Weld-RS-0731 Rockshelter ,Weld Range, Western Australia – a Mid- to Late Holocene Sequence in Local, Regional and Inter-Regional Context

Australia: The Land Where Time Began

According to Winton et al. the Aboriginal archaeology of the inland Mid West region of Western Australia has remained poorly synthesised. In this paper Winton et al. consider a sequence that has been excavated at Weld-RS-0731 Rockshelter dating to the mid- to late Pleistocene in the local context of surface archaeology as well as other excavated sites. In terms of the size of artefacts, technological types and lithologies, which indicates repeated patterns of site function and specialised task activities. A paucity at the inter-regional scale has resulted in models of cultural and linguistic shifts in the Late Holocene have inadequately considered the role of the inland Mid West, in spite of the discoveries of some sites that are highly significant such as Walganha, Wilgie Mia and the ochre mines at Little Wilgie. The earliest absolute estimate of age for the use of the Little Wilgie Mine of 2,500 cal. BP was provided by an ochre fragment that was excavated at Weld-RS-0731 and provenanced geochemically to Little Wilgie. Winton et al. discuss the possible role of the inland Mid West during the Late Holocene inter-regional relations and the resultant cultural and linguistic shifts.

Charles Dortch collaborated with François Bordes and others during his time at the West Australian Museum on ground-breaking field work in the Mid West of the state (Bordes et al., 1983). Little has changed in the understandings of the Aboriginal past in the Mid West over the next 30 years. There are apparent associations between stone artefacts and extinct megafauna in deposits dating to the Pleistocene at the Murchison River and Greenough River is still being investigated (Murszewski et al., 2014) and (Bordes et al., 1983) work at Billibilong Spring and Walganha still provides the basic framework for the Holocene (cf. Baynes, 1984; Bindon, 1986; Crawford, 1980; Davies, 1961; Davies et al., 1977). It was found that site and artefacts frequencies in the Holocene in the Mid West increase over time and, after 4,000 cal. BP, contain evidence of the emergence of microlithic technologies. A reliance on this unrefined record and very little comparative data has hampered new research in the region. As a consequence there is a poorly developed understanding of the role of the Mid West in inter-regional cultural and linguistic shifts (McConvell. 1996; Veth, 2000). In this paper Winton et al. help to rectify the situation by the use of lithic artefacts data to investigate the patterning in the Late Holocene in part of the Weld Range, which is 50 km to the north of Walganha, contextualising the results of the excavation of a rockshelter with regional comparisons and by the development of arguments around the possible peoples in the Mid West in the Late Holocene regional cultural and linguistic change.

The Weld Range Project (WRP)

Refining archaeological and cultural understandings of the Weld Range, an arc of rocky, banded iron formation (BIF) hills, 60 km long and oriented in a southwest-northeast direction, located 50 km west-northwest of Cue, was the main focus of the Weld Range Project. The Weld Range ties together a local landscape of claypans, that hold water, seasonal creeks and wash areas, sand and gravel flats, laterite breakaways and granite domes, as one of the Mid West’s greenstone belts, which are described (Conservation Council of WA, 2007) as ‘islands in an otherwise flat landscape’. This landscape is rich in Aboriginal heritage, most notably the Aboriginal ochre mines at Wilgie Mia and Little Wilgie which are listed as national heritage. The Weld Range was an important ceremonial centre in the past, with several law grounds recorded in the vicinity (O’Neill & Jordan, 2007).

Together with traditional owners, archaeologists have carried out a systematic pedestrian survey in and around the Weld Range which has resulted in the description of a rich and varied surface archaeological record that has been minimally disturbed (Winton et al., 2010). They have identified a wide range of site types, though the most common are artefact concentrations and quarries of the various available siliceous volcanic and metamorphic rocks (Elias, 1982). It is indicated by work that had been undertaken on the WRP to date that an intensive use of the different habitats of the landscape, with a signature that is particularly strong in, and of, riparian environments (Byrne et al., 2013: 104-105).

Weld-RS-0731

During CHM fieldwork the research potential of Weld-RS-0731 (Department of Aboriginal Affairs Site 28793) was recognised. At 38 m² this site is a moderately sized rockshelter that faces northwest, situated within a rock face that is highly visible on the northern flank of the Weld Range. On the walls of the site there are 24 pigment motifs that have been recorded, including 18 hand stencils appearing to include hand stencils of men, women and children. The reason the site was chosen for excavation was its potential archaeological deposit, as a locale that is associated with the chaines opératires of ochre use, which is a major theme in the archaeology of the Weld Range, and because it is at risk of disturbance from mining. There are 3 AMS age determinations from charcoal recovered from this site that have been published, as well as an anthracological analysis (Byrne et al., 2013). A sample (WK 32139) recovered from the lowest level at Weld-RS-0731 produced an age estimate of 5,645-5,470 cal. BP now provides the earliest evidence for Aboriginal occupation of Weld Range. Within a trace element characterisation and provenance study (Scadding & Watling, 2012) there was a fragment of ochre. This fragment dated to about 2,500 cal. BP, and was found to match the trace element chemistry of Little Wilgie, which is 5.5 km east-southeast of Weld-RS-0731 from where it was assumed to have been procured.

Building on the approach that was applied by Byrne et al. (2013), Weld-RS-0731 is understood as a node that past Aboriginal activity can be explored from. Importantly, according to Winton et al. contrasting with the assemblage that was recovered from Walganha and Billibilong Spring, which was dominated by quartz (Bordes et al., 1983; Webb & Gunn, 1999), the assemblages that were recovered from Weld Range contains a vast range of raw materials, with quartz being a ubiquitous but minor component. Lithic assemblage analysis that is landscape focused was used to exploit the rare opportunity that is provided by an extensive comparative surface survey data as well as the results of a low power magnification lithic use-wear study were integrated by Winton et al. into a discussion of Late Holocene interactions on an inter-regional scale. A description and sediment analysis of the Weld-RS-0731 excavation included on-line supplementary material.

Discussion

Landscape analysis

There were 3 basic models for the potential form and content of the assemblages at Weld-RS-0731 were derived from analysis from nearby surface sites. The assemblage from Weld-RS-0731 differs from all of the local analogues in terms of artefact raw materials, technological types or sizes. At Weld-RS-0731 there is a much higher proportion of fine-grained materials, though the sources for most of these are known to be present within 5 km of the site. An exception is a bladelet made from brown silcrete (ID90) from an unknown source, and the only clear example of the manufacture of a specialised blade in the assemblage, which demonstrates different treatment of a rare, exotic raw material. The assemblage shows clearly that fine-grained rock types were specifically selected for use at this site, though Weld-RS-0731 conforms to the general pattern for preferred use of materials that were available locally.

Included among the artefact types at Weld-RS-0731 are higher proportions of flakes, microflakes in particular (<10 mm), broken flakes and very low numbers of cores, when compared to other assemblages at Weld Range. A fair representation of the flake sizes that were preferred (15-44 mm) is given by the size of flakes that had been discarded at A-sites. Only limited evidence of daily subsistence tasks at all other local sites is provided by the low proportion of flakes at the Weld-RS-0731 site, in combination with the difference in proportions of artefact types between Weld-RS-0731 and those in all other surface assemblages within 5 km.

It is indicated by these characteristics that Weld-RS-0731 was not used as a residential base, too few of the elements recovered from among the general surface assemblages support the interpreting of Weld-RS-0731 as a site where a wide variety of general domestic tasks were occurring over a period of many days or weeks. Also, the description by Binford (1980) does not fit the description advanced as a forager ‘location’, a place that was occupied briefly for resource extraction where few cultural remains were discarded. Winton et al. suggest, that as the occupants made fires and produced rock art, which suggests that ate least part of the significance of Weld-RS-0731 must have derived from the properties of the place itself “as a comfortable place to sit, seek refuge or communicate symbolically through art” the significance of the site was more like a place itself, and not the importance of the available subsistence resource. Also, Winton et al. suggest it is indicated by the preference for fine-grained raw materials at Weld-RS-0731 that there was forward planning involved in the occupation of this site, such as could be expected for the type of location occupied by small numbers of skilled, knowledgeable individuals in order to exploit particular resources. Binford (1980) used purely subsistence terms to describe field camps, which may not be appropriate for all periods of use at Weld-RS-0731.

Though it is indicated by consideration of Binford’s (1980) model of site type, mobility and residential organisation that at Weld-RS-0731 occupation was more at the strategic end of the spectrum, it does not factor-in any possible non-subsistence reasons for the site. The archaeological signature of historically occupied rockshelters in the Western Desert has been assessed (Nicholson & Cane, 1991). They found by ethnographic consultation that rockshelters had either been used a habitation sites or for sacred activities. At habitation sites material culture included flaked stone artefacts for a variety of domestic tasks, with small portions of grinding material, hammerstones and ochre (Nicholson & Cane, 1991:304). There were very few flaked stone artefacts or grinding stones, a small number of hammerstones and many manuports, though ochre was common, and there was also rock art (Nicholson & Cane, 1991:304) at sites that were associated with sacred activities. The evidence for use of Weld-RS-0731 as a habitation site, flaked stone artefacts are common, though in a range of sizes that may not have been detected easily in the surface survey (Nicholson & Cane, 1991). Some evidence of woodworking may be provided by the adze (ID27) and by use wear on artefacts ID247 and ID21, though there was no grinding material that was identified. Though the rock art could indicate the site had been used for ritual purposes. Manuports were not recorded at this site, though they might be difficult to identify because of the scree of boulders in the rockshelter. According to Winton et al. there is a mix of habitation and ceremonial assemblage characteristics at Weld-RS-0731 (Nicholson & Cane, 1991). It could be suggested by this there were differences between the Western Desert and Weld-RS-0731 in terms of material culture associated with habitation of ceremonial sites. An alternative suggestion is that the use of Weld-RS-0731 may have varied in such a way that ritual use and habitation occurred at different times. The site may have been used by different groups for varying purposes over time, as the hand stencils for women, men and children, which are often groups who are segregated from each other during ritual activity, would suggest.

Conclusion

This study contributes a case study of the mid- to Late Holocene site from the Weld Range, providing comparative data and a discussion staged at local, regional and inter-regional scales. Winton et al. concluded that Weld-RS-0731 was not used as a residential camp or a forager location, though it does show some similarity to a field camp (sensu Binford, 1980). A task specific, logistical use of the site is suggested by the unusually high proportion of fine-grained materials form a diverse variety of sources, and there is little evidence that would suggest long-term occupation. Comparison of the assemblage from Weld-RS-0731 against the analysis (Nicholson & Cane, 1991) suggests however, that the site may have been used for both ceremonial and short-term habitation purposes.

A high proportion of artefacts from both Weld-RS-0731 and Walganha is comprised of microflakes, which differed from lithic assemblages from other sites in the region. Winton et al. tentatively suggest microflakes may have been used in particular ways during occupation of rock art sites, possibly for wood carving or body scarification, though much more research is needed to be more confident in this suggestion. A crystal flake recovered from Weld-RS-0731 displays use-wear that provides broad support for the idea that very small flakes that have not been retouched were used as tools in their own right, possibly in the ways suggested previously in this paper.

At Weld-RS-0731 the diachronic pattern of occupation concurs broadly with other data from the region that exhibits increased artefact frequencies from about 4,400 cal. BP, and the highest discard rates in the last 1,100 cal. BP. Following the lead from their Wajarri colleagues, who attest to strong affiliations with Western Desert culture, Winton et al. suggest this is an important avenue for future research. In the rock art at Walganha there are hints of a relationship and the inter-regional expression of microlithic technology. They suggest it is also likely there are inter-regional relationships that have incorporated the exchange of ochre from Wilgie Mia and Little Wilgie, with the latter being used since at least 2,500 cal BP.

Sources & Further reading