Terminal - Transition

 

The first clear archaeological evidence for human occupation of the South West Coast of Victoria is from after the Meltwater Pulse (MWP 14,600-14,300 ybp), an event which marked the end of the long Last Glacial Maximum (LGM peak 22,000-19,000 ybp).

The general post-MWP period is known as the Terminal Pleistocene, the end of the Pleistocene epoch, and as the Holocene Transition, the start of the Holocene epoch. While technically the Pleistocene finishes and the Holocene begins at 10,000 years ago, the Terminal-Transition stretches from 13,000 to 8,000 ybp. It is a broad timespan of prehistory covering the millennia of recovery, when the climate became wetter and warmer. In turn, plants, animals, and humans, all looked to expand their territory.

There is an increase in human activity during this time, with new sites appearing in new areas. Prior to the LGM, the Victorian SW Coast and the adjacent South East Coast of South Australia seem to have been uninhabited, at least from an archaeological point of view. Explanations for the absence of occupational evidence have been discussed in previous posts (Human Era 1). In summary they are:  

          (i) Submergence by rising sea levels (MWP),

          (ii) Small Aboriginal population (<20,000) in big Australian continent,

          (iii) Cold and arid conditions (LGM) caused contraction of numbers into refugia.

The existence of hunter-gatherers is signalled on land nearest to the 'drowned' Bonney Shelf, in sites at either end of Discovery Bay. Victoria's far SW and South Australia's lower SE is neatly divided by the Glenelg River, which runs south to the sea from its source in Grampians/Gariwerd. The Glenelg can be imagined not so much a physical barrier as a gateway into the coastal districts on each side. 

In the east is an open site of shell middens at Cape Duquesne and an enclosed rock shelter called Bridgewater South Cave. In the west are Koongine Cave and Malangine Cave. These investigated sites had two phases of occupation, Terminal-Transition and mid to late Holocene. Even within those phases, it is important to note that occupations were not continuous. Radiocarbon dates indicate intermittent and unique use.

The four sites emerge in the context of climate change. After the peak of the LGM, the region endured a period of acute dryness, lasting from about 17,000 to 14,000 ybp. Temperatures gradually rose but rainfall did not. Barely vegetated steppes were 'normal' and woody plants were uncommon. From about 14,000 to 11,500 ybp, rainfall increased in step with warmer temperatures, transforming the local landscape with grass and woodland ("dry sclerophyll forest"). The SW and SE became much more habitable.

Koongine and Malangine Caves in South Australia are both substantial limestone shelters on the edge of the coastal plain, 4 kms north of the present sea level and 10-15 kms from the paleo-shorelines contemporary with their occupation. Koongine has radiocarbon dates on charcoal between 11,050 and 9,200 ybp, with emphasis on the period 11 to 10.4 thousand years ago (ka). 

Animals consumed at this site include possum, bandicoot, potoroo, wombat, wallaby, and grey kangaroo. Minimal amounts of marine mollusc shell indicate this was a terrestial hunting camp. Some emu egg shell suggests it was used in winter.

Bridgewater South Cave is similar to Koongine but with older dates and a longer time span of use in the Terminal-Transition. Four radiocarbon ages range from 13,250 to 9,350 years before present (ybp). The same array of large and small mammals were consumed, but very little seafood beyond one seal and one fish. Despite being only 2-4 kms from the paleo shoreline, this was another autumn-winter camp, concentrating on terrestial prey. 

An archaeologist assessed the excavated remains of bone and limited flakes of stone as evidence of "an ephemeral use of the site as a hunting bivouac". Both Koongine and Bridgewater South suggest seasonal use with little interest in diving for shellfish in the colder months.

Cape Duquesne is, on the other hand, a wind-eroded shell-midden site on the top of a 50 metre high cliff. In the Terminal Pleistocene - Early Holocene time-frame, the Southern Ocean was only 1.4 to 1.4 kms from this site, connected to the cliff by a sand ramp. A total of 18 radiocarbon dates on shell stretch from 11,800 to 8,750 ybp, with the main period of occupation from 10,500 to 9,850 ybp.

There is no real sign of any economic subsistence activity beyond the collection, cooking, and eating of shellfish. 62% of the catch was Wavy Turbo, 24% was Variegated Limpet, 7% Wavy Top Shell, and 2% were Pipi. These are all rock platform species found on "high energy coasts", and at mid to sub tidal depths. Harvesting them requires immersion in rough ocean waters so it is sensible to assume seasonal, (i.e.summer) visits.

The site revealed 8 middens and a number of hearth features (burnt stones and charcoal),as well as flaked stone artefacts. Archaeologist comments are along the lines of, "small groups, on the scale of extended families, camped at this location for short periods of time, for durations of a few days to a few weeks at most."

Conservative interpretations of Cape Duquesne are informed by comparison with the nearby Cape Bridgewater shell middens. The more recent site, dated between 4,000 and 400 ybp, has nearly 200 times the amount of shell deposits, and when the relative timespans are taken into account, 100 times the annual discard rate of shell. The implied difference in population density from the Late Pleistocene to the Late Holocene is telling.

The 4 sites near the sea describe a very small population, moving seasonally between camps situated close to food sources. It is essentially a story of family groups, hunters and gatherers familiar with environmental opportunities, who returned to each place (winter cave or summer camp) at certain times of the year. This pattern is consistent with what is known of inland Terminal-Transition sites at Grampians/Gariwerd.

There is a reasonable possibility that the Grampian mountains provided sanctuary for reduced human numbers during the LGM, and that the river Glenelg supplied a migration corridor out there when conditions recovered after the MWP. The idea is supported by dates from Drual Rockshelter in the Victoria Range.

Drual lies at the bottom of a prominent outcrop of rock, 20 metres above the valley floor and 200 metres from the permanent water of Deep Creek. Its lower unit of excavated material has returned (uncalibrated) dates of 22,160 +/- 150 ybp, 22,140 +/- 160 ybp, and 15,840 +/- 120 ybp,  which clearly shows occupation of the rock shelter during the Last Glacial Maximum.

The shallowness or small quantity of the deposits indicates that visits were infrequent and transient. This pattern of short, occasional stays at Drual is not picked up again until 3 dates from the Early Holocene ― 8,170, 8,050, and 7,720 ybp ― matching local sites Billimina Rockshelter at 8,940 ybp and Blackfellows Waterhole at 8,790-9,010 ybp. 

Scientifically, connecting the Grampians-refuge people with the post-glacial colonists on the coast is speculation. But the the fact remains that Drual gives the only dates from SW Victoria before the Melt Water Pulse, making it a very important site in Aboriginal archaeology.

 

Miniscule quantities of red ochre have been found at Drual and Billimina rock shelters. Similarly tiny amounts were found at Cape Duquesne but these shared traces of distinctive cultural material are not enough to link the inland and shoreline groups.

Both Drual/Billimina and Koongine/Malangine have rock art, the former painted on hard mudstone, the latter carved into softer limestone. There isn't, however, a reliable method to connect each act of art-making on cave walls and ceiling with the particular levels of occupation buried in the floor below. None of this artwork can be confidently assigned to the Terminal-Transition, which is needed to demonstrate a shared tradition between the northern and southern sites.

The coastal communities do show some similarities with the inland ones (especially with Blackfellows Waterhole), but these are more general themes than an identical sameness.

The archaeology of Discovery Bay is one of short term camps associated with the exploitation of specific food types. It is also about repetition, with people returning to the same place to harvest the same style of prey (even to the same species of shellfish). This implies habitual behaviour within the 'boundaries' of familiar territory.

There is a sense of custom, of established patterns of behaviour and a regular relationship with particular parts of the landscape, from these Late Pleistocene people. This is more than simple nomadism. And in their geographical connectedness they are consistent with the Early Holocene people in and around Grampians/Gariwerd, who also revisited 'occasional' camps with a definite purpose in mind.

Although there is no proof of a post-glacial migration down the Glenelg River from the Grampian Mountains, the theory explains the absence of human occupation evidence before then. From 13,050 ybp onwards to the present time, there is never any question about the Aboriginal 'footprint' in SW Victoria and SE South Australia.

__________________________________________________________________________

REFERENCES

Thomas Richards, 2012, 'An early-Holocene Aboriginal coastal landscape at Cape Duquesne, southwest Victoria,

Australia', in Peopled Landscapes: Archaeological and Biogeographic Approaches to Landscape, ANU E-Press

Thomas Richards, 2013, 'Early Holocene Aboriginal occupation at Blackfellows Waterhole, Barrabool Flora and Fauna Reserve, south-west Victoria, Australia', The Artefact, 36, 32-38

Robert Bednarik, 1994, 'Malangine and Koongine Caves, South Australia', The Artefact, 17, 43-60

Robert Bednarik, 1998, 'Direct Dating Results from Australian Cave Petroglyphs', Geoarchaeology, 13.4, 411-418

Caroline Bird and David Frankel, 2005, 'An Archaeology of Gariwerd: From Pleistocene To Holocene in Western Victoria', Tempus, 8, 1-116