Aboriginal Shelter

Australia: The Land Where Time Began

Aboriginal Shelter (see Aboriginal Stone Huts, Georgina River, South West Queensland)

In the Simpson Desert the use of wood for the construction of mia mias (humpies or wiltjas) by the Wangkangurru made it a valuable commodity. Mia mias were mainly made from gidgee.

Stone structures

Stone structures of Aboriginal Australia were well known in the 19th century, as can be seen from the ethnographic literature from that time. The known structures were of a number of types, stone-walled windbreaks, stone-walled residential buildings, stone foundations for houses and shelters, hunting hides, food storage buildings, as well as storage sites for sacred objects that must only be seen by fully initiated men.

Engineering works of stone were also known, such as marine and freshwater fish traps, canals, ovens, protective coverings for sacred objects and path liners. There were also stone layouts on the ground of geometric and abstract design that had spiritual significance. See Aboriginal Astronomy. Early references to stone structures were made by a number of authors (Worsnop, 1897; Kenyon, 1930).

An Aboriginal artist made drawings of stone houses still being used by Aboriginal People in the Australian Alps in 1840. In the Warringah area, north of Port Jackson, now the northern suburbs of Sydney, the stone bases of winter houses of the Gai-Marigal people could be seen, and would no doubt have been known to the early settlers of the Sydney area.

Basedow described a house with a stone roof being used by the Yaurawarka people from the lower Cooper Creek, Warburton Creek and the Strzelecki Desert in the northeast of South Australia (Basedow, 1925, in Memmott, 2007). Basedow did not specify the type of stone but the description as slabs make it sound like slate. According to the description it may have been domed on a frame of heavy timber, the gaps between the slabs being plugged with clay.

The Gunditjmarra tribe of western Victoria were observed to be using structures of circular plan form with the entrance to the east.

Stone houses with a roof seem to have been first recorded in the ethnographic literature in march 1842 by George Robinson, Aboriginal Protector. He mentioned a village of the 'Nillen gundidj' clan in the Eumeralla River region , near Bessibelle. He described a number of houses of 2 designs, either stone or 'dirt'. The dirt houses he mentioned are believed to refer to houses with a domed frame covered with a sod cladding. This type of house was known to be in use in this region.

Historical records indicate that about 500 people lived in this village at times during the period of early contact. Robinson stated that the area was comprised of stony rises as well as swamps and thickly wooded areas. A number of circular stone walls were seen by a surveyor, Alex Ingram, as reported by Kenyon, in about 1898, around Mt Eccles and in the south of Lake Condah. An Aboriginal man at the Condah mission station told him they had been roofed like 'an ordinary hut', with boughs and bark. He was told by another man that his grandfather had spoken of 'decaying bark and sapling rooves on stone houses that had been on his property near Louth Swamp.

Among anthropologists and archaeologists up to the 1970s (Memmott, 2007), when the settlement at Lake Condah came to their attention, which eventually led to the Victorian Archaeological Survey 1992, the existence of stone houses wasn't well known. It has since been shown to be a unique region of stone structures. It is in an area of volcanic rocks, draining into Portland Bay, southwestern Victoria. It is open hilly, grassy areas, that are often treeless, with scattered stones and boulders, locally called 'stony rises', streams and swamps in the lower parts. The surrounding woodlands are comprised of stringy bark, manna gum, red gum, yellow box, sheoak and swamp gum. A number of swamps in the basalt lava flows are connected to each other by Darlot's Creek and the Eumeralla River. The lava flow, Tyrendarra Flow, is about 50 km long, from Mt Eccles and Mt Napier in the north, extending to the coast.

The Gunditjmara people's territory covered much of the drainage basin prior to contact, their modern descendants continuing to inhabit the area. This people made use of the plentiful supply of rocks to build stone structures that were integrated with the landscape. The complex of stone structures in the area have been studied by archaeologists for 30 years, but they are still to finalise their work to the point where they can reach a definitive conclusion on the use and purpose of many of the structures in the complex.

There is a degree of agreement between the archaeologists and the living descendants of the Gunditjmarra people that there is a complex system of creeks, ponds, dams, dykes, races, channels, weirs, traps and gates that have been engineered to extend, join and manage the wetlands with the aim of farming the short-finned eel, Anguilla australis. The structures they built were stone, stone-lined and earth. Openings were built into walls to hold conical eel traps made of plaited reeds, each about 3 m long and with a mouth about 600 mm wide, tapering to 100-125 mm. The eel harvest took place in the winter, possibly lasting as long as 9 months, as the result of engineering such a complex structure. At the time of year of the harvest, the area experiences 3-4 months of very cold weather, with intermittent frosts at night. They moved to the coast for the summer months, where they depended on abundant food from the sea and mutton bird chicks and eggs.

The eel harvesting technology sustained a large group of people who lived in sedentary/semi-sedentary huts and villages for up to 9 months each year. Throughout the cold, wet winters they worked at harvesting and smoking eels, but still had time for ritual and social activity, and no doubt enjoyed their summer holidays at the beach.

Information accumulated from Aboriginal, as well as anthropological evidence, suggests that local groups of the tribe each owned an estate, each of which contained a set of eel traps and the associated structural complexes, including a village. These estates were passed on to the descendants of the group owning each estate.

The smoking of eels in hollow trees has been demonstrated by Builth (2002). She has hypothesised they were for export, being traded to other parts of Victoria and as far as South Australia.

Archaeologists use the term intensification to describe the development of a combination of intense exploitation of local resources and semi-sedentary village construction. This economic strategy is usually only applied at the intergroup level in Aboriginal Australia where sizeable gatherings are facilitated.

At a number of sites in western Victoria, such as Lake Condah (or Allambie), Wallacedale, Kinghorn, Darlot's Creek, Stony Rises, Toolonda, Mt Eccles, Mt. Napier, Gorrie Swamp, Lake Purrumbete (the south side), Ettrick, Homertown, Mt William, Lake Bolac and Salt Creek. These are sites where stone structures have been mentioned in the literature, places where it is believed intensification may have occurred.

The interpretations from the archaeological evidence has been disputed by Dr. Annie Clarke, in a paper 'Romancing the Stones', where she presents a systematic critique, suggesting alternative explanations for the circular stone remains. She suggests they may have been foundation walls for timber-domed structures, windbreaks or daytime hunting hides. According to her interpretation of the stone circles they were more likely to be a foundation to hold the base of the sapling ends in place for the construction of a domed structure. In other, less stony areas, the sapling ends would be inserted into the ground. In her paper she warned of 'romancing the stones', seeing the natural structures as being of human construction. Following the publication of her paper archaeologists were more reluctant to claim human engineering in the formation of some structures.

Further evidence for human involvement in the structures have resulted from the Victorian Archaeological Survey of 1992. Dr. Heather Builth (2000, 2002) has also provided evidence for the constructions being man-made. In a 40 Ha area at Darlot's Creek Builth has found at least 25 weirs and channels that had been constructed by the Aboriginal people, as well as the remains of 103 dwelling and storage structures that occurred in clusters, some of which had shared walls. She found that the stones of uniform size had been used in the construction of hypothesised shelter sites. Many of the stones averaged about 200 mm long by 100 mm in width and height. The areas surrounding these areas were found to be depleted in stones of the size used, the stones both smaller and larger being still present. The remains of the simple wall structures were all found to have an opening to the east or northeast, away from the prevailing winds in winter. Among the structures she found was a 1 m high circular wall, 6 courses high, enclosing a space that was 2 m across. The height of the wall decreased towards the entrance, definite evidence of the presence of higher walled structures, as well as the windbreaks that were low.

Builth has identified a number of plan forms among the remains of stone structures:

Circles of various diameters, many with an average diameter of 2.5 m. These circular structures were either isolated or in clusters of 4-5, in some of which was an internal pit that is believed to have been a hearth,
Up to 5 intersecting or interlocking circle segments in a structure of a cellular type
An oval exterior wall with an internal cross partition - a bicellular form
Single circular wall which had a small circular chamber attached near the entrance
A spiral wall enclosing a circular space. This type was known to the living Gunditjmara as a 'number 6' shape. It has been suggested this may have been a way of protecting the interior from wind of changing direction.
A normal circular wall with a smaller diameter circular wall attached at the entrance. It has been proposed that this may have been another way of protecting the interior from winds of changing direction.

See Memmott, 2007

There are other stone structures of shapes that were unlikely to have been used as houses, but their use is unknown at present. Builth has identified 1 such structure that is composed of 2 walls, at most 1 m apart, that were almost 6 m long. There is also a structure of a dumbbell shape, composed of 2 circles, the smaller having an internal diameter of 1 m and an external diameter of 2 m. 2 parallel walls connecting the 2 circles has been described as a narrow race that may have been a corridor.

Land was taken from the Gunditjmara in the 19th century by the colonists and they were eventually moved to the Lake Condah Mission where they remained from 1867-1918. They put up a fight against the invaders, carrying on guerrilla warfare 1834-1854. Many of the tribe now live in the towns of Heywood, Portland and Hamilton. They are keen to preserve and even re-develop the stone culture of their ancestors.

In the meantime, many of the stone structures have been damaged or destroyed, partly by walls being knocked over by cattle and the stones being taken to build walls around paddocks, as well as the squatters own houses. Hence the necessity for extensive archaeological work to gain a true appreciation of what has been lost over the years of destruction and neglect.

The roofing of the walls

As bark sheeting would not have been available in all parts of such a large area covered by the Gunditjmara country where the eel farming was practised, it is not certain what material they would have used to roof their houses.

A few of the methods proposed for roofing the circular walls, are based on methods observed in other parts of Australia, as well as by the squatters of the 1840s on the Gunditjmara land to roof their own houses. These are earth sods, thatched tussock grass and woven reeds with a clay or mud plastering. Sods could have come from many of the swamps in the area that dry out in summer, producing a thick green grass suitable for use as sods. High thick swards of kangaroo grass grew in the lava areas. Just to the northeast and south of the eel harvesting area the use of grass thatching or sods have been reported. Where the camps were located near woodlands, the cladding could be supported by a frame of heavy limbs. It has been suggested a free-standing roof structure could have been made from reeds, if the required limbs were not available in the area. It is believed the same technology used for eel traps may have been used. The ethnographic literature reported these eel traps as being up to 3 m long. Grass of clay could have been used to cover such a framed structure.

Timber-framed domes western Victoria

In the 1840s villages of up to 30 domed winter (wet-weather) huts of a bee hive shape were observed that were waterproofed. A village near Caramut in southwestern Victoria that was observed in about 1840 was the site of the best example. These are illustrated as well as with written descriptions. The residences have been estimated to have been 3.0-3.5 m in diameter and about 2.5-2.8 m high with a 1-m high semi-oval opening of a vertical orientation. At the apex was an opening about 20.0-22.5 cm diameter, to let out the smoke from the fire. It was covered with a sod during rain. The interiors of the domes were on mounds of raised earth. The construction was of limbs with mud cladding. These structures were strong enough to bear the weight of a man, allowing men to climb onto the roof to carry out maintenance. There is some evidence that these villages were used at times of large social gatherings of the people from the surrounding areas, such as were held throughout Aboriginal Australia when rituals and ceremonies were held that involved a number of neighbouring groups, as well as when trade took place. As with other places where large numbers of people gathered for limited periods of time, it was a place where there was plenty of food and water, that would support a large population, at least for a limited length of time (Williams, 1984, 1987; Crichett, 1984). Apparently a stone age convention centre.

At Great Swamp, Konnung-i-yoke, past Lake Linlithogow to the north of Mt Napier at the Grange, believed to be near Hamilton, according to the map, George Robinson saw a number of camps, describing one village of 13 houses. According to Robinson (in Memmott, 2007: 196) :

"The Great Swamp is skirted by low hills and well grassed open forests. The natives are still the undisputed occupants, no white man having been there to dispossess them. The people who occupy the country have fixed residences. At one village were thirteen huts. They are warm and well constructed. In shape of cupola or kraal. A strong frame of wood is first made, and the whole covered with thick turf with the grass inwards. There are several varieties. Those like a kraal are sometimes double, having two entrances; others are semicircular. Some are made with boughs and grass and last are temporary screens. One hut measured 10 feet in diameter and 5 feet high, and sufficiently strong for a man on horseback to ride over."

Another observer writing of the same region, about the Wannon River area in particular, mentioned that sometimes porches were fitted that were often oriented to take advantage of gentle breezes from the northeast.

James Dawson wrote of a standing architectural pattern in western Victoria, people living in base camps over-winter in tall domes with internal partitions. Each dome had a central fire as well as a separate apartment for each nuclear family, with a separate one for the single young women and and a separate one for single young men. Small shelters, not as waterproof, but usually dome-shaped, were used when travelling or during fine weather, and windbreaks were also used, a separate one for each of the above-mentioned groups. The low dome was about 1 m high and at the front was a peaked ridge. Limbs were arranged in a 3/4 circle and bent in to support each other. There was an open wall on 1/4 of the plan.

According to Dawson, "habitations ...wuurns...are of various kinds and are constructed to suit the seasons. The principal one is the permanent family dwelling, which is made of strong limbs of trees stuck up in dome shape, high enough to allow a tall man to stand upright underneath them. Small limbs fill up the intermediate spaces, and these are covered with sheets of bark, thatch, sods and earth till the roof and sides are proof against wind and rain. The doorway is low and generally faces the morning sun or a sheltering rock. The family wuurn is sufficiently large to accommodate a dozen or more persons; and when the family is grown up the wuurn is partitioned off into apartments, each facing the fire in the centre. One of these is appropriated to the parents and children, one to the young unmarried women and widows, and one to the bachelors and widowers. While travelling or occupying temporary habitations each of these parties must erect separate wuurns. When several families lived together, each builds its own wuurn facing one central fire. This fire is not much used for cooking, which is generally done outside. Thus in what appears to be one dwelling, fifty or more persons can be accommodated, when to use the words of the Aboriginal People, they are 'like bees in a hive".

He goes on to describe the houses as comfortable and healthy, saying they are occupied by the owners of the land in the neighbourhood. He said they were built on the edge of a lake, stream or 'healthy swamp', but they avoided places he described as a 'malarial morass' or beneath large trees that could fall or be struck by lightning. Gum trees are also prone to drop branches, even big ones, especially in windy weather. He said that whenever they left their house to travel they closed the doorway with bark sheets or bushes and placed a crooked stick above it pointing in the direction they intended to travel. He also said that as they left they always uttered the words that translate as "close the door and pull away".

According to Dawson even temporary habitations were dome-shaped. He said they were made of limbs, gum tree bark and grass, and not very waterproof, being erected with less care, and were smaller and more open than the permanent huts. These huts were not used for permanent or semi-permanent residences, only in summer or when they were travelling. They had a fire in front of a large opening on one side. A semicircle of small green bushes was often used as a windbreak in warm weather.

Women constructed the smaller shelters, but the men build the permanent shelters, all the men of the group helping. The men kept their smaller weapons with them when they enter the huts, leaving their spears standing on either side of the doorway for easy access if they are required at short notice. He said the huts were made of flat stones when wood and bark was not available, and the roof was made from limbs and thatch. Karn Karn was the name of a stony point on the southern side of a lake near Camperdown, the name translating as' building of stones', though no evidence of a stone building could be seen by Dawson.

Dawson said these buildings were cool in summer and warm in winter, making them better, at least in comfort, for the occupants than the wooden cottages used at the 'Government aboriginal stations'. A fire is kept burning 24 hrs/day in cold weather in the centre of the floor, only a small fire being required to keep the interior of the hut warm. He states that to keep a moderate steady temperature, the ends only of the sticks meet in the centre of the fire, and as they burn slowly away they are pushed inwards. Any other method would be a waste of fuel and would raise too much heat.

'In the event of the habitation being burned down by a bushfire, or occasionally accidentally - which often happens in the absence of the inhabitants - the debris are levelled, and a new wuurn erected on the same spot, which is always preferred...'

Dawson also refers to houses of stone walls roofed with limbs and thatch. .

A watercolour by George French Angas, painted in the 1840s, that is in the South Australian Museum, depicts a structure from Portland Bay in western Victoria. The structure appears to have a high, long-spanned roof, that was probably a permanent building, though the precise details of the structure are no clear. It is believed this type of structure was occupied in winter, a season that in this part of Australia is generally wet and cold. Woven mats can be seen in use.

The caption of the painting is:

'Native Encampment at Portland Bay. 'Cold Morning and his family: In the southwestern portion of the Province, and about the Portland Bay district, the natives build larger and warmer huts than those to the northward: the native name among the Portland Bay tribe for these huts is miam miam. The plate represents one of these encampments inhabited by a native who has assumed the ludicrous name of 'Mr. Cold Morning'. At the period of my sketch he was lying sick upon a round grass mat within the hut, and his wife, with a numerous family of dirty, naked, little 'Cold Mornings' were about him. Spears, shields and baskets were lying about on the roof of the miam miam, which was constructed of boughs of Banksia and eucalyptus, thatched with reeds and dry grass. The scene is in the woods near Portland Bay'. (from Memmott, 207: 198).

Ngalawuru or High Cliffy Island

This island is a member of the Montgomery Island complex, in the Buccaneer Archipelago, to the northwest of the Kimberley coast about 10 km from the mainland. Evidence of stone houses has been found on this island by a post-graduate anthropologist from the University of Wisconsin (Blundell, 1975) while carrying out research in the Kimberleys. The find was made at about the same time as the study at Lake Condah was being undertaken by the Victorian Archaeological Survey. The island was owned by the Yawijibaya people (Montgomery Islanders).

The climate of this region is of a semi-arid monsoonal type similar to that of Arnhem Land and Cape York, temperatures being high throughout the year, reaching as high as 30^oC in the cooler months, the minimum temperatures hardly ever going below 18^o C. At nearby Cockatoo Island the mean annual rainfall is 919 mm. The seasons in this region are mostly the Wet and the Dry, with about 1 month of transitional conditions at the change over. During the Dry, about May to October, the easterlies are the prevailing winds, rain usually not falling between July and September. During November the winds swing around to become westerlies, but fluctuating between easterlies and westerlies until December, when the westerlies become fully established and the Wet season begins, lasting until about March. At this very wet time of year campsite location is determined by the availability of shelter, and there is no shortage of surface water.

In what appears to have been a settlement of relatively high density, there are the remains of hundreds of stone structures. The island, about 15 m above sea level, is about 1 km long and at its widest point about 300 m wide, adding to the surprise of finding a settlement of what must have been many people. It has been suggested that the island was probably used throughout the year, though mostly as a camp during the monsoon season, as there is apparently no permanent water on the island. It would have been relatively free of annoying insects that are common at this time of the year on the mainland. The people are believed to have used rafts for crossing water prior to European contact. Their economy is thought to have been based on the availability of many birds eggs, turtles, dugongs and fish. Quarrying was also thought to have taken place, the products being exported (traded) to the mainland. Evidence has been found of many stone walls of quartz sandstone that are laid out in a stretcher bond, of a rough type, up to 1 m high and up to 50 cm at the base. There was an opening for entry that was narrow, and the interior was about 3 m wide. The remains of fish and turtles, as well as stone artefacts have been excavated inside this structure suggesting it may have been a house (of domiciliary function).

It has been suggested that the circular wall structures were clad with spinifex (O'Connor, 1987, 1999). This and stunted eucalyptus and acacia associations were the only vegetation on the island. The soil is relatively thin on the rocky top of the island, so post holes would have been very shallow, probably too shallow to be of any use.

According to 2 of Blundell's informers, one of whom said they were windbreaks that were covered with paperbark, Melaleuca sp., brought from the mainland on rafts, and spinifex, Triodia sp. Another informer believed they were wet weather shelters, as well as providing a way of hiding fires from view, as the island was prone to attack from parties of Worrorra who raided at night. This informer also believed the lines of stone probably had ceremonial associations. The island has been identified as 'the sacred place of the Yanjbai' (Love n.d., 68).

Photographs suggest that the walls with wide entrances were windbreaks and those with narrow entrances were probably roofed as wet weather shelters, in at least 1 case both types appear to form a complex. Whatever the roofing material used it would require some sort of frame to support it. A requirement for a roof in this location would be to withstand strong wind and heavy rain. A problem on the island was the lack of suitable timber, for both structures and firewood. If it is accepted that paperbark sheets were brought to the island by raft, timber could just as easily have been brought in by raft.

Georgina River

On the Georgina River, about 40 km south of Boulia, near the airstrip of Marion Downs Station (pastoral property), a village of 17 circular, or curvilinear enclosed, stone houses have been found. It appears each circular structure had an opening less than 1 m across on one side, and facing north. There are also a number of structures with a rectangular floor plan. The area of the internal space averaged 7.1 m².The structures were studied by Dr. Ian Davidson in 1989. He assumed they were huts. One had a collapsed group of gidgee boughs, that may have been either over it or above it, which suggests it probably had a roof structure. It is not known if they were being used prior to European contact.

Stone walled game hides

These hides or blinds used for hunting have been described from the Western Desert and Central Australia. They usually consisted of a section of dry-stone wall structure that often abutted a natural feature of the landscape, often near a waterhole. The area inside the hide, about 3 m wide, was usually cleared of stones to allow the hunters to sit comfortably as they waited. The rocks from the cleared area were used to build the walls of the structure, about 1.2 m high. These hides are mostly found in rocky high ground in arid sand country near waterholes where water is held for long periods because of the impervious rocks and some degree of protection from the sun. The animals hunted by this method were mostly kangaroos, wallabies and emus that came from the waterless sand country to drink. Such structures have been found near Yuendumu that were used by the Warlpiri. They were also found at Ampilatwaj, used by the Alyawarr on the Sandstone River in Central Australia.

According to Memmott, the best description in the ethnographic literature was written by Richard Gould (1968, 1977), an ethno-archaeologist working with the people of a dialect subgroup of the Pitjantjatjara language group, the Ngatatjara, from the Warburton and Rawlinson Ranges area in the Western Desert. He described a rock wall hide at Mularpayi, a part of the Barrow Range situated between the point of intersection of the borders of Western Australia, the Northern Territory and South Australia, and the Warburton community.

'The principal feature of the site is a [57 metre] long gorge with unusually regular rock walls ranging from [1.8 to 4.5 metres] high. The floor of the gorge slopes downwards to the entrance, while at the upper end there is a small but deep pool of freshwater. In a few places along the side walls of the gorge there are natural crevices large enough for a man to hide inside. The gorge is renowned among the Ngatatjara as a natural game trap and its use reflects similar practices at the 13 other sites visited during the survey ... on the northeast rim of the gorge, [10.8 metres] from the waterhole is a circular hunting blind constructed of local rocks piled neatly on one side in a row, with natural rocks comprising the back and sides. The blind is [1.8 metres] in diameter, the rocks had been cleared from the interior and used in the wall construction. The earth floor of the blind contained a few pieces of butchered bone and some stone flakes. My Ngatatjara guides explained that the hunters crouched in the hides at night, waiting to spear kangaroos and emus as they approached the water. Once inside the gorge there was no escape except by means of the entrance. This route was usually blocked by additional hunters ..'

Gould also described rock art, both secular and ritual, in the vicinity of the hide. There were small openings in the sides of the hides so the hunters could watch the game. They could also throw their spears through these openings, over the walls or around the outside of the hides. There were also bird hides from which the hunters catch birds through a opening in the roof.

Stone-walled bird hides

Darrell Lewis reported circular stone walls covered with a cladding of grass and spinifex on timber rails being used by bird hunters in the basin of the Upper Victoria River in the Northern Territory. They used a dead pigeon impaled on a stick that they poked through the cladding of the roof, attracting certain species of hawks and eagles to the bait by lighting a fire. When a bird lands to eat the pigeon the hunter grabs its legs and pulls it through the covering and kills it, then pokes the pigeon through the roof again to attract more birds. George Augustus Robinson described birds being caught by the same method at Macquarie Harbour in Tasmania in 1833, the the birds being hunted were crows. In Tasmania the hides were made of pliable saplings. The stone-walled hides have also been reported from Esmeralda Station near Croydon, to the southeast of the gulf of Carpentaria. These hides were similar to those structures being used as residential shelters in western Victoria and on High Cliffy Island, though they were smaller.

Covers protecting sacred objects

In Central Australia various tribes have retained their beliefs in ancestral Beings from the Dreamtime (Agterre) and creation myths of their ancestors. They believed that many sacred objects still contain the powers imbued in them by the ancestral Beings. These objects were often placed in clefts and holes in the rocks of the landscape, and camouflaged with layers of stones. One of these sites is known to also have had a stone wall built around it. These sacred objects were taken out only when they were to be used in sacred ceremonies and rituals connected with the Dreamtime Being who left them for the people.

A sacred site, Lyabe, in the western MacDonnell Ranges in the Northern Territory, is believed by the Arrente to have been created by Honey Ant Beings, Yerrampe. The atywerrenge are sacred objects they left in a depression in a rocky part of the range. Surrounding this site are upright stone slabs that are said to be the transformed Ancestral Honey Ant Beings that were turned to stone at the close of the Dreamtime. A horseshoe-shaped stone wall surrounds the pit-like depression. The wall, open at the western end, is about 1 m high and at the base is at least 1 m thick. The outer diameter of the wall is about 5 m, the 2 m-wide storage pit continues below the wall base. Honey Ant increase rituals were held at this site by the Arrente. The secretive nature of such sacred sites prevents more detail being provided (Memmott, 2007: 204).

The evidence for stone structures

According to Memmott (2007: 204) Worsnop was the first ethnographer to attempt a summary of the evidence for stone structures, including the lava flows region of western Victoria, in Arnhem Land and the Prince Regent River basin and stone arrangements for ceremonial purposes and the use of low walls.

The 2 examples of stone villages of a sedentary use, one in the Kimberleys and one in western Victoria, both used at times when resources were in plentiful supply, and on a regular seasonal basis. They reflect the process of socio-economic intensification. Evidence indicates they were used pre-contact, as well as for some time after contact. There is a high level of similarity between the structures on High Cliffy Island and the Lake Condah complexes of stone structures, as noted by O'Connor, stating 'these use the same walling techniques, are of approximately the same size and are clustered in the same way ... [and] are located immediately adjacent to a resource rich zone ...'. She noted that at Lake Condah fish traps were constructed , while the reef and rock pools at High Cliffy Island formed natural fish traps. The rock type differed between the 2 sites, those on High Cliffy Island being fractured, flat slabs of sedimentary rocks. At Lake Condah, the rounded, igneous rocks were of less regular shape. They used comparable architectural styles at both sites, adapting it to allow for the rocks available.

The structure of any roofing present has still to be determined, at least to the satisfaction of the scientific community, though there are a number of references to roofed walls appearing in the ethnographic literature. These often clearly describe huts with circular stone walls with roofs of various materials.

Whale bone house structures

Among many groups of coastal Aboriginal people whales were important in their economy and culture, particularly in the south and southeast of Australia. An example was the people of the Nullarbor Plain and the people of the coastline of the northern section of the Great Australian Bight. They identified with the Whale Dreaming (Burgoyne, 2000). Whale bones are known to have been used in the construction of coastal houses, as depicted in a painting by W. A. Cawthorne (1842) in which 2 whale bones are employed as a structural frame for an Aboriginal shelter. Around the time Cawthorne painted the scene he was writing a manuscript about the local Aboriginal People. In 1847 George Angas published a book, South Australian Illustrated, in which he included copies of Cawthorne's painting and his subjects, re-working them in his own style. The Caption included for his copy of Cawthorne's painting :

'Natives of Encounter Bay. The view here given represents a part of the shores of Encounter Bay with a native hut formed of the ribs of a whale. Numerous carcasses of whales being cast upon the shores adjoining the fisheries, most of the native huts are constructed with a framework of bones, the interstices being filled inn with boughs and dried grass. The present group consist of a man called Ginnginnana, and his two wives, Kundarkey and Wuddagar, Kundarkey is chewing reeds, while her husband is twisting the chewed strips into a cord upon his thigh, for the purpose of being made into baskets and fishing nets, one of which lies on top of the hut.'

According to Memmott the location of the hut is suggested by the caption to be adjacent to an early European whaling station. In Tasmania there is evidence of the use of whale bones in construction of shelters in pre-contact times. In the 1820s, Jorgen Jorgenson described a house at Venable's Corner, on the northwest coast of Tasmania, that is a sheltered bay:

'On the following day they reached the Venable's boat harbour. On passing along they observed a very neat and compact native hut. It bore all the marks of simple rudiments of Gothic architecture; it rose in the shape of an oblong dome, and might easily contain from sixteen to twenty persons. The wood used for the principal supports was bent into a curve, and seems to have been rendered hard by fire. It was uncommonly neatly thatched, and the door-way was about two and a half feet high. Necessity is the mother of invention, and therefore the Aboriginal People on this coast have been compelled to construct compact huts to screen from the inclemency of the cold, and the boisterous winds, especially where fuel is so scarce as it is here ...'.

Also in Tasmania, John F. Jones discovered a hillside camp site at Bluff Point on the same stretch of west coast as Sandy Cape. He visited the site in 1930 and 1945, finding on his second visit that the wind had eroded the site to reveal whalebone structures. He compared his findings with those described by Jorgenson:

'On the top I noticed, standing erect at the edge of one of the hollows, a portion of a rib of a large whale. On pulling it out of the sand I found it to be about [600 millimetres] long. The portion that had been exposed to the air had been decayed and the end that had been forced into the ground had been cut on two sides by a chopper by the natives. It was evidently part of one of the roof supports of the hut. Not far off I found a complete rib which showed little sign of decay having probably been uncovered more recently. It appeared that in one, at least of the ten huts that had been built on the hillock a whale's rib had been used for the support. I tried to realise the effect, I suddenly realised how Jorgenson's Gothic vault and dome might be quite comprehensible. The number of ribs with one end embedded in the sand at intervals around the hollow, with the other end gathered and fastened together over the middle of the hut would, when completed with light material and thatched, make a very perfect dome. Inside the dim light and blackened by smoke, the bones might easily be mistaken as wood while their smoothness and evenness of the curve would give the impressions of timber that had been artificially shaped (Jones, 1946 in Memmott (2007: 207).

Dugong bone grave site at Stewart River

A grave site was recorded near the mouth of the Stewart River, on the east side of Cape York, in 1928-1929 by Donald Thomson, an anthropologist. It was constructed from the bones of 8 dugongs, and was also decorated with their bones.

'The vertebrae and small bones were piled up on the centre of the grave and were completely encircled by the ribs arranged with vertebral (thick) ends inwards'. At the head of the grave the dugong heads were placed facing towards the sea.

Spinifex houses of the Western Desert

In pre-contact times the people of the Western Deseret built their domed huts with timber frames with a cladding of spinifex (hummock grass) Triodia spp. The method used for collecting spinifex for cladding has been described 'squatting on the ground and leaning backwards, the weight of their bodied supported on their hands, they would thrust their feet forward, and using the tough callused skin of their heels, they would gouge the clumps out bodily complete with the roots'.

According to Memmott the settlement patterns of the Aboriginal people of the western Desert were integrated with their social, economic and religious organisation.

Professor Bob Tonkinson has said 'Extending over a million square miles, the Western Desert ... covers a vast area of the interior of the continent. It extends across western South Australia into central and central-northern Western Australia (south of the Kimberleys) and south-western Northern Territory and it includes most of the hill country in northern South Australia. The area is marked by an overall similarity in both climatic conditions ... and physiographical characteristics. More important, however, is its delimitation as a distinct cultural area ... its Aboriginal inhabitants share a common language (with dialectical variations), social organisation, relationship to the natural environment, religion and mythology and aesthetic expression. The term Western Desert, then, refers to both a cultural bloc and a geographical entity.

Features of the cultures of the various groups of Western desert people that are shared or similar cover a wide range of activities such as camp layout, shelter construction and materials, there is some variation between the different groups.

In the part of the Western Desert comprising the northern section - the Gibson Desert, the Great Sandy Desert and the Little Sandy Desert, the Aboriginal inhabitants refer to themselves as Martu. These people now live in places such as Newman, Jigalong, Parnngurr, and on the Old Canning Stock Route, Kunawarraji. In the south and east, comprising the Great Victoria Desert and the Nullarbor Plain, they refer to themselves broadly as Anangu. This term includes people such as the Pitjantjatjara, Yankuntjatjara and other related dialects. These people mainly live in such places as Amata, Pipalyatjara, Fregon and Iwantja (Indulkana). The people identifying themselves as Western Desert people include the Kukatja, the Pintupi, and the Ngalia subgroup of the Warlpiri, as well as several other groups, inhabit the north-east part of the Western Desert. The northernmost parts are occupied by the Walmatjarri, and in the southwest are groups such as Ngaanyatjarra, Mandjindja and Nyanganyatjara. They live in places such as Warburton, Wiluna, Kalgoorlie and Laverton (Memmott, 2007: 210).

The Giles weather station near the Rawlinson Ranges provides data that gives an indication of the range of temperatures that can be experienced in the Western Desert, that can range from below freezing in winter to occasional summer maximum temperatures of 46^oC. The peoples of the area wore no clothes at any time of the year, cold or hot. In winter the only they avoided the coldest times was with fire and shelters.

Very limited study of the ethno-architecture and camps of the area has been undertaken. The only study was apparently undertaken by an architect, Peter Hamilton at Mimili in the Everard Ranges in 1970-1971. He camped with them for more than 16 months. The Mimili settlement was on a flat, open area near a water bore. The surrounding country was rock outcrops in a large area of sand ridges. The shelters, called Wiltja, numbered 19. The layout of the village was cantered on a public space, around this common the village was divided into a number of living spaces, 'niches', each about 18.5-22 m². These were of a segmented design. Gender specific rituals were carried out in peripheral places around the edge of each living space. These places were also used for defecation. Each Wiltja, with its cleared space, were moved around a central point that was normally unchanging. A number of reasons were given for such moves, such as the materials had to be physically replaced, as with structural components that had broken, and disintegrating cladding, as well as a response to sickness.

Couples with children occupied 10 of the wiltjas. 6 were for childless couples. 1 had 2 widows, 1 contained a widower and 1 housed 5 unmarried adolescent men. The number in the settlement varied up to 84 during the period of study - individuals or families leaving for varying lengths of time from days to months. Hunting and rituals were reasons for leaving, as well as visiting relatives who had moved to the Indulkana settlement set up by the government, that was nearby. Fregon, Ernabella and Alice Springs were also places where relatives had moved to. The people were no longer living the free ranging traditional life, receiving social security payments and doing occasional stock work on the cattle stations. Tea, flour, sugar and tinned food had entered their diet. No traditionally living groups were studied in connection with their shelters.

Models of camp sites and architecture have needed to be pieced together from references in the literature from widely separated places throughout the Western Deseret.

Because of the hostile nature of the environment of the Western Desert, the early contacts with Europeans were very limited.

Sources & Further reading

Shephard, Mark, 1992, The Simpson Desert: Natural History and Human Endeavour, Reed
Memmott, Paul, 2007, Gunyah, Goondie + Wurley: The Aboriginal Architecture of Australia, University of Queensland Press

Links