded frenzy Pigeon dropped a parcel of seventeen cartridges, tied up in bark from the paper-bark tree. That left him only three cartridges; and Pigeon died, snarling defiance at the agile black trackers whose bullets finished him off.

Until a few years ago, Pigeon's lubra was still alive." (p.134).

4. A maban ('doctor') from Roebourne called Minko Mick, according to Banjo Worrumarra, personal communication, 1979.

5. Howard Pederson, personal communication.

6. As readers we 'relax' in the sense that our training in literature has made us familiar with such devices, present equally in Biggles books or Jane Austen. Traditional Aborigines are not potential readers of Johnson's book because we must assume that they don't have this training.


Barthes, Roland: "Le discours de l'histoire", reprinted in Poetique, 49, 1982.

Benveniste, Emile: Problemes de Linguistique Generale, 2, Paris: Gallimard, 1974.

Foucault, Michel: The Order of Things - an Archaeology of the Human Sciences, London: Tavistock, 1970.

Foucault, Michel: Language, Counter-Memory, Practice (trans. and ed. by D. Bouchard) Oxford: Blackford, 1977.

Idriess, Ion: Outlaws of the Leopolds, Sydney: Angus and Robertson, 1952.

Johnson, Colin, Long Live Sandawara, Melbourne: Quartet, 1979.

Love, J.R.B. "The Tale of the Winking Owl", Mankind 3/9, 1846.

Nowell-Smith, Geoffrey: "A Note on History/Discourse" Edinburgh '76 Magazine, No. 1, 1976.

MacCabe, Colin: "On Discourse": Economy and Society, 8/4. 1979.

Marshall, Jock and Drysdale, Russell: Journey among Men: Melbourne, Sun Books, 1966.

Muecke, Stephen: "Australian Aboriginal Narratives in English - A Study in Discourse Analysis" Unpublished Ph.D. thesis, University of W.A., 1981.

Pederson, Howard: "Pigeon: An Aboriginal Rebel. A Study of Aboriginal-European conflict in the West Kimberley, North Western Australia during the 1890s." Unpublished BA Honours thesis, Murdoch University, 1980.

New: 20 May, 1997 | Now: 22 April, 2015