The Gardens of the Dreaming

TL;DR

• For fifty years the comparison of Australia and New Guinea turned on one question: why did Aboriginal people never farm, when their neighbours across the Torres Strait did?
• Gosden and Head (1999) answered that subsistence was the wrong thing to compare. Compare attachment to land, and the two shores share a history: the “ritual tracks of the Huli are extremely similar to the Dreaming tracks”.
• The yam proves them right. On both shores an ancestor made the food at a place, his descendants keep the place, and a rite makes the food come up: by singing in central Australia, by copulation and sperm on the New Guinea coast. Farming is a difference of kitchen. The country is one.

It is strange that these people have never learnt to cultivate the earth and build houses. — Captain John Moresby, of the Aborigines of Cape York, 1876

Why did Australians never farm? #

Moresby sailed through the Torres Strait in the 1870s and saw two worlds. On the islands, people built “good huts” and kept gardens of yam, taro and banana. On the mainland just to the south people dug wild yams and moved on. A century later the archaeologist Peter White put Moresby’s remark at the head of a paper and gave the puzzle a name: “New Guinea and Australian prehistory: the ‘Neolithic problem’” (White 1971). His question was not why Aboriginal people failed to invent agriculture. It was why they did not take it from neighbours who had it.

White saw that the Strait was a slope, not a wall. The Meriam of the eastern islands “were gardeners in the Papuan manner”, gardens thinned to the west, and “there is clearly no sharp break between Papuan agriculturalists and Australian hunter-gatherers.” Then he reached the mainland: “With Cape York, however, the contrast becomes complete.”

That sentence set the terms for a generation. New Guinea had farmed since the early Holocene: at Kuk Swamp in the highlands, people were planting and draining a wetland about 10,000 years ago (Denham et al. 2003). Australia had not. Every comparison of the two began from that line and asked what had held Australia back. The field’s habit of looking away from New Guinea altogether is its own story, told in Australian Archaeologists Can’t Handle the Truth. This essay is about what you see when you stop asking White’s question.

What did Gosden and Head propose? #

In 1999 the archaeologists Chris Gosden and Lesley Head wrote a chapter with a question for a title: “Different histories: a common inheritance for Papua New Guinea and Australia?” Their target was the frame itself. “Central to the division of Australia from its northern neighbour,” they wrote, “is the notion that it is a continent of hunter-gatherers next to an island of farmers.” The comparison “has not been open-ended, but has centred round one question: why did Aboriginal groups not adopt agriculture when they must have known about it through contacts with agriculturalists?”

Their answer was to stop treating food as the master key. “The production of food is not the motor for history,” they wrote, and, more bluntly, “There is, then, no such thing as subsistence.” Food is one strand in a whole way of living with land, and it cannot be pulled out and ranked above the others. Gosden’s introduction to the volume put the argument in two sentences. If subsistence is the main point of comparison, “it is only possible to emphasize differences between Australia as a continent of hunter-gatherers and Papua New Guinea, a country of farmers. However, if different elements of attachment to the landscape are emphasized then more of the shared history of the two areas comes into focus.”

They also did the arithmetic the old frame ignored. People had lived in Sahul for at least 40,000 years, and the land bridge drowned only about 8,000 years ago, so “five-sixths of the period of the settlement of the continent occurred while the two areas were joined.” That “raises the possibility of long-term historical continuities which still have some influence today, ancient inheritances being reworked to suit local exigencies.”

Then they named one. Among the Huli of the Papua New Guinea highlands, ritual and trade follow lines of power called dindi pongone, “the tracks of ancestral spirits who created the present-day features of the landscape”. Rites must be performed at the right places along them. Across the Arafura Sea, the Kunwinjku of Arnhem Land walk the Dreaming tracks of the ancestors who made their country, and, citing the economist Jon Altman, Gosden and Head note that they “make no differentiation between the ceremonial system and the economic system of hunting and gathering.” The Huli raise pigs and sweet potato in ditched gardens. The Kunwinjku hunt. Yet “the ritual tracks of the Huli are extremely similar to the Dreaming tracks of the Gungwingu, as are the complex of ceremonies surrounding them.” Their conclusion: “The obvious example of a possible form of ancient inheritance is the existence of Dreaming tracks in New Guinea, as well as Australia.”

They ended with a question nobody followed up. Is it an accident that “a stress on Dreaming tracks is most pronounced to the north and south of the Arafura Sea, the area which saw most environmental change due to rising sea levels between 15,000 and 8000 BP”? Were the tracks “some means of maintaining continuity in an otherwise fast-changing world?” That is the coast where the flood stories cluster, as Myths That Remember Sahul maps them.

Gosden and Head offered a frame and two examples. They did not test it on the shore where it matters most, the Torres Strait itself. The rest of this essay does.

Was the line ever real? #

Even on its own terms the farmer–forager line was blurrier than White’s sentence. Aboriginal people did not keep gardens, but they did a great deal to make plants grow. A. C. Haddon, who knew the Torres Strait better than anyone of his time, noted a century ago that in Western Australia “the natives invariably re-insert the head of the wild yams they have dug up so as to be sure of a future crop.” A recent survey of the ethnography finds that “the re-planting of sections of tubers (including yams, lilies, sedges, legumes, ground orchids and numerous other plant groups) after harvesting is documented throughout Australia” (Silcock 2018), from Cape York and Arnhem Land to the Kimberley and the south-west. On Cape York, Hynes and Chase argued that some plant communities “may have been not merely modified but created by Aboriginal cultural activity”, and coined the word domiculture for it (Hynes and Chase 1982). Rhys Jones called the burning of country “fire-stick farming” (Jones 1969).

The New Guinea side was not all gardens either. The Sengseng of New Britain, one of Gosden and Head’s cases, “cultivate gardens, but gain 60 per cent of their nutritional requirements from wild animals and plants.” David Harris, who studied the Torres Strait islands in the 1970s, “very explicitly argued for a continuum, in subsistence terms”, as Gosden and Head put it, while keeping the two categories. Lesley Head and her colleagues later traced plant use across all of Sahul as one generalist pattern that only split into regional specialisms “primarily from the mid-Holocene” (Denham, Fullagar and Head 2009), after the sea had come in. White himself, revisiting his paper in 2011, said he had “unquestioningly accepted” the contrast.

So the line is a gradient. But a gradient is still a comparison of kitchens. Gosden and Head’s point was to compare something else.

Who made the land? #

Start with the ancestors.

In central Australia, Baldwin Spencer and Francis Gillen found that the Arrernte country was divided into “localities, each of which is owned and inhabited by a local group of individuals, and each such locality is identified with some particular totem.” The ancestors of the Alcheringa, the Dreaming, “are associated with special localities in which they became changed into spirit individuals” (Spencer and Gillen 1899, pp. 277–278). The ancestor walked, made the country, went into the ground at a place, and the people of that place descend from him and are named for his species.

Now cross to the south coast of New Guinea, to the Marind-anim, whose country runs from Merauke to the Bulaka River, just west of the Torres Strait. Their ancestors are the dema. Jan van Baal, who wrote the great book on them, defines dema as beings of “the mythical era” who “are the ancestors of the clans and subclans and are associated with their totems. Sometimes they created the totem, e.g. Wokabu’s wife, whose excrement became the sago, or changed into it, such as Yawi, out of whose head the coconut grew” (van Baal 1966, p. 179). “The link between the clan and its totems is the déma: the clan-ancestor who procreated, made or changed the natural objects which are the clan’s totems” (p. 191). When the dema’s work was done they went into the ground, and where they lie, “the totem species associated with the déma abounds in the neighbourhood.” The missionary Verschueren reported that a Marind man “would never give up these places, not for all the world… He needs them for his magic” (p. 183).

That is the Dreaming, under another name. The early ethnographer Paul Wirz, who had read the Arrernte material, saw it at once: “In fact there is great agreement between the totem rites of the Australians and the Marind-anim,” and the Marind origin myth was so like the Arrernte one “that one might think of a common origin” (Wirz 1922, pp. 40, 130, my translation). Van Baal himself, in 1963, found the main likeness in “the totemism of the Marind-anim and the Orokolo on the one hand and that of the Australian tribes on the other”. A. C. van der Leeden then set the Marind beside the Nunggubuyu of south-east Arnhem Land and found the same four-part order of moieties and clans on both sides of the Arafura Sea, and the same view of land: “Territorial names refer primarily to particular sites, which are considered as the creations of mythical ancestors” (van der Leeden 1975, pp. 151, 155). This site’s earlier survey covers those comparisons.

The relation to the ancestor is not reverent at a distance. It is kinship, and it is eaten. A Marind man calls his totem species amai, grandfather. The Dutch missionary Vertenten heard men at a meal exclaim, “Amai, her ah… oh!, i.e. Grandfather, you taste delicious” (van Baal 1966, p. 192).

How do you grow a yam? #

Here the comparison of land becomes a comparison of food after all, and the line disappears.

Download the data (CSV)

In Australia the rite that makes food abound is the increase ceremony. Spencer and Gillen defined the Arrernte intichiuma as “sacred ceremonies associated with the totems, and the object of which is to secure the increase of the animal or plant which gives its name to the totem” (1899, p. 167). Each totem group does it for the others, so that “taking the tribe as a whole, the object of these ceremonies is that of increasing the total food-supply” (p. 169). For a flowering hakea, the men sang at a stone that was the ancestor’s tree, inviting it to bloom and the blossoms to fill with honey. Then a young man opened a vein in his arm and let “the blood sprinkle freely over the stone” (pp. 184–185).

Among the Warramunga and Worgaia, north of Arrernte country, the yam people “arose… close to where is now Tennant Creek. They were sent away to the east by the great ancestor of the black-snake totem, taking their yams with them.” Their women, the Munga-munga, “walked away towards the east, dropping yams as they did so… two or three of them going further on still in the direction of the Gulf of Carpentaria.” The yams they dropped turned to stone churinga (Spencer and Gillen 1904, pp. 274–275, 303–304). Their descendant, the head of the yam totem, keeps those stones. When the yams are wanted, the other moiety asks him to “go and walk about in the bush and ‘sing’ the yams, as they want them to grow.” He goes out every day for about two weeks with a churinga under his armpit. When the plants are up, he bites a small yam and throws the pieces in all directions. Then: “I have made plenty of yams for you to eat” (pp. 296–297).

Cross the Strait. On the Marind coast the dema Aramemb, at a Mayo feast “so exceedingly grand and licentious… as no mayo celebration had ever been before”, “anointed his body with sperma instead of coconut oil, whereupon yams of all known varieties grew from his head and shoulders. He planted them on a large field east of the Torassi” (van Baal 1966, p. 284). The Torassi is the river that marks the border. Aramemb’s yam field lies in the Trans-Fly, on the Papua New Guinea side, facing Cape York.

The Marind did not only tell this. “Human sperma is widely applied in fertility magic,” van Baal writes. “The sperma of a Geb-zé has a special effect on coconut trees, which are a Geb-zé totem; that of a Mahu-zé is particularly favourable to a sago palm” (p. 194). In the fertility rites “a number of women have intercourse with as many men as possible, while the excreta are collected in a coconut-bowl,” and the sperm is put on the buds of the plants (p. 548). The clan’s seed feeds the clan’s totem. At the climax of the Mayo rites, Wirz was told, the initiates raped a girl and then killed and ate her; “a bone is buried of the iwag who was eaten” beside each newly planted coconut palm, “and the trunks of the palms are painted red with her blood” (p. 543).¹ The girl of the Mayo, the mayo-iwag, shares her name with “a small, brightly coloured snake”, and a grass-snake was buried around the ritual ground (p. 541).

In the Torres Strait the culture hero Sida came out of New Guinea, “from Pab, on the mainland of New Guinea beyond Boigu”, loaded with “coco-nuts, taro, yams, sweet potatoes, sago”, and sowed the islands as he went. On Mer the people gave him a girl, Peker. On the beach “he try to do bad.” “No,” she cried, “you must not do it that way.” “So they turned round ‘and do bad,’ and immediately coco-nuts and every kind of vegetable food and various fruits sprang up” (Haddon 1904, pp. 28–29). Haddon noted that “the association of the sexual act with agricultural fertility… is implied in all the versions of the story of Sida” (p. 347). When he had sex with the virgin Zabaker at Warwe on Mer, “the blood flowed from her in great quantity; then he erected a shrine” on the spot (Haddon 1908, p. 20).

The Islanders’ clans had their own increase rites. When the first turtle of the season came in, the men of the turtle clan “danced round the turtle whirling bull-roarers” while a length of creeper “was inserted in the cloaca of the turtle and pushed up and down several times… to ‘make him (that is, all the turtle) proper fast’.” Haddon’s footnote: “‘Fast,’ in the jargon English that is spoken by the islanders, means the act of copulation” (1904, pp. 183–184). In Mabuiag gardens the wooden guardians, the madub, came alive at night and “went round the garden swinging the bull-roarers to make the plants in the garden grow” (p. 346). At the Fly mouth the Kiwai bullroarer called Burumamaramu was swung for the boys “when the yams are ready for digging”; its name, Haddon explains, means “the mother of yams” (p. 219). Another Kiwai blade announced itself: “My name is madubu. I am the father and mother of all yams,” as A New Word for I tells.

Put the three shores side by side. The ancestor carries the food over the land and leaves it in the ground at a place. Her descendants own the place. A rite at the place, done by the people who share the ancestor’s substance, makes the food come up: the yam man’s song and his bitten yam, the Marind clan’s sperm on its own palms, the turtle clan’s creeper and its roaring blades. Haddon, looking at the yam shrines of Mer, drew the same conclusion: they were “originally… rituals connected with the increase or control of totems by the elders of the respective clans” (1908, p. 255).

A garden is a place where an ancestor’s food is made to grow by the people who belong to him. So is an increase site. The Mer islander and the Worgaia yam man were doing the same thing. One of them also dug the soil.

What does a garden change? #

Not nothing. Gosden and Head are careful here. The Huli keep their ancestral tracks “in tension with an emphasis on exchange of pigs and other valuables as a route to personal prestige quite antithetical to an Aboriginal disdain for material wealth.” Gardens feed pigs, pigs make big men, and big men build a politics Aboriginal Australia did not have. Van Baal, for his part, thought Marind gardening poor and the Mayo rites not mainly about crops (pp. 543–544).

But the difference sits on top of the shared thing, not under it. The garden on Mer is a new kitchen in an old country. The ancestor still made the food, the clan still keeps the place, and the rite is still where the food comes from. That is why the Islanders could take up yams and taro from New Guinea without changing their religion, and why Cape York people could watch them do it for centuries without needing to. They already had a way to make yams grow.

White asked why agriculture did not cross the Strait. The better question is what crossed it long before, so that gardening could arrive on one shore as a small addition and be refused on the other as unnecessary. The answer is the relationship itself: ancestors in the ground, clans named for what they left, and the rites that call it up.

How old is the shared country? #

Old enough to be one country. The last land bridge drowned between about 9,000 and 7,500 years ago (Old Sahul), after people had already been planting at Kuk. For most of human history, the yam ancestors of Tennant Creek and the dema of the Bulaka River walked one continent, and the Munga-munga women were walking toward its middle. The archaeologist Michael Rowland, surveying the contact evidence on the Queensland coast, made the same inference from the other end: the peoples of Sahul “likely shared some aspects of a common history over a period of perhaps as much as 50,000 years”, and “it would seem unlikely that this shared community of culture and ideas would have ended abruptly with the rise in sea level” (Rowland 2018).

The rest of the shared country is told elsewhere on these sites, and it lines up. The same men’s cult with its bullroarer, stolen from women and swallowing the boys, runs from the Sepik to Arnhem Land (One Sahul, Nggwal). The languages of Sahul name tree, firewood and fire as one family of things, as the people who burned and replanted their country might, and on the coast facing Saibai they still say “I” the Australian way (The First Person of Sahul). The bullroarer that eats the boys in Australia grows the yams in the Strait. It is one inheritance seen from different sides.

Gosden and Head asked for “tales of the unexpected” in place of the old history of farmers and foragers. Here is one. On both shores of the drowned bridge, the food grew where the ancestors had lain down, and people made it come up by giving the ancestor back what was his: blood on the stone, seed on the palm, a song walked through the bush. Only one shore also picked up a digging stick.

Sources #

1. Baal, J. van. Dema: Description and Analysis of Marind-anim Culture (South New Guinea). The Hague: Nijhoff, 1966. PDF.
2. Baal, J. van. “The cult of the bull-roarer in Australia and Southern New Guinea.” Bijdragen tot de Taal-, Land- en Volkenkunde 119 (1963): 201–214. doi:10.1163/22134379-90002136. Quoted via van der Leeden 1975.
3. Denham, Tim P., et al. “Origins of agriculture at Kuk Swamp in the highlands of New Guinea.” Science 301 (2003): 189–193.
4. Denham, Tim, Richard Fullagar, and Lesley Head. “Plant exploitation on Sahul: from colonisation to the emergence of regional specialisation during the Holocene.” Quaternary International 202 (2009): 29–40.
5. Gosden, Chris. “Introduction.” In Gosden and Hather (eds), The Prehistory of Food: Appetites for Change. London: Routledge, 1999.
6. Gosden, Chris, and Lesley Head. “Different histories: a common inheritance for Papua New Guinea and Australia?” In Gosden and Hather (eds), The Prehistory of Food: Appetites for Change, 227–245. London: Routledge, 1999.
7. Haddon, A. C. (ed.). Reports of the Cambridge Anthropological Expedition to Torres Straits, vol. 5: Sociology, Magic and Religion of the Western Islanders. Cambridge University Press, 1904.
8. Haddon, A. C. (ed.). Reports of the Cambridge Anthropological Expedition to Torres Straits, vol. 6: Sociology, Magic and Religion of the Eastern Islanders. Cambridge University Press, 1908.
9. Haddon, A. C. The Races of Man and Their Distribution. Cambridge University Press, 1924, p. 22. Quoted in Gosden and Head 1999, p. 233.
10. Hynes, R. A., and A. K. Chase. “Plants, sites and domiculture: Aboriginal influence upon plant communities in Cape York Peninsula.” Archaeology in Oceania 17 (1982): 38–50.
11. Jones, Rhys. “Fire-stick farming.” Australian Natural History 16 (1969): 224–228. Reprinted in Fire Ecology 8(3) (2012): 3–8.
12. Leeden, A. C. van der. “Nunggubuyu Aboriginals and Marind-anim: preliminary comparisons between south-eastern Arnhem Land and southern New Guinea.” In van Beek and Scherer (eds), Explorations in the Anthropology of Religion: Essays in Honour of Jan van Baal, 149–157. The Hague: Nijhoff, 1975.
13. Rowland, Michael J. “65,000 years of isolation in Aboriginal Australia or continuity and external contacts? An assessment of the evidence with an emphasis on the Queensland coast.” Journal of the Anthropological Society of South Australia 42 (2018): 211–240. Abstract.
14. Silcock, Jennifer L. “Aboriginal translocations: the intentional propagation and dispersal of plants in Aboriginal Australia.” Journal of Ethnobiology 38 (2018): 390ff.
15. Spencer, Baldwin, and F. J. Gillen. The Native Tribes of Central Australia. London: Macmillan, 1899.
16. Spencer, Baldwin, and F. J. Gillen. The Northern Tribes of Central Australia. London: Macmillan, 1904.
17. White, J. Peter. “New Guinea and Australian prehistory: the ‘Neolithic problem’.” In Mulvaney and Golson (eds), Aboriginal Man and Environment in Australia, 182–195. Canberra: ANU Press, 1971.
18. White, J. Peter. “Revisiting the ‘Neolithic problem’ in Australia.” Records of the Western Australian Museum Supplement 79 (2011): 86–92.
19. Wirz, Paul. Religion und Mythus der Marind-anim von holländisch Süd-Neu-Guinea und deren totemistisch-soziale Gliederung. Dissertation, Basel, 1922.