iDIDJ Australia Didgeridoo Cultural Hub

Shadows of song: Exploring research and performance strategies in Yolngu women's crying-songs.

Article from:Oceania
Article date:December 1, 2001
Author: Magowan, Fiona

**********

Heart-wrenched wailing suddenly pierced the night air, as a commotion of distressed voices rumbled closer in the darkness. From the house opposite, the distinctive high-pitched cry of a woman was soon joined by several more, the wailing intensifying in descending cascades of sorrowful melodies, weaving a garment of sadness that seemed to cloak the town of Galiwin'ku (Eleho Island). (1) The wailing continued, polyphonically wrapping itself around the mourners and those who stayed in their houses, waiting to hear of the death, sung by men to the community through the Ancestral Law. In waterfalls of song, poetic laments of loss, longing and ancestral belonging echoed across the street, invisible cords of kin-song weaving a web that tied the deceased to the mourners and their ancestral identities. Each melodic wave of grief served to bind singers to listeners and the family of the deceased, enveloping them in a multitude of sentiments and memories. As the dawn met the last rays of the morning star, a lone woman's voice continued the lament, her thin chords straining the tenuous links remaining between life and death. Hoarse, partly sobbing and partly singing, her lament became less and less frequent until silence reclaimed the street, awaiting the funerary proceedings to follow.

These entangled webs of song pose particular problems of meaning-making for researchers of music culture, an experience that has been identified as 'chasing shadows' (Cooley 1997:3). These shadows are elusive and as evanescent as the multiple, ambiguous and open-ended conundrums that this performance analysis addresses. I argue that Yolngu women's crying-songs, ngathi-manikay, create a shadow-dance of making meaning between performers and listeners, variously twisting and turning around a performative nexus of music making, recording and the writing of music. In this examination of one senior woman's songs a series of shadow-dances between self-reflection and strategic positionings were played out between ritual performances and research recordings. For the purposes of the analysis that follows, it must be recognised that each song is only one element of the much larger and longer song series from which it comes and from the complexity of other series that interact with it. (2)

In one regard, shadow-dancing suggests an ever-mobile revelation of meaning between performers and listeners where images grow longer and fuller as they are sung, like a shadow stretching out from a person as they move in relation to the sun's rays. Just as shadows change shape depending on the position of the person and their movements to the light, so, too, do the songs change until one fades and the next begins. In another regard, shadow-dancing is also a technique within song composition. Just as shadows require light in order to create them, so singing is the medium of ancestral brightness, evoking the spiritual power of ancestral beings. As one ancestral being comes into view so another may be partially alluded to, its shadow hiding in the background, waiting to be revealed in the next part of the singing. As such, ancestral beings are always in the land and sea covered and concealed, their forms but a shadow until they are sung into vision once more.

Discursive analyses in the West have all too frequently produced hegemonic narratives about cultural others. However, it is nowhere more pertinent than in a performance context that performers should be recognised as not just having begun to write themselves back into representations of their performances but that they have always done so. The 'ground of musical knowing' has increasingly been recognised as an intersubjective ground of being and being known where performers are integral authors of their performance experiences. (3) Writing the discursive and non-discursive practices of singers back into their songs is an embodied process of cultural performativity, not just a tool of 'knowing about' music research. This intercultural and intersubjective nexus of creativity is a highly complex one as Rabinow (1977) and Dumont (1978) recognised early on, where the fleidworker's personal relationships constitute the primary data of anthropological research and their self-reflections form the epistemological base for analytic development.

This paper explores the extent to which the effectiveness of musical anthropology (Seeger 1987) depends on how various sorts of reflexivity might be recognised and managed in relation to the performances the researcher seeks to evoke, analyse and recount. It poses questions for the discipline about how the experience of performance might change the performance of anthropology. It also seeks to show how strategies of knowing and being known can be conveyed in their immediacy, intimacy and particularity without heading over the cliffs of either romanticisation or irony. (4) Crying-song knowledge cannot be entirely separated from the learning experiences of the anthropologist or ethnomusicologist, nor from the processes of recording and interpreting because 'personal experience is neither free nor individual; it is constrained by interaction with the tradition' (Rice 1994:308). Therefore, rather than viewing song as exuding an authority whose dynamics of production remains largely unquestioned, this shadow-danc e is partly a play of 'how power enters into the process of "cultural translation", seen both as a discursive and non-discursive practice' (Asad 1986:163), and about 'scanning sound energy for patterns' (Clayton n.d.:4) in order that a tuning-in to the world of the performer might allow a series of reflections to flicker, grow, dance and extend into and from the singer to the researcher and back again.

In order to illustrate fluidities of consciousness within the research context, I borrow Said's (1978: 20) notions of 'strategic formation' and 'strategic location' in an analysis of a selection of one woman's ritual songs. (5) In this case, 'strategic formation' could be said to be a means of analysing the relationship between individual performances of songs and their transmission to the researcher where performances acquire a particular mass, weight and referential power among themselves in the research context, as well as through multiple performances in public gatherings. The performer comes to be situated in a strategic location between other performers and the researcher, each positioned in relation to the production of particular song meanings and their effects. I examine these performance strategies as ways of knowing and coming-to-know, manifest in a small sample of ngathi-manikay to show how my understanding of the learning context changed over time. For as Sangren notes:

'Meaning' and 'culture' are not merely the negotiations 'between' subjects in acts of 'communication'; such acts of communication are inevitably embedded in encompassing systems of power and meaning. These encompassing systems are related dialectically in the process of social and cultural reproduction to the 'experiences' of the subjects that they encompass and that are necessary in their reproduction. (Sangren 1988:417)

Different aspects of the performer-researcher relationship were revealed and concealed over seven years of performances and recordings. By reflecting on my engagement in this fieldwork context, it will become apparent that the singer deconstructed and reconstructed song texts as shifting narratives of herself, simultaneously locating me as strategically as possible in her own world. In recognising these strategies, the importance of a reflexive approach to the latent, and often taken-for-granted dynamics of the research process is highlighted, as it is only 'once 'informants' have begun to be considered as co-authors, and the ethnographer as scribe and archivist as well as interpreting observer, we can ask new, critical questions of all ethnographies'. (Clifford and Marcus 1986:17)

The critical questions that will be posed here are: how is the complex experience of self and other danced out in a web of mutuality, reciprocity and obligation in the process of music making and, in particular, of learning crying-songs; what dynamics of respect, power, mutuality, love and obligation are generated by discursive and non-discursive practices between the performer and researcher; and how have these song strategies of location and formation shaped our relationship over time? These questions are situated firstly in a historical trajectory of performance research in Arnhem Land.

ARNHEM LAND PERFORMANCE RESEARCH

The only detailed analysis of ngathi-manikay, from northeast Arnhem Land in the Northern Territory of Australia is Catherine Berndt's 1950 Oceania article, 'Expressions of grief among Aboriginal women' which compared women's song texts from Melville and Bathurst Islands with those from northeast Arnhem Land. (6) This article discusses eight texts of four Yirritja and four Dhuwa songs; two by the same woman while the others are from women of different groups. As no tape recordings exist of these songs, (7) it is probable that Berndt transcribed the texts phonetically discussing their semantics with women afterwards. (8) Apart from Berndt's analysis, ngathi-manikay have only occasionally received general mention in broader discussions of ritual genres. (9) This analytic lacuna arises from a combination of academic trajectories and cultural restrictions in the research context: firstly, apart from the ethnomusciological writings of Moyle (1974) and Stubington (1978), there were few female anthropologists in nort h east Arnhem Land exploring the contexts of ritual, art and performance at that time. Since the 1990s, however, there has been an increase in research on dance, ritual art and song by female researchers (see Tamisari 1998; Slotte 1997; Hutcherson 1998; and Bonfield 1998 located primarily at Milingimbi, Ramingining, Yirrkala and Gapuwiyak, respectively); (10) secondly, although male anthropologists in this region have not always treated 'relationships which flow through women as secondary and secular' (Bell 1983:240), (11) their discussions have tended to focus less upon women's s views, due to the cultural limitations of gender involvement in ritual; and thirdly, as women's crying-songs constitute a period of intense personal grief, reflection and soul-searching that conjure feelings of loss and sorrow when research discussions focus upon them, women are often reluctant to sing other than at times of deep grieving. (12)

Crying-songs are solely the domain of senior knowledgeable and respected women but northeast Arnhem Land is not the only region of Australia where women's crying-songs can be heard. (13) It is common knowledge that, prior to European contact, women would wail, sing and sometimes hurt themselves (14) to show sorrow and deep love for their kin when a death or other ritual occurred. (15) In northeast Arnhem Land, both men's songs, manikay (16) and women's crying-songs relate the travels and actions of the ancestors embodied in and emergent from the Arnhem Land topography. Women's ngathi-manikay use the ritual words of men's songs but the texts are performed in quite a different fashion. In a broad sense, anyone may cry but ngathi-manikay is a specific form of emotional release from anger or grief or loss.

Unaccompanied by clapsticks or didjeridu, women cry in long slow lines, revealing the travels of ancestral beings as a series of related ancestral events. Men perform the same ancestral subjects but in three to four repeated segments with breaks between each subject. They also sing a sequence of subjects comprising an ancestral journey of the land and the sea before the song series is finished, whereas women may focus their singing upon just one major (yindi) song subject. While these events speak of the creative actions of ancestral beings, they are also a commentary on the physical and spiritual nature of the singer's ties to places, to other individuals and to key groups. Each ancestral being holds the essence of individual and group identities inside their own country and waters whilst connecting groups with different ancestral beings at other places. Places are spiritual containers of human life and of the possibilities for healing, illness, conception, birth and increasing spiritual strength through inv oking the power of the ancestors in song. For women it is 'an individual matter about what to cry' (Berndt 1950:308) although there are certain occasions when it is obligatory to do so, for example, immediately after death, at a funeral, during or after a circumcision, or in preparation for a Ngarra or Gunapipi ritual. A woman may also cry when she remembers loved ones far away, or particular attributes of the environment such as the morning star. (17) For women, as for men, music is power -- a social, spiritual and political force. Thus performance is a means of making powerful statements by authorizing, teaching and controlling the place of the researcher within the community through song.

SEEING MUSICALLY

From 1990-1992, I sought to learn and record women's crying-songs on Galiwin'ku, working with a range of senior women from different groups. This engagement was marked by a shadow-dance of increasing trust as it took several months before senior women were willing to share these songs of sorrow. The senior Datiwuy leader had adopted me on my arrival and insisted that he must teach me men's songs first as the 'inside' knowledge of ritual songs is controlled by men. (18) Once he was confident that I could dance (19) (and I had recorded and transcribed many songs of his own group, his mother's group, his mother's mother's group and his sister's son's group) he allowed me to approach women. There were problems in this shift. Moving from the status of 'honorary male' to female was a difficult transition as women were unsure how much 'inside' knowledge I had been given. Eventually, those who kindly agreed to teach me their songs were often unable to talk in depth afterwards for the pain of the memories they caused. It was also a slow and difficult process to find women who 'felt like singing' but the word gradually spread around Galiwin'ku that I was keen to learn ngathi-manikay.

Murukun Dhamarrandji was acclaimed as one of the most knowledgeable and adept performers of crying-songs in the region and it was early February 1991 when this gentle but vivacious woman knocked at my door. She said, 'I will teach you milkarri (crying). Yolngu know my voice. It is clean and hard, dal. It has a good sound, manymak rirrakay. Her stress on the quality of her voice was qualification of her right to know and to pass on her knowledge. The deeper a woman's voice, the greater the respect, as the timbre of the voice is consonant with the depth of song knowledge. Young women sing in higher pitches, garramat, while older women and men sing in a lower pitch, ngoy. Pitches are identified along a continuum of relativity to the singer's voice and to others known to them. Younger women would say how ashamed, gora, they felt to sing in public as their voices had not deepened, (20) an indication of the 'outside' and lesser status of their knowledge and the right to sing. Murukun and I met daily and we would si ng, talk and record until sunset. Our days were more than song and more than musical knowledge, they were a sharing of one another in a delight of knowing and coming-to-know. Her melodies of mourning constituted a gift of musical embodiment; a generous exchange of musical knowing for my coming-to-know, that I might be enfolded in her web of musical being-in-the-world.

Of course, many town events intervened: visits of dignitaries; graduation ceremonies for Batchelor College graduates; the regular welfare pay day rush and shopping sprees; trips to Yirrkala and other homelands; and sadly, all too frequent funerals. The latter proved constant throughout 1991 with 11 funerals during the year. On some occasions Manydjarri would often request his songs to be recorded, on others, I would ask his permission to record songs of other groups, and at times, Murukun would take me to sit with the women mourning near the funeral shade to listen to their songs.

On my first trip to Arnhem Land, I was a young, unmarried, childless woman and therefore not in a position to sing crying-songs either publicly or privately without deep embarrassment. At the start, Murukun would sing as I recorded. She worked through the crying-songs of her own Djambarrpuyngu inland and seaways song series; her three groups of mother's song series, Lamamirri, Warramirri and Wangurri; her mother's mother's song series, Golumala; and those of her sister's group, Galpu. Her decision about what to sing was always one that related to recent events and memories. If her daughter had just called from Darwin she would sing either her mother's songs or her sister's son's songs; when she would recount her patrilineal connections she would sing her father's and father's brothers' songs. She always began by saying, 'Listen to my voice', whilst I tried to memorize the timbre, pitch, melodic line and juxtaposition of her words. My initial efforts to sing were often met with a patient silence and a kind smi le as she would begin to sing again... She would repeat the song tirelessly without identifying what had been wrong in my rendition. On each of her repeats there would be variation in the words or in the length of the melodic line making it difficult to perfect.

In our initial days together, Murukun would not explain songs word for word. She would give a gloss and sing them again, despite my attempts to quiz her tediously on each individual phrase. I had hoped Murukun might be concerned with my questions of accuracy: had I understood the meaning of individual words; had I created the correct images from each phrase; had I captured the essence of meaning in the long lists of names so common in crying-song structure? Instead of responding analytically, Murukun would say 'listen again', as if it would become self-evident to me. It appeared as Ewing (1990:264) has noted, 'as if the utterance and the self-representation it expressed or implied were her total experience of self' and that I could automatically share this. Due to the tendency to sing alternate words on each repetition, I had taken to playing the recording back, listening to the song and then singing it with her together with the tape. After a couple of attempts Murukun cried elatedly, 'Yo, manymak dhuwali", 'Yes, that's right'. On this occasion, the only change I had made was to the placing of the breath mark, a controlling factor in making meaning in crying-song performance. Gradually, I ceased wanting to ask, instead, I wanted to hear again and in doing so, I came to feel how she wanted me to see musically, the colourful movement and detail of the bird or animal that was flying in sound.

STRATEGIES OF FORMATION AND LOCATION

In the initial stages of song instruction, she adopted one main strategy of formation: condensed repetition of song couplets. When I had difficulty following the text she chose which segments and sections of music should be repeated for the research context, representing them in a condensed form in comparison with their elaboration in funeral contexts. This shift in the processes of contextualisation was reflected in the songs themselves, and by focusing more on these processes, I came to recognise how Murukun used particular forms of 'poetic patterning' to interpret the structure and significance of her song narrative (Bauman and Briggs 1990:70). In the learning process, when I failed to grasp the logic behind the phrasing of the melody, she would adopt a pattern of condensed repetition in every couplet, singing each line the same way twice. However, whether I had managed to correct a line or not, each repeated couplet would only be sung once before moving to the next and if I had still failed to grasp the s ong she would perform all of it again. In this example, of the bar-shouldered dove, the bird cries out 'gukguk' as it sees the shadow of a shark moving slowly in the shallow waters, warning Yolngu that danger is near:


Ngapalawal -- The Bar-Shouldered Dove

Ngapalawal gukguktja ngathi
Ngapalawal gukguktja ngathi
The bar-shouldered dove is crying

Dhatharram ngathi gukguk
Dhatharram ngathi gukguk
The bar-shouldered dove called Dhatharram and Gukguk is crying

Yimingida gukguktja
Yimingida gukguk
The dove, called Yimingida, Gukguk

Gapu nhangal djarrarranmirrnydja
Gapu nhangal buwananamirrnydja
He sees the muddy water called Djarrarran and Buwanana

Ngathi dhatharram gukguk
Ngathi dhatharram gukguk
The dove, Dhatharram and Gukguk cries
In 1991, Murukun's first lessons began with song instruction from the perspective of her own subgroup whose patrilineal identity comes from the King Brown snake, Darrpa, at Ngurruyurrtjurr. This subgroup is one of five other subgroups that join to make the one Djambarrpuyngu group. All the subgroups speak the same dialect and share the same main ancestral being, shark, as well as having rights in other ancestral beings peculiar to their own homelands. (21) Although shark is the group's primary ancestral being, Murukun chose to teach me her own subgroup's identity first, making only passing reference to the shark. As my mari, grandmother, Murukun was in the kinship position of key ritual authority to hold and pass on knowledge to a granddaughter and she sought to establish a sense of order and authority, teaching respect for her ancestral King Brown snake identity. In addition to the technique of condensed repetition, Murukun cried short extracts to illustrate how crying-songs are constructed in thematic segme nts. In this excerpt of the text, the King Brown snake lies sleeping in the shade, the shade that is also hers, the shade of the Ngurruyurrtjurr Djambarrpuyngu clan. Just as the snake sleeps, so do the people in her shade where the snake and people are one:


Warraw' (1991): King Brown Snake Shade (Ngurruyurrtjurr Djambarrpuyngu)

Ngarraku dhuwali Walarritj Lumbilumbi
This is my snake shade called Walarritj Lumbilumbi

Ngalimurr dhu wanganhami dhuwal
We all talk to each other here

Wulanayngu Wurrpurrnydja Riyariya Manganiny
We are the King Brown snake people called Wulanayngu Wurrpurr Riyariya
Manganiny (the deep ritual names of the subgroup that are the essence
of their identity in the land)
Throughout the learning process, we were strategizing each other from our own self-referential viewpoints; Murukun wanting to establish a sense of authority through webs of kinsong, while I wanted to understand the elements of crying-song structure. Her strategy of location was one of 'discursive formation' (Foucault 1972:49) (22) whereby she situated my kinship to her as part of our mutual becoming. We were remaking ourselves in the process as 'cultural poesis -- and politics -- is the constant reconstitution of selves and others through specific exclusions, conventions, and discursive practices'. (Clifford and Marcus 1986:23-24) Strategically speaking, Murukun had 'tuned-in' to my infancy in understanding Yolngu culture, locating me in techniques of Yolngu education. This was evident in the differences between her crying-songs recorded at funerals and those in our private sessions. In each mode of tuning-in different streams of consciousness emerged, for example, in the discursive formation of her own cryin g-song identities in relation to those of other women singing and in the personal relationship between her knowledge and remaking of the deceased through her crying-songs. Women share a lifetime of streams of song consciousness collated from successive funeral events where, sitting together at the funeral shade, they simultaneously create their own personalised narratives of the deceased through the webs of kin-song, a quite different kind of tuning-in from that which happens between the performer and researcher in the learning process. In the educational context, the layering of her own or other women's personal references to the deceased are often omitted.

Murukun frequently sought to emplace her own subgroup identity as the focal point of the performance in these first lessons. Speaking of her authority to teach me the song, she said, 'My father was Mayamaya and his ancestral identity is Manggani Wurrpurr (snake). Another brother was Badinggu but he is Darrkamawuy [shark]'.(23) By outlining her genealogical connections, Murukun organised both the performance and the discussion of her own identity as a strategic response to me as her pupil, and her granddaughter. I was placed in the situation of not only learning to respect her, but learning to respect the importance of her clan songs in relation to my own Datiwuy shark songs. She would begin with her song of the Cloud that rises over the King Brown snake shade where people are talking.

Here, three main thematic ideas are alternated:

X Yolngu talking at the shade.

Y the cloud forming from the breath of the groups talking to one another.

Z the deep ritual names bundurr of the Ngurruyurrtjurr subgroup.


Warraw' ga Mangan
Clouds over the Ngurruyurrtjurr Djambarrpuyngu Snake Shade

X) Dhiyal ngalimurr wanganhamirrnydja
Here we all talk to each other

X) Warraw 'ngur lumbilumbingurnydja
At the shade Lumbilumbi

Y) Malarrmalarr gudji ngatiti
Speaking in dhangu and dhuwal languages to the cloud

Y) Gudji ngatiti raliyi buwangbuwangdja
The breath from our speech rises to form the cloud called Ngatiti,
Buwangbuwang

Z) Raliyi Wurrpurrtja Wulangayngu
It comes over to us, to the Wurrpurrtja Wulanayngu Djambarrpuyngu

Z) Djanggadpaniny Mangani Wurrpurrdurrdu
Our bundurr is Djanggalpani, Manggani Wurrpurrdurrdu

Z) Djanggatpaniny Manggani Wurrpurrdurrdu
(We are the snake people) Djanggatpaniny Manggani Wurrpurrdurrdu

Y) Manganthinan dharuk ngalawurr
The language is going to the cloud called Ngalawurr

Y) Malarrmalarr Gudji Ngatiti
The dhangu language and the dhuwal languages called Malarrmalarr and
Gudji form the cloud called Ngatiti

X) Dhiyal ngilimurr wanganhamirrnydja
We talk together here

X) Walarritj'ngur Walarritj'ngur
At the shade called Walarritj

X) Lumbilumbingur warraw 'ngurnydja
At the shade called Lumbilumbi

Y) Ngdtil ngatilma Gudij Gudji Ngatiti
In the Gudji language to the cloud Ngatiti
In this song the alternating segments create shifting images. The image of Yolngu sitting talking in the shade is contrasted with their breath being transformed into clouds rising above the shade. In a funeral context, the cloud rising over Nurruyurrtujurr is an indication to other Yolngu that a Ngurruyurrtjurr member has died. When they see the cloud travelling to Yalangbara -- Rirratjingu homeland, and then to Rarrakala -- Barrarrngu and Barrarrparrarr land in the Wessel Islands, they may be able to establish the identity of the deceased because of the implicit links in the song. Clouds are the breath of Yolngu. Their essence drifts over homelands to indicate a death in another group's land by the direction in which they are travelling. While the shapes of Dhuwa and Yirritja clouds are different, the shapes of clouds are shared between groups of the same moiety enfolding multiple identities of group members.

As both a knowledgeable senior leader and a practising Christian, Murukun would always sing these crying-songs two ways: with grieving for her lost relatives and absent loved ones and as parables of faith. She would tell of how the song of her cloud at Nurruyurrtjurr is the breath of people praising God, joining Barrarrngu and Barrarrparrarr with the different Djambarrpuyngu subgroups at her Wotjara snake shade, just as in funerals, groups come together to sing their respect and love for their deceased. Their songs rise to the cloud as they unite to form the body of the family or the church. She remarked:

I am singing about each group. This one He called Dadayana, Marawunggu, Burruwalka, and this one Gululuwanga, Gadulkirri, Wadulnyikpa and this one Gulkalirrngu. Our group is Wotjara Yolngu (King Brown snake people) and those others are Gundangur from Garrata (Shark people) and Miliwurrwurr from the Wessel Islands. God made each mala (group) to be together so they are singing praises to Him.

Although the ritual words of the song remain as they were in the previous rendition, Murukun is seeing inside the cloud, to change its colours of meaning in a shadow-dance of Christian interpretation. (24)


Warraw' ga Mangan
Shade and Cloud

Raliyi dhu Minirringa, Baliyana Rrapudja
The cloud Minirringa, Baliyana, Rrapudja is coming

Miliwurrurrnydja gudji' yunmirr
From the Wessel Islands belonging to Barrarrparrarr and Barrangu where
they used to speak gudji language (25)

Ngatiti Banggilpanggilyurra
The cloud called Ngatiti Banggilpanggil

Ga raliyi Burruwalka Dadayana Letileti Marawunggu
It comes from the Djambarrpuyngu Marapay group called Burruwalka,
Dadayana Letileti Marawunggu

Walarritjngur warraw' ngurnydja
At the shade of the King Brown snake called Walarritj

Warraw' ngurnydja Gumukulngur
At the shade where the Gumuk cloud is

Nhe raliyi Wotjangurnydja
You are coming from Wotja

Gulkalirriyunydja Wotjangur Banggilpanggilyun
The cloud, Gulkalirri, Banggilpanggil is coming from Wotja

Dhiyal Walarritjngur
Here to the king Brown snake's shade
Thus, the process of learning the melody, pitch, timing, timbre and song semantics was a process of recognising sounds as a plurality of patterned meanings. The task of understanding was also one of accumulated absorption rather than strict interpretation: while Murukun would provide glosses on the songs and identify names of places or ancestral beings, eliciting their relationship to one another and to the broader context, her singing was like a kaleidoscopic sphere turning constantly to view changing colours providing additional glimpses of meaning into her extending song web. (26) Lists of names for each song subject would evoke a myriad of nuances of movement, colour or sound relating to the ancestral being at a particular place. Often the names would be synonyms for a place or ancestor -- part of the bundurr (deep ritual names) of the ancestral being. The layering of names allowed multiple aspects of ancestral beings to be viewed simultaneously. (27) And as my learnin g progressed, a freedom of improvisation and creativity emerged that had been lacking in the early sessions. Gone were the condensed repetitive phrases and instead flowing lines emerged, pictures of vitality and intensity, mosaics of ancestral action captured in the essence of improvisation. (28)

In September 1998, I returned to work with Murukun on her song texts. Whilst Murukun was away in Yirrkala visiting relatives, Rrikawuku, a teacher linguist at the school came to see me to say that Murukun was now involved in a music project with the school, Shepherdson College on Galiwin'ku, and Murukun had been recording crying-songs for Yolngu teachers to use in classes providing an additional body of comparative material. When Murukun returned she agreed to record songs with me once more. Now, time, distance and increased familiarity created a new set of dynamics between us, as well as a performative continuum between the earlier research and current performance context. With a shared understanding of Murukun's crying-songs, our discussions about songs were as much part of music making as the performances themselves. In this new context, 'performance provided a frame that invited critical reflection on [previous] communicative processes' (Bauman and Briggs 1990:60). When I had first gone to the field, I h ad asked Murukun if she would teach me to sing from the knowledge of a child to the understanding of a senior woman. Now I could witness the early teaching strategies of condensed repetition mirrored in the texts of recordings made for school-children seven years later. (29)

This time, however, Murukun began with the song of the shark shade, not the snake shade, a shift that was determined by changing contexts as we became more familiar with one another, 'implicitly redefining ourselves and each other [once again] during the course of interaction' (Sangren 1988: 273). (30) This change in emphasis, from singing her subgroup identity to that of the ancestral shark identity, reflected three new strategic responses: 1) my welcome home as her Datiwuy granddaughter; 2) recognition that I already knew the Djambarrpuyngu subgroups joined together under the one ancestral shark; 3) and an acknowledgement that this song was of deep personal significance to our ancestral identities.

In the text the reference to the shark, Gululuwanga, indicates that there are members of the Gundangur Djambarrpuyngu ancestral shark subgroup present in the shade but they are sitting apart from Murukun's snake subgroup. With an identity that encompasses both ancestral beings, Murukun could choose between the song of the snake and the shark, although on this repeat occasion, her song of the main ancestral being bound us together in an emotive reunion of mutual musicality:


(1998) Warraw'
The Gundangur Shark Shade

Warraw' nydja ngarraku dhuwal
This is my shade

Dhawurruwurru rongiyinan
(The shark ancestor is) coming back to Dhawurruwurru

'Gong-djalk yurr ngarraku warraw'nydja
Creator of my shade

Minitjpu dhuwal warraw'
This shade is called Minitjpu

Ngarraku dhuwal warraw'
This is my shade

Gadulkirri wadulnyikpanydja
The shark shade called Gadulkirri Wadulnyikpa

Dhamalamirri Gululuwanga
Dhamalamirri Gululuwanga, (the deep ancestral names of the shark at
Gundangur)

Go nal'yurra ngarraku warraw'nyda
Let's go to the shade

Minitjipurrnydja Dhawurrurr rongiyinan
Returning to the shade called Minitjpu, Dhawurrurr

Ya ---

Warraw' ngarraku bilyurra
Coming back to my shade

Ga minitjpurrnydja
Called Minitjpurr

Go ngalyurr ngarrakuy warraw'nydja
Let's go to my shade

Ngarra dhuwal Yolnguy
I am the Yolngu (for that shade)

Dhamalamirr dhuwal
(A person from) Dhamalamirr

Gadulkirri Wadulnyikpa Yolngu
A shark person called Gadulkirri, Wadulnyikpa

Nhina ngarra marrtji warraw'
I am going to sit at the shade

Nhina ngarra marrtji warraw'nurngydja
I am going to sit at the very shade

Dhawurrurr rongiyiny ngarraku go nal'yurra warraw'nydja
Returning to Dhawurrurr, let's go to my shade

Minitpurrnydja.
Called Minitjpurr
This time, after each crying-song, Murukun's glosses painted vivid pictures of the shark lying in the mouth of the river, its shadow just visible in the shallow water; or of it making its way along the riverbed, hollowing out the creeks. Her songs were 'tone poems' of ecology and the senses they evoked inscribed feelings of her own history and experience as well as memories of past shadow-dances reaffirmed as presently shared friendship. (31) I understood that seeing meant that Murukun might project at any particular moment what Ewing (1990:251) has identified as, 'multiple, inconsistent self-representations that are context-dependent and may shift rapidly'.

As I began to understand how her songs had been strategically formed and located, I became aware that I had not been oblivious to the technical differences between ritual and non-ritual performance contexts. I was fully aware that women cried in non-repetitive lines outside the research context, and I was equally aware that Murukun was using condensed repetition to teach me the structure of these songs. What aroused my interest latterly, however, was the extent to which the strategies that Murukun employed at that time had been consciously adopted by her and employed by other senior women also. I came to ask myself: to what extent were they tuning-in to the political and emotional weight carried by the song and its pertinence for the listener, as well as placing the song in relation to themselves and their memory or recounting of others?

Consequently, there is an intensive reflexive critique that is required in evaluating the nature of the product that we come to label song, and, in turn, a corresponding ambiguity about how we create a musical anthropology. (32) Song is more than the sounds and contexts of performance; it is a way of being musically related, feelingfully entwined by patterns of sound through strategies by which researchers attempt to experience music as a performative process and by which performers engage with us in a reciprocal web of experiential exchange. There is the need for both critique of the experiential language of musical knowing and coming-to-know and an unravelling of the nature of power relations between performers and researcher between what is sung and what is said, and between what is expressed and what is inexpressible. As it is in the music-making process that musical knowledge exists in an uneven exchange with various kinds of power -- political, intellectual, cultural and moral, there is a multivocality inherent in these power relations raising questions about 'how distinct groups ... imagine, describe and comprehend each other'. (Clifford 1988:260)

CONCLUSION

In this comparative analysis of one woman's crying-songs over a six year time span, I have sought to deal with the strategies of Murukun's crying-songs in relation to my own position as a researcher allowing reflection upon her perspective, as well as to engage in a generalist discourse about one woman's principles of singing. In contrast to personal narratives, it has been argued that generalist discourse tends to be a cultural-political configuration of a kind of textual attitude, yet, for Murukun, a generalist discourse was often included in her own song narration. And though I have tried to write Murukun's empowering strategies back into the text, I acknowledge that this act is a disempowering one for her. To deconstruct and reconstruct a text is an act of control and 'in regard to the differential exercise of such control the issue of power arises'. (Bauman and Briggs 1990:76)

Muruknn did not always choose to start every song session in the same way or with the same song. Nor were her glosses and narratives of kin and family ties always logical and cohesive. Through generalist discourse, Murukun made sense of indeterminacy, variation and her own personal strategies in performance practice. She tended to tidy up the discrepancies between performers and song variations, constantly reordering and making sense of the unruly process of music making. Yet, her performance variations did not reflect her tidy rhetoric, leaving open the possibility for future shadow-dances of mourning for her loved ones.

In this discussion, I have attempted to avoid reducing Murukun's crying-songs to one definitive musical moment but, rather, to recognise the performative strategies that account for the researcher in the moment of performance. The disparate nature of ritual and non-ritual recording contexts offers a continuum of performative paranoramas where researching and music making are mutually evolving and constitutive processes. I have argued that we need to uncover and account for the strategies and power relations between performers and researchers that go into generating musical understanding in order to understand the nature of the musical event. By writing about music, we are inextricably involved in another performance of music making, one that comes to have a related, but different tune. Thus, the sounds in our recorders are a product and process of the anthropologist actively conducting fragments of Aboriginal music into notation and text as orchestrated by the performer. The task of translating experience in to textuality when there are multiple subjectivities, strategies and political constraints operating within and beyond our control is to recognise that there is a continuous 'tacking between the 'inside' and 'outside' of events: on the one hand grasping their sense of strategic occurrences and performative positioning empathetically, on the other stepping back to situate these meanings in their wider contexts.' (Clifford 1988:34)

Consequently, the production of this text is itself a reflection upon reflection. Sadly, Murukun died in March 2000 and others are now reproducing her songs in her absence. (33) The writing of this piece is a shadow-dance of a different kind and another layering of strategical positioning is being played out in the knowledge that her musical gift will raise new research questions for the future: how is her memory living on in the reproduction of her song performances by the younger generation who will learn from her recordings at the school; how much of her song technique will be adopted by these younger women; and how will the learning process feed into ritual performance as younger women take up the obligation to cry for those who have passed on?

ACKNOWLEDGMENTS

I am grateful to the Berndt Foundation, University of Western Australia for funding that partly assisted this paper. I also wish to acknowledge the School of Anthropological Studies, Queen's University, Belfast and the Department of Music, York University, Toronto which gave me the opportunity to present this paper at two symposiums where Suzel Reily, Andree Grau, Ellen Koskoff and Beverley Diamond all provided extremely valuable comments. My thanks also go to Suzi Hutchings and Marian Thompson for their assistance and suggestions as well as to Martin Clayton for his advice and comments from three anonymous reviewers.

NOTES

(1.) Elcho Island is known today by its Yolngu name, Galiwin'ku, and is situated six kilometres off the coast of north east Arnhem Land. It is approximately 55kms long and 35 kms wide with a population of about 1200 Yolngu and 100 balanda (non-Indigenous Australians).

(2.) Intersections of ancestral song tracks are beyond the scope of this paper and I cannot do justice to the incredible depth of all Murukun's kin-songs and their links.

(3.) See Clifford (1988:21-54) for a detailed discussion of the authorial presence of the 'other' in the 'experiential, interpretive, dialogical and polyphonic processes at work' in the making of ethnography. Berliner (1993) also explores the nature of knowing as privileged information in his relationship with an mbira musician

(4.) See Kisliuk (1997: 38) for a discussion of techniques of evocation in writing about performance.

(5.) Said (1978:20) posits that 'strategic formation' is a way of analyzing the relationship between texts, and the way in which groups of texts, types of texts, even textual genres come to shape cultural thought and therefore carry their own referential power. He also argues that 'strategic location' refers to an author's position in a text in relation to the material of production.

(6.) While Berndt gives the singer's group identity and relationship to the deceased, no details are provided as to where these songs were recorded, although Yirrkala is a likely location. It does not appear that Catherine Berndt made tape recordings of these songs as none exists in the Berndt collection.

(7.) I am grateful to Dr. John Stanton for his generous assistance in accessing the Berndt collection for the purposes of listening to the Berndt's recordings from Arnhem Land.

(8.) I have recorded and transcribed versions of two Dhuwa songs of the Wawilak cycle, djulwadak, friarbird and wujal, honey ancestral being, that were also transcribed by Berndt, although space does not permit discussion of them here.

(9.) For example, speaking of the Madayin ceremony with Gunapipi, Keen (1994: 250) notes that women had their own interpretations of songs that were similar in form to men's.

(10.) In addition, there have been a significant number of women anthropologists and linguists writing on other topics, e.g. land tenure, Williams (1986); linguistic variations and dialects, Morphy, F. (1977); health and sorcery, Reid (1983).

(11.) Morphy, H. (1978) demonstrates the centrality of women and their circulation in a cycle of exchange and Rudder's (1993) analysis of Yolngu cosmology posits the 'inside' essence of Yolngu identity as intimately related to women.

(12.) They may be sung out hunting and gathering, or at home following a death, but the most common occasion on which they may be heard is at funeral rituals.

(13.) For some excellent recent work on Aboriginal women's music see Mackinlay (2000).

(14.) Bodily harm was inflicted with sticks or rocks in a desire to share in the suffering of the deceased because of the remorse and regret of not having seen or spoken to the person prior to their death; not fulfilling kin duties while the deceased was alive; or a wish for revenge or blame to be attributed to a family in order to avenge the death. In many places these practices are now extant.

Berndt (1950:307) notes how women were criticized in the past for not showing sufficient respect by cutting themselves and received hostile comment. Today the church on Galiwin'ku dissuades women from this practice and prayers are said before the announcement of the deceased is made.

(15.) In the Eastern Kimberleys, Kaberry (1939:209-211) reported that women's wailing comprised mainly sobbing, and ritual words were rarely included, whereas on Melville and Bathurst Islands women's songs were composed for illness and funerals (Berndt 1950:289-305).

(16.) Manikay is the generic term for song. Ritual songs performed by men at funerals are known as bapurru manikay, or funeral songs. For a detailed description of a variety of men's song genres in Arnhem Land see Anderson (1992), Knopoff (1992) and Toner (2000).

(17.) For a more detailed analysis of environmental memory and identity in Yolngu crying-songs see Magowan (2001).

(18.) As all Yolngu are related through the asymmetrical marriage system, balanda who work closely with Yolngu are frequently adopted by a family member and placed within a particular group, creating rights and limitations upon their knowledge and behaviour.

(19.) Berndt (1950:306) also reports that women should 'be well acquainted with the various steps and dances as well as the relevant songs ...'

(20.) For a discussion of women's song as one element in the relational aesthetics between singing, dancing and the processes of didjeridu production and playing in Arnhem Land, see Magowan (1998),

(21.) The shark is also connected to Djapu and Ngaymil because it travelled through these areas on the way to Datiwuy and Djambarrpuyngu homelands.

(22.) Foucault (1972:49) notes that "discourses ... [are] ... practices that systematically form the objects of which they speak".

(23.) Murukun's father, Mayamaya, is recognised as an important leader because he was the son of Djumbala, as was his stepbrother, Badinggu. Djumbala was renowned for his wives, 9 in total, which gave him increased ritual importance. He was also an apical figure in the continuity of the Djambarrpuyngu clan, having produced 35 children. Mayamaya increased the snake subgroup's strength fathering 28 children from 6 wives, while Badinggu supplied the shark subgroup with 17 children from 3 wives.

(24.) For a fuller discussion of Christian interpretations of Yolngu ritual songs and Aboriginal theology see Magowan (1999).

(25.) Barrarrngu and Barrarrparr members are now deceased although the Ancestral Law of these groups is held by the Murrungun aggregate of Naymil, Datiwuy, Murrungun and Golumala.

(26.) For a brilliant exegesis of the complexity of language in Yolngu men's songs see Keen (1977).

(27.) As the allusions and significances of song names are polyvalent there are many layers of meaning that cannot be extrapolated here. Keen (1994:275) has noted that radical variability in interpretation between song meanings is possible because songs are not anchored in ordinary language. Also, for a highly sophisticated analysis of a multiplicity of meanings in ancestral landscape, art and identity see Morphy (1991).

(28.) Berndt (1950:306) also identifies 'some individual modification in the form of variation on an existing rhythm, or alteration of certain words whilst the traditional basis remains relatively stable...'

(29.) The text of the bar-shouldered dove on p.94 was taken from the 1998 recording done for the school.

(30.) Bauman and Briggs (1990: 68) note that 'shifts in social interactions can be discerned by attending to the "contextualisation cues" that signal which features of the settings are used by interactants in producing interpretive frameworks'.

(31.) The concept of the 'tone poem' is borrowed from the symphonic term for compositions that conjure up images of landscape imbued with national sentiment, such as Sibelius' Finlandia.

(32.) Within the Context of Australian Aboriginal studies of music, there have been a variety of approaches to these problems of reflexivity. Barwick (1989:12) has argued that music is capable of generating new ways of understanding Aboriginal thought and culture, whilst also pointing Out that no ethnomusicologist has been able to provide a comprehensive theory of Aboriginal music due to the complex relationship between the melody, text and spoken explanations of the song (see also Ellis and Barwick 1987).

(33.) NB I have been given permission to write Murukun's name but it is still not allowed to be spoken, the proper practice following a death.

REFERENCES

ASAD, T. 1986. The Concept of Cultural Translation in British Social Anthropology. In J. Clifford and G.E. Marcus (eds) Writing Culture: the poetics and politics of ethuography. Berkeley, London: University of California Press.

ANDERSON, G. 1992. Mularra: A Clan Song Series From Central Arnhem Land. Unpublished PhD thesis, University of Sydney.

BARWICK, L. 1989. Creative (ir)regularities: the intermeshing of text and melody in performance of Central Australian song. Australian Aboriginal Studies. 1:12-28.

BAUMAN, R. and C. BRIGGS. 1990. Poetics and Performance as Critical Perspectives on Language and Social Life. Annual Review of Anthropology 19:59-88.

BELL, D. 1983. Daughters of the Dreaming, 2nd edn. Minneapolis: University of Minnesota Press.

BERLINER, P. 1993. [1978] The Soul of Mbira: Music and Traditions of the Shona People of Zimbabwe. 2nd ed. Chicago: University of Chicago Press.

BERNDT, C.H. 1950. Expressions of grief among Aboriginal women. Oceania XX (4): 286-332.

BONFIELD, A. 1998. The Dhanbul Ritual. Unpublished PhD thesis. Manchester University.

CLAYTON, M. n.d. Experience, discourse and meaning: thoughts on musical meaning in India and elsewhere. Introduction to Special issue of British Journal of Ethnomusicology (forthcoming).

CLIFFORD, J. 1988. The Predicament of Culture: Twentieth Century Ethnography, Literature, and Art. Cambridge, Mass.: Harvard University Press.

CLIFFORD, J. and G. MARCUS. 1986. Writing Culture: the poetics and politics of ethnography. Berkeley: University of California Press

COOLEY, T. 1997 Casting Shadows in the Field: an introduction. In G. Barz and T. J. Cooley (eds) Shadows in the Field: New perspectives for fieldwork in ethnomusicology. Oxford: OUP: 3-19.

DUMONT 1978. The headman and I: Ambiguity and ambivalence in the fieldworking experience. Austin: University of Texas Press.

ELLIS, C. AND L. BARWICK 1987. Musical syntax and the problem of meaning in a Central Australian Songline. Musicology Australia 10: 41-57.

EWING, K. 1990. The illusion of Wholeness: Culture, Self and the Experience of Inconsistency. Ethos: 18:251-78.

FOUCAULT, M. 1972. The Archaeology of Knowledge and Discourse on Language. New York: Pantheon.

HUTCHERSON, G. 1998. Gong-Wapitja: Women and Art from Yirrkala. Aboriginal Studies Press. Canberra.

KABERRY, P. 1939. Aboriginal Woman, Sacred and Profane. London: Routledge.

KEEN, I. 1977. Ambiguity in Yolngu Religious Language. Camberra Anthropology 1(1): 33-50.

1994. Knowledge and Secrecy in an Aboriginal Religion. Oxford: Oxford University Press.

KISLIUK, M. 1997. (Un)Doing Fieldwork: Sharing songs, sharing lives. In G. Barz and T. J. Cooley (eds.) Shadows in the Field: New perspectives for fieldwork in ethnomusicology. Oxford: OUP. 23-44.

KNOPOFF, S. 1992. Yuta Manikay: Juxtaposition of Ancestral and Contemporary Elements in the performance of Yolngu Clan Songs. Yearbook of the International Council for Traditional Music. 24: 138-53.

MACKINLAY, E. 2000. Maintaining grandmothers' law: female song partners in Yanyuwa culture. In Musicology Australia 23:76-98.

MAGOWAN, F. 1998. Singing the light: sense and sensation in Yolngu performance. RES: Anthropology and Aesthetics 34: 192-207.

1999. The joy of mourning: Resacralizing 'the sacred' music of Yolngu Christianity and Aboriginal theology. Anthropological Forum 9 (1): 11-36.

2001. Crying to Remember: Reproducing personhood and community. In B. Attwood and F. Magowan (eds) Indigenous Memory and History in Australia and New Zealand. Sydney: Allen and Unwin.

MORPHY, F. 1977. Language and Moiety: Sociolectal Variation in a Yurlngu Language of North-east Arnhem Land. Camberra Anthropology 1(1): 516.

MORPHY, H. 1978. Rights in Painting and rights in women: A consideration of some of the basic problems posed by the asymmetry of the 'Murngin' system. In J.Specht and P.White (eds) Trade and Exchange in Oceania and Australia. Mankind Special Issue 11(3): 208-219.

1991. Ancestral Connections: Art and an Aboriginal System of Knowledge. Chicago: Chicago University Press.

MOYLE, A. 1974. North Australian Music. Taxonomic Approach to the Study of Aboriginal Song Performances. Unpublished PhD thesis. Monash University, Victoria.

RABINOW, P. 1977. Reflections on fieldwork in Morocco. Berkeley: University of California Press.

REID, J. 1983. Sorcerers and Healing Spirits: continuity and Change in an aboriginal Medical System. Canberra: Australian National University.

RICE, T. 1994. May it Fill Your Soul: Experiencing Bulgarian Folk Music. Chicago: University of Chicago Press.

RUDDER, J. 1993. Yolngu Cosmology. Unpublished PhD thesis. Canberra: ANU.

SAID, E.W. 1978. Orientalism: Western Conceptions of the Orient. London: Penguin Books.

SANGREN, S. 1988. Rhetoric and the authority of ethnography: 'postmodernism' and social reproduction of texts. Current Anthropology June 29(3): 405--425.

SEEGER, A. 1987. Why Suya Sing: A Musical Anthropology of an Amazonian People. Cambridge: Cambridge University Press.

SLOTTE, I. 1997. We are Family, We are One: An Aboriginal Christian Movement in Arnhem Land, Australia. Unpublished PhD thesis. Australian National University.

STUBINGTON, J. 1978. Yolngu Manikay: Modem Performances of Australian Aboriginal Clan Songs. Unpublished PhD thesis. Monash University.

TAMISARI, F. 1998. Body, vision and movement: in the footprints of the ancestors. Oceania 68(4): 249--271.

TONER, P. 2000. Ideology, influence and innovation: the impact of Macassan contact on Yolngu music. Perfect Beat. 5(1): 22--41.

WILLIAMS, N. 1986. The Yolngu and their Land: a system of land tenure and the fight for its recognition. Canberra: Australian Institute of Aboriginal studies.