Ownership and resource use on islands off the Liverpool River, Northern Territory

Peter Cooke and Gowan Armstrong

During 1995 and 1996 the Northern Land Council (NLC) Native Title Unit commissioned preliminary surveys of Aboriginal interests in seas and submerged lands at a number of locations along the coast of the Northern Territory. In May 1996, Peter Cooke provided the NLC with a draft report covering an area extending eastward from de Courcy Head, on the eastern boundary of the current marine native title claim lodged by people at Minjilang (Croker Island) to the mouth of the Blyth River in Central Arnhem Land. From de Courcy Head to the Blyth River mouth is a distance of about 200km in a straight line. However, the length of coastline between those two points (including the coastlines of North and South Goulburn Islands) is about twice that distance.

The first part of this chapter summarises the results of archival and field research undertaken by Cooke, while the second focusses on a predominantly marine estate in the mouth of the Liverpool River. Complex arrangements for access and resource use under local customary law are described and discussed, drawing on data collected by Armstrong during the 1960s and early 1970s and by Cooke during the 1996 survey.

Aboriginal people have been in Australia a minimum of between 40,000 and 30,000 years BP and less conclusive evidence points to site dates as early as 60,000BP (Bowdler 1993:60). Between 26,000 and 11,000 BP the sea level was between 65 and 130m lower than today: the lowest levels are estimated at 18,000 years BP (Smith et al frontispiece). When these lower sea levels are plotted on modern marine charts the dynamic nature of the coast becomes apparent: the coastline has retreated landward more than 100km over the last 10,000 years. 286

Within the oral history of this region are stories, framed in mythological terms, which may relate to these geological/hydrological events that happened thousands, rather than hundreds, of years ago. At Goulburn Island there is a mythologically interpreted record of the separation of Weyra and Warruwi, or North and South Goulburn Islands. At Maningrida, ancestral beings are deemed responsible for the separation of Entrance Island from the mainland. A large freshwater swamp on the former ‘land bridge’ connecting it with the mainland is now shallow sea where fresh water rises to the surface from submarine springs.

Defining the seaward extent of marine tenure

Defining the seaward extent of Aboriginal marine territories is not straightforward. Of the generalised descriptions put forward by various Aboriginal people interviewed during the survey that advanced by Ralph Gurmurdul, a principal Aboriginal consultant, seems to deal best in a comprehensive way with this issue. Gurmurdul says Aboriginal people’s sea territory extends to:

1. Those places where they go hunting or travelling.

2. Those places for which Aboriginal owners have names.

3. Those places known to Aboriginal people as being inhabited by ancestral beings.

The seaward limits for travel by dugout or bark canoe are not clear. However, during the survey Albert Wurrdjal (Worrdjol), an experienced canoesman, indicated that it was not uncommon for people to travel by canoe in a straight line from the mission at Warruwi (Goulburn Island) to around Cuthbert Point, a distance of about 45km. The midpoint of this journey is about 12km from the coastline.

Of sites named during the survey, the most distant from shore would appear to be Wulurunbu and Lingardji, regarded by people from the Goulburn Island area as islands inhabited by the spirits of the dead. These islands, which are said to remain hidden under the sea by day and emerge by night, are regarded as ‘real’ places. Some older people can 287indicate the direction of the islands from various points along the coast. A rough ‘triangulation’ of those directions suggest that they may be a large area of shoal at about 5–6 fathoms lying about 20km east of North Goulburn Island in waters generally at 13–15 fathoms.

While camped near Cuthbert Point, the survey party watched lightning visible far out to sea. Senior Aboriginal consultants said this lightning indicated the presence of a particular spirit being and they interpreted the westward movement of the lightning activity as the movement of that spirit being across the sea.

During the 1996 survey, 350 named coastal and marine locations were mapped between the Blyth River and de Courcy Head.

Identifying owners of marine territory

Aboriginal people within the 1996 survey area asserted that there are no differences between owning land and sea. Estates and interests on land extend into adjoining seas and an estate may be comprised of mainland terrestrial, littoral, marine and insular components. These elements make up a single area which is usually known by what is referred to in English as a ‘big name’. Such a big name is often that of a pre-eminent spiritual site within that territory, although it might also derive from a focal occupation location, such as a modern outstation. Sites of spiritual significance may be found in any of these ecological zones within an estate.

Aboriginal people in the survey area usually indicate land ownership by identifying a group defined by patrilineal descent as having primary responsibility for a named estate. These patrilineal groups are known by a number of different names — yuwurrumu, kunmukurrkurr, namanamadj, yákkarrarra and bábburr. Nguya is a more generalised word for ‘group’ used often in most western Arnhem Land languages as a synonym for one of the more particular names for a patrilineal group, locally referred to in English as a ‘clan’. 288

Although the more specific words are to an extent associated with particular languages and locations, they are also known and used interchangeably and widely throughout the region. This is particularly so along the coast. Cooke and other researchers working for the NLC have found the word namanamadj most commonly used at Minjilang for the patrilineally defined primary land owning group. Armstrong (1967) found the word namanamadj used by Kunibidji (Gunavidji) people from the Liverpool River in a way in which the meaning of the word signified not only the landowning group but also their estate. A Kunibidji man might speak of a particular named section of Kunibidji territory, with its known boundaries and which is owned by his father’s father/father’ father’s sister, his own father/father’ sister and brothers/sisters and all relatives traced in the patriline, as his namanamadj.

Elkin, Berndt and Berndt (1951) translate namanamadj loosely as ‘follow my father’ and this seems apt. Children normally belong to the patrilineal descent group to which their father belongs. However, only males can pass this affiliation on. Women of the group bear children who ‘follow their fathers’ from other nguya. In special cases where a woman may have had husbands belonging to different nguya, the children may identify with their biological father, or perhaps accept bestowal of membership of the group of a stepfather. Similarly, children who may have been the result of a liaison rather than a marriage, may, as they grow up, choose between the nguya of their biological or social father. People may be adopted into a nguya without it involving a sexual relationship between their mother and a man of a particular nguya. This is the case in one instance, where a childless man, the last of his descent group, decided to bestow membership of his group on another man’s children. In this case and others, a person may follow two fathers — claiming affiliation with two nguya, through biological or social paternity, including adoption. Dual affiliation is certainly possible, but it may provide the grounds for other people to suggest unreasonable opportunism, particularly if the dual membership brings financial gain.

The survey identified 32 named, patrilineally defined descent groups with primary claim to estates along the 400km of coastline 289surveyed. Information on some 1200 living and deceased members of these groups was entered onto a genealogical database file (Reunion software) enabling easy formatting of lineages from apical individuals or purported siblings. Data was drawn from field recording, the NLC Land Interest Reference and 1979 printouts from the (then) DAA Aboriginal Population Record database. Most of these genealogical charts extend back to individuals born about 1900 but in some cases they extend perhaps as far as the mid nineteenth century.

About half of the 32 groups have a distinctive nguya name with a primary interest in one estate. In the other cases the same nguya name, or a cognate term, may apply to a number of lineages associated with separate estates. In one case in 1996, three slightly different versions of one name were recorded for lineages associated with five separate estates. In another case, a second name is used in addition to a common nguya name to distinguish amongst four lineages with primary interests in four separate estates. Armstrong (unpublished field notes) observed that it is common in the Liverpool River area for a group to have a ‘big name’ and a ‘small name’. There seems reasonable grounds to suggest that in some cases at least, a second name preserves the name of an extinct group whose territory has been subject to succession.

Thus, as the coastal physical environment is dynamic, so too is the coastal social and cultural milieu. People must deal with the decline and extinction of groups and find ways to accommodate long term ‘migrants’. Armstrong (unpublished field notes) observed that there is evidence that some yákkarrarra (patrilineally defined groups) merge and that in situations when the members of a yákkarrarra all die, their estate is incorporated into the territory of another yákkarrarra. The 1996 survey recorded five estates where the processes of succession were at various stages. Some cases are straightforward andthe new owners easily and quickly find acceptance locally, but in other situations, succession may be contentious and may take decades to be resolved.

Lamilami (1974:136) noted that connections between same (and similar) named groups associated with different languages and different estates are widely recognised. He talked about his coastally 290situated Maung group, Maiirwulidj and the similar-named Neinggu (Kunwinjku) inland group, Mairrgulidj and also a Yiwaidja group of the same name. As he observed: ‘I think they must have been related way back in their ancestors’.

In some cases ancestral connections between same-name groups with separate estates may be remembered clearly and this may lead members of the two groups to assert shared interests in one another’s lands. One lineage may assert seniority within such an aggregation of people and estates but over time a ‘subordinate lineage’ is likely to assert its own independent identity and rights of decision-making in respect of its particular estate.

Nguya with quite distinctively different names and separately named estates may also propose that they regard these estates as held in common to some degree at least. Such arrangements are widely referred to in western Arnhem Land as ‘company’. In one example from the 1996 survey, the basis for a strong company arrangement between adjacent, same moiety nguya was that senior members of both nguya had the same mother. She had been married first to a man from nguya A and then after his death she married a man from nguya B. A senior member of one of these two groups identified the implications of this company arrangement as:

1. Both groups discuss proposals affecting either parcel of land and make decisions, as a group, but recognising to a degree separateness of interest in each parcel.

2. One partner clan does not make a decision on major issues by itself for its own land or for the other ‘half’ of company land.

3. Each party recognises the right of the other group to be asked for access to their land.

4. Members of the clans share equally in income derived from either parcel.

5. Members of both clans have free access to each other’s lands for normal resource use.

This example also perhaps demonstrates one way in which matrifiliation may strongly influence land ownership, despite the acknowledged 291primacy of patrifiliation in land ownership matters. While the patrilineal descent approach realistically represents the primary contemporary determinant for land ownership in Western Arnhem Land, behind that lies a complex web of other factors which may ultimately decide an issue where a patrilineal connection is not completely straightforward. It is difficult to generalise about such situations because the parameters of legitimate process in the Western Arnhem Land system of tenure are broad and allow many paths for a person to press a claim where there is no clear and extant nexus between a patrilineage and an estate.

People in Western Arnhem Land generally talk about who country belongs to by reference back to focal ancestors who lived on the land and were associated with it in a mundane way, as well as also celebrating its spiritual attributes in ceremony. Descendants of the male line can call this ‘my country’ and hold the broadest range of rights. The descendants of the women of the group also enjoy considerable rights.

While it is difficult to compare or categorise in a hierarchical way the rights of the patrilineal sons and daughters of a clan and the descendants of women of that clan without inadvertently freezing ongoing social process, it is possible to identify various incidents of native title for descendants of the earliest recalled members of the land-owning nguya, whether the connection be through mother or father. These incidents are mediated by various factors, not the least of which are age and gender.

Some rights and responsibilities of the patrifiliated recorded during the 1996 survey include:

1. A member of an owning group has the right to be asked for permission for someone to enter his or her land or sea country. A senior person has the right to expect that they, rather than a junior member of the group, would be asked.

2. Members of an owning group have the right to grant or refuse a person permission to enter or stay on land or seas within their estate (again seniority is relevant).

3. A senior member of a group has the right to expect to be offered a share of any resources harvested in, or income derived from, land or sea 292within their group’s estate. Where non-traditional activites are involved, it may be that all members of the group are entitled to a share, i.e. ‘royalties’ or ‘clan money’.

4. Members of an owning group have the right to be inducted into the religious life of their group and to be a joint holder of the group’s sacred property, such rights being mediated by age, experience and gender.

5. Members of an owning group have rights to free access to their lands and waters, constrained at times only by temporary closures to some members of a group because of ceremonial activity or because some sites may be ‘too dangerous’ for their owners to approach.

6. Senior members of an owning group have the right to close all or part of seas and land within their estate after the death of important people and subsequently, the right to open access again.

7. Senior members of an owning group have the right to allocate names associated with their estate to members of their own group or to people in more distantly related groups.

8. Members of an owning group have a responsibility to see that their country and the spirits of that country do not harm visitors on the country.

9. Members of an owning group have a responsibility to pass on knowledge of their clans’ physical and spiritual resources to their descendants.

Some rights and responsibilities which operate through the sons and daughters of women of a clan include:

1. A responsibility of men in particular, but also women, to progressively learn the ritual associated with the land and sea country of their mothers and mother’s brothers and to assist members of that group to perform ritual associated with the country.

2. A responsibility as above in respect of sacred sites on mother’s and mother’s brother’s sea and land country.

3. A right to free use and access to resource in one’s mother’s and mother’s brother’s sea and land country. These use rights extend but with diminishing strength to more distant matrifiliates.

4. A right to challenge strangers found on ones mother’s and mother’s brother’s sea and land country and demand to know who has given 293them permission to be there. A right to insist that such people leave the estate if they have not been authorised to be there by appropriate members of the patrilineal group.

5. A right to speak up for one’s mother’s and mother’s brother’s sea and land country in contentious issues such as succession, with an expectation that one’s views will be respected in any outcome. Again strength of interest is strongly mediated by age.

Rights in one’s mother’s mother’s country are strengthened by the fact that (conventionally) it is of the same patrimoiety as one’s own country. The major contemporary ceremonies of Kunabibi and Yabburddurrwa, each have strong prohibitions against ‘mixing up’ moiety determined roles. Unconventional marriages which result in fathers and children being of opposite moiety therefore present some problems in the ritual sphere. However, the increasing regularity of performances of these ceremonies in the survey area shows that such problems are not insoluble. At the time of writing, a Kunabibi was in progress at Minjilang, two more near Maningrida as well as a Yabburddurrwa near Maningrida. As many as a thousand people may attend the conclusion of such events in this region.

In the east of the survey area there is a greater tendency for people to emphasise the importance of intra-clan patrimoiety homogeneity than there is in the west where there is more emphasis on correctness of matters relating to matrimoieties and the matrilineal semimoieties, although not so great as to override the importance of patrilineally defined local groups in land matters.

As well as identity within a language group and within a patrilineally defined nguya, people also belong to social groupings which are somewhat more difficult to define. Lamilami (1974) discussed a number of these, in some cases using the word ‘wara’ in connection with a name. These names often mean ‘mob from such and such a place’. They do not refer only to members of the particular landowning group from that place and it appears they are usually names which might apply to groups which come together in particular places during the seasonal cycle, often to pursue some joint economic strategy, whether 294fishing or hunting geese or so on. Some examples of these words are: Maindjinaidj (Manjdjinadj), a very extensive coastal grouping; Muwamal, people from the coast inside Junction Bay who moved to around Hall Point in the wet season to escape mosquitoes; Mabárnad, coast dwelling Ndjébbana-speaking clans; and Márro, coast dwelling Nakkara-speaking clans. They do not say much about land and sea tenure except perhaps that they indicate groups who shared access to one another’s lands and sea estates on a regular basis and therefore established fairly persistent rights of access and use.

Some groups may use one word in respect of both a people and their language, e.g. Maung language, Maung people. In other areas, different terms are used for people and for their language, e.g. the Kunibidji people from the Liverpool River area speak Ndjébbana1 and the Márro people on the coast to their east speak Nakkára, while the Warlang to the west of the Kunibidji speak Kunbarlang. Within the region, language may be an important qualifying factor in various issues, including succession. Some people might, for example, propose that a patrilineal owning group with the same name as an extinct group might succeed to that extinct group’s estate. Other people might oppose the succession on the grounds that those people are from ‘too far away, different language’, if the genealogical connections between the two groups are lost in time and the distant group has little experience or knowledge of the land and sea country under succession.

Language is also an important factor in defining individuals and groups with resource rights associated with particular areas of land. Speaking the ‘right’ language for a place is not just of mundane importance — it is central to the relationship between people and the ancestral spirits of their estates and the land and sea resources within those estates. Amongst the Ndjébbana-speaking Kunibidji of the Liverpool River, these ancestral spirits are known as múya (singular)

295or barramúyiba (plural) and are associated with particular estates and nguya. The Kunibidji believe that these spirits are watching over their country and the actions of people on the country and that they have the power to affect outcomes in everyday life, particularly failure or success in hunting. They can be persuaded to be generous with natural resources, but only if asked in their own language.

The lúrra fishing expedition

Armstrong was the first to document the ritualised fishing expedition called lúrra amongst the Kunibidji. The primary qualification for participation in this ritual is language.

The lúrra takes place within the estate of the Wúrnal nguya who, in company with Karddúrra and Malandjárridj, are the owners of a large area of sea country in and off the mouth of the Liverpool River—see Figure 11:1. The Malandjarridj primary interest is on the mainland, east of Ndjúdda (North East Point), the Karddúrra primary interest around Ndjúdda and Kabálko (Entrance Island), while the main Wúrnal terrestrial interests are the two islands, Kabálko and Ngarráku (Haul Round Island). Kabálko is the larger of these islands, about 1.5km along the east-west axis and about the same north-south. On the northern side are a series of stone fireplaces the Macassans, used for processing trepang, until they were prevented by legislation from continuing their trade with Australia in 1907.

There are 11 Kunibidji nguya extant, all speaking Ndjébbana language. Eight coastal nguya speak a njárlkkidj (hard) form of Ndjébbana and three inland groups speak marndálangurrnga (soft) Ndjébbana. The coastal groups refer to their language as ‘seagull language’ or malamaldárra (generic name for tern) language. There also seem to be minor differences between djówanga (duwa) and yirriddjanga (yirridjdja) moiety speakers (Graham McKay pers. comm.). There are six clans regarded as djówanga and five clans regarded as yirriddjanga. However, 296the majority of adults in two clans, are of the opposite moiety because of unconventional marriages in previous generations.

A map of the Liverpool River Estuary. There is an inset map of Australia with the larger map area marked.

Figure 11:1 Liverpool River estuary

Gowan Armstrong was at Maningrida when the last lúrra took place at Kabálko and made this record of it:

There are two main qualifications for participation:

1. The men had to speak the ‘sea language’ dialect of Gunabidji (Kunibidji). Men from southern/inland clans whose dialect had affinities with Gungaragone (Gungurrgoni) and other inland groups were excluded—unless their mothers came from a ‘sea language’ speaking group and had transmitted their mother tongue to their sons. A certain Gunwinggu (Kunwinjku) man and a Walang (Warlang) man were able to qualify for this reason.297

2. Only fully initiated youths and men could participate, and for a youth this usually meant being present at three sacred ceremonies. Among other things lúrra was an age grading ceremony.

The ceremony takes place in duwa (djówanga) territory but men from yiritja (yirridjdjanga) country have an active role in helping to organise it. The senior yiritja men choose a period toward the end of the wet season (April in 1964) and others listen for them to ask, ‘you like fish?’ In 1964 Mangiru (Dúkurrdji leader) talked to everybody and ‘pushed’ them. Five ‘new men’ participated for the first time. Four older men who were eligible to go, and were by no means incapacitated, did not go and stayed in the camp.

The men travelled from Maningrida Settlement to Juda Point (Ndjúdda) where they began collecting the black berries or ‘plums’ from the malayman (marlémarla) trees (Diospyros calycantha).2 They also gathered paperbark, which they bound together to make torches (djit [djird]).

Late that afternoon they loaded the dilly bags full of berries and the bark torches into their canoes and travelled in line ‘level’ across to the beach Namalála on the eastern side of Entrance Island Gabalgu (Kabálko). Two other possible sites for lúrra could have been Nagayela or Ngaraku (Ngarráku, Haul Round Island). In 1964 at Namalala the men told me that they smeared their bodies with mud, and kept their eyes closed as they hammered the berries with hammer stones to protect themselves from the burning effects of the juice. The hammering broke the skin of the berries to release the juice. The berries treated thus were put into dilly bags.

As darkness fell the men were seated on the sand and the new men in particular were seated on one side of the rocky area where the hammering was done and told not to talk or laugh. They had to sit with legs outstretched and with their hands on their knees. They were told the múya (the Dreaming one, [múya]) listens and if he hears the new men speaking he will not give fish. They were permitted to use sign language.

298Fires were also lit on the sand. When the tide began to go out, they would listen for the calls of the sea birds. When birds began to call, a leader would say ‘oh, we are lucky. A lot of fish. You young men, don’t talk’.

One man with firestick in hand would then be sent into the knee deep water, ringed by rocks, looking ‘like a billabong’. As he confirmed that the fish were there everyone took up either a dilly bag full of broken berries or a triangular fish net or a djit which he lit. They moved forward ‘like balánda (white men) quick march’. As they entered the water the dilly bags were immersed and whirled about to spread the toxic juice, moving particularly around the sides of the rocky depression. As the fish were paralysed and gathered into the nets, they were placed in the dilly bags. When the leaders are satisfied that all the fish have been taken, they are brought back to the rocks where the berries were crushed and spread out. The older men then addressed the new men ‘You people can talk now. The muyu spirit belonging to this country hears you now’. The older men give some fish to the younger ones to cook and eat. They say, ‘the people of this country are one relation’. Presently the fish are loaded into the canoes and taken across to the women waiting at Juda Point. They are happy and say ‘plenty fish’.

Men told me that if someone was present who could not speak the sea language of Gunavidji the fish would not come and the men would return empty handed.

A lúrra ceremony was carried out ‘in Mr Drysdale’s time’, possibly 1958 and in our time in 1964 when we could see the red flares of the torches across on Entrance Is around 8pm. Another was planned for 1965 but because of continuous rain it was postponed and finally not held. A couple of older men complained that European staff people were going out to Entrance Island to fish and would sometimes camp out overnight. Because of the possibility of Europeans being on the island they were reluctant to plan the ceremony.

No singing or dancing are involved in the lúrra ritual. Jimmy Borwarg and Johnny Godawa’s (Garduwa) son said, ‘We born from Juda 299Point. We talk like seabird Mula-mulara (malamalárra)’. I believe this to be one of the terns (Armstrong, fieldnotes).

During the 1996 survey, Cooke identified ten men who did participate in lúrra and who are still alive. Two of these men, Albert Wurrdjal and Peter Marralwanga were interviewed and provided descriptions of lúrra generally consistent with Armstrong’s report.

Wurrdjal, in particular, stressed the importance of the ‘language test’ for participation.

But some people they’re not going. Some djówanga and some yirriddjanga (moieties). My language, I’m talking really Ndjébbana. Really soft one Ndjébbana orright. My wife (from inland), or even her brother they not allowed. … This is very danger(ous). Like ceremony one. If someone (does) wrong you get trouble, same as marddáyin (highly sacred and dangerous ritual). No allow balánda, no allow Nakkara, no Kunbarlang, no some Kunibidji. Only Kunibidji from here (seaside). … They (spirit beings) can’t put fish if wrong people. If they see nother blokes, all the devil will say, ‘different blokes, we can’t put that fish’. Boss people talk to all the spirit. They tell them about that new man, ‘you saw him (when he was) little boy, him here now, we show him here tonight’. … New man can’t talk (uses whistles and tongue clicks only). We had to be very quiet otherwise that djówanga debil from there (Kabálko) and Ndjúdda come out. He comes from jungle way on the point and two points and looks after the lighthouse (on Ngarráku). He belongs to that country.

Wurrdjal’s recollection is that the young men were still prohibited from speaking until after fish had been cooked and eaten. When some of the fish have been cooked and eaten by the young men, the leader asks them if they are getting full. He asks again, ‘Full?’ and then takes off the prohibition on speaking by wiping sweat across their mouths.

The main lúrra site is in a ‘pocket’ on the northern side of Kabálko where rocks make a natural enclosure at the right tidal elevation. Peter Marralwanga says the lúrra was held on a full moon towards the end of the wet season, while the grass was still long and standing. 300

The full moon is accompanied by a spring tide rising during the night. Towards the end of the wet season, large schools of a number of species of mullet come in around the island. Of these Wurrdjal says the main target species in the lúrra was the large mullet barrábarrakabulúyara, which has a darkish body and eyes reflecting red in the light from paperbark torches. This is probably the diamond-scaled mullet, Liza vaigiensis, although other mullet species were also caught. The diamond-scaled mullet has been recorded up to 4.5kg in weight (Grant 1982:517). Wurrdjal and Marralwanga say ‘might be hundreds’ of fish were caught.

After participation in a lúrra young men were then allowed to participate in egg harvesting on Ngarráku, Haul Round Island.

Sea bird egg harvesting on Haul Round Island

At high tide Haul Round Island is perhaps half the size of a football oval. There are no trees—only patches of long grass, tangles of Ipomea vine and areas of the prickle caltrop (Tribulus sp). Jagged rocks appear from the sea over a large area to the east and north as the shelf of ironstone on which the sand island sits emerges on the falling tide. Five years ago biologists from the Parks and Wildlife Commission of the Northern Territory began to collect information on seabird breeding on Haul Round Island as part of a larger project across north Australian coastal waters. Ray Chatto, the principal biologist involved, has made five or six aerial surveys of the island and three or four ground visits. His records indicate that during the breeding season (late wet, early dry) there are many thousands of bridled terns, more than 5000 roseate terns, more than 30 pairs of silver gulls and small numbers of crested and black-naped terns on the island. Pied cormorants breed on a rocky reef/island with mangroves which is only a few hundred metres from the sand island. This is one of only six or eight such sites known in the Top End. The breeding population of roseate terns is regarded as being of international conservation significance (Chatto 1996:pers com). 301

Not surprisingly, the island and its resources are also regarded as highly significant by the local community. Even before people had the tools and technology to make dugout canoes they used folded and sewn canoes of stringybark (Eucalyptus tetrodonta) to travel to Ngarráku, six kilometres north of Kabálko, which in turn is three kilometres from Ndjúdda. From the coast around Hawkesbury Point to Ngarráku is about nine kilometres. The stability and efficiency of these canoes was noted by Thomson (1952:3):

A large canoe made from a single sheet of bark, which I saw at Rolling Bay on the north coast of Arnhem Land in 1943, had an overall length of 15 feet and measured 12ft 9 inches along the keel. This canoe had a beam of 3 feet and a maximum depth inside the hull of 1 foot 11 inches. It carried a crew of four men comfortably and made journeys at least two or three miles out into the sea even under choppy conditions without any difficulty. One such bark canoe under actual test made better time in a choppy sea than a ship’s dinghy driven by two experienced men each using two oars.

The oldest living male Kunibidji today is Johnny Naliba, from the coastal clan adjoining that of Wúrnal and Karddúrra. In 1926, while Naliba was still in utero, his father and his father’s cousin went to Ngarráku to gather sea eggs, travelling by bark canoe from Kabálko. They never returned and it is presumed the canoe broke up and they drowned. Older people today assert that someone must have sabotaged the canoe or they would never have been lost: such experienced seafarers don’t just drown! The dangerous nature of the journey is one obvious reason why women and children were not allowed to go to the island.

Hunting pressure on the island escalated astronomically with the growth of the government ‘settlement’ at Maningrida after it was established in 1957 and with the coming of dinghies and outboard motors in the early 1970s. The patrol officer Gordon Sweeney noted about 130 Kunibidji at two locations east of the Liverpool in 1939. One might presume the usual presence of, say, a maximum further 70–100 people on the western, Warlang side of the estuary. In 1960 the population at 302Maningrida was 485 and by 1965 it had risen to 800 as more groups were drawn into the new settlement. By 1972 there were about 1100 Aboriginal people and more than 150 non-Aboriginal staff and their families. In 1996 the Maningrida clinic carried health records for more than 2200 people but with the return to country since 1970 the population of Maningrida in the dry season is now about 800 Aboriginal people, rising to perhaps 1200 in the wet season. As well as the Kunibidji, the more or less permanent population now includes close countrymen such as Gungurrgoni, Nakkara, Warlang and some Kunwinjku speakers. More distant groups with longterm residential association include Maung, Rembarrnga, Dangbon, Djinang, Burarra and Gun-nartpa, as well as a smaller number of families from further east.

Non-Aboriginal staff brought with them western notions of the sea as a commons belonging to everyone and they generally ignored the traditional ownership of sea country. Similarly a Government-sponsored fisheries project in the late 1960s introduced Aboriginal people to this exotic philosophy and gave encouragement to unregulated use of the seas and island resources.

By the second decade after settlement, Burarra people from the east had attained a fairly dominant position at Maningrida in terms of jobs and of political power within the village council. However, from the early 1970s the Kunibidji people began to reassert their traditional authority for country, reminding people that Maningrida was not a Balánda (non-Aboriginal) town, nor a Burarra town, nor a Djinang town.

During Armstrong’s time at Maningrida, the Wúrnal man George Yarlmurr was the senior leader for his group and the territory around the two islands. After his death and the later death of his brother, the Wúrnal were without senior adult members. A senior Karddurra man became the principal spokesman for the two groups and to an extent Karddurra became the senior clan in the ‘companyship’. This man died two years ago and the senior Wúrnal, now in their late 30s and 40s, reasserted their primacy for the islands and surrounding waters. 303

A few longterm non-Aboriginal workers at Maningrida have set a better example to new staff by making a point of asking permission to fish or camp in various places and have honored the principle of sharing catch with estate-owners. Although not all non-Aboriginal people show respect in this way, the situation is vastly improved over that through the 70s and early 80s.

It has been somewhat more difficult for the Wúrnal and Karddurra to control pillaging of the egg resource on Ngarráku. Dissatisfaction about egg harvesting and a general lack of respect for traditional owners led to a ‘showdown’ over the issue about 1990. The senior Karddurra man, Billy Yirinyin (Yírrkin, since deceased), and Yarlmurr’s daughter Helen Williams addressed a large crowd of people gathered for the unloading of the Maningrida supply barge and stated their demands as traditional owners, particularly in respect of the egg resource. Helen Williams also later made a video broadcast over the local BRACS system and arranged for a sign to be placed on the island. She said:

I spoke in public to the people who had boats. Next time whoever got boats must come and ask us before they go and collect eggs off the island. That way, we’ll know who is going out there. ‘We want eggs too’, I told them. ‘You guys have got boats and we haven’t. But next time come and ask us’. Long time ago people shared lots. A little bit here, little bit there. And I told people there’s gotta be a limit. Leave some eggs to breed. The reason why it upset me was that when I walked around the camp every afternoon I could see shells. ‘Heh! How many eggs have this mob taken?’, I thought. ‘How come they don’t come and tell us or remind us or bring some to us, this family, landowners?’. That’s how I started to realise some aren’t sharing. Countrymen too, I don’t care.3

304During the 1996 survey, a number of the Aboriginal consultants, in particular Albert Wurrdjal, Jockey Bundubundu and Johnny Nalíba provided previously unrecorded information about the traditional arrangements concerning the egg harvest. Wurrdjal confirmed Helen Williams’ statement that the owning clan should collect the first eggs ‘to open the season’. This would be done in consultation with the senior person to call the island ‘mother’s country’. These principal Aboriginal consultants also described the division of egg gathering rights amongst eight clans, the Wúrnal, Karddúrra and six other clans whose estates abut the Wúrnal/Karddúrra estate. These comprise seven djówanga clans and one yírriddjanga clan. Five are Ndjébbana speaking clans and three are Kunbarlang speaking.

The survey party visited Ngarráku in March 1996 early in the egg laying season. Johnny Nalíba and Jockey Bundubundu walked around the island indicating discrete boundaries between areas of the island where each of these clans had exclusive rights to gather eggs, except for Wúrnal and Karddúrra, who shared one area. The allocation of rights appears to follow proximity to country in most cases, that is, each clan’s egg gathering area was situated on the part of the island closest to the land of their home estate. The exception was one clan whose estate is due west of the island but whose egg gathering rights are on the north-eastern side of the island, but from which the land of the clan’s estate is visible across the low elevation of the island. There are no physical markers marking the boundaries. Yet as Nalíba walked around indicating boundaries, he appeared to be taking bearings on various parts of the mainland before providing an unequivocal position for a boundary.

The Aboriginal consultants say emphatically that the apportionment of egg gathering rights does not carry with it ownership of the land from which the eggs are gathered. The island, the reefs and water surrounding it are all for Wúrnal and Karddúrra. As Albert Wurrdjal says, ‘all that nguya, that just for egg. That island for Wúrnal and Karddurra properly’.

The Wúrnal and Karddurra seem to have slowed the free-for-all pillaging of eggs since they aired their grievances in 1990. A quick survey 305of likely hunters with boats suggests that there were only a few egg gathering expeditions in 1996. The landowners gathered a few hundred eggs during our survey of the island and later in the season a senior Kunibidji man, camped near Rolling Bay, also paid a visit.


Ndjébbana words are written with an accent on the syllable which carries the main word stress. In this paper the authors have attempted to follow the locally established Ndjébbana orthography and are grateful for the advice of linguists Dr Graham McKay and Carolyn Coleman on this and other matters in the paper. Where words are quoted from written text or where an ‘official’ spelling differs from Ndjébbana conventions, the conventional spelling appears in brackets afterwards.


Armstrong, G. 1967. Social change at Maningrida. Unpublished thesis for Diploma in Anthropology. Sydney: University of Sydney.

Bowdler, S. 1993. Sunda and Sahul: a 30 kyr culture area. In Sahul in review (eds) M.A. Smith, M. Spriggs, and B. Fankhauser. Canberra: Australian National University. Pp. 60–70.

Cooke, P. 1996. A sea of meaning: a preliminary survey of Aboriginal interests in seas and submerged lands along the central and western Arnhem Land coast. Unpublished confidential report. Darwin: Northern Land Council.

Elkin, A.P, R.M. and C.H. Berndt 1951. Social organisation of Arnhem Land, 1. Western Arnhem Land. Oceania 21:253–307.

Grant, E.M. 1982. Guide to fishes. Brisbane: Department of Harbours and Marine. 306

Lami Lami, Lazarus 1974. Lami Lami speaks. Sydney: Ure Smith.

Smith, M.A., Spriggs, M. and B. Fankhauser (eds) 1993. Sahul in review. Canberra: Australian National University.

Thomson, D. 1952. Notes on some primitive watercraft in northern Australia. Man 52:1–5.

1 Graham McKay (pers. comm.) points out that Kunibidji is not a Ndjébbana word, but is what others call the people who speak Ndjébbana. However, it is not uncommon for Ndjébbana speakers to refer to themselves as Kunibidji, particularly when talking with non-Aboriginal people.

2 A more recent botanical evaluation identifies these trees at Ndjudda as Diospyros maritima (Carolyn Coleman, pers. comm.).

3 In June 1997 David Bond, who has been visiting Ngarraku for more than 20 years, reported that he had never seen so many young birds on the island as this year. This perhaps indicates that the wishes of the landowners are again being respected, one small but significant victory for culture and conservation.