Academic journal article Archaeology in Oceania

Change or Decay? an Interpretation of Late Holocene Archaeological Evidence from the Hamersley Plateau, Western Australia

Academic journal article Archaeology in Oceania

Change or Decay? an Interpretation of Late Holocene Archaeological Evidence from the Hamersley Plateau, Western Australia

Article excerpt


Data collected from the Hamersley Plateau over the last four decades are examined for patterns in the archaeological record. Data relating to the timing of the archaeological appearance of backed artefacts, seed-grinding technology and rock art are currently too few to indicate major cultural changes with certainty. Increases in numbers of radiocarbon dates from archaeological sites on the Hamersley Plateau are evident in the late Holocene. This can be interpreted as a pattern of cultural change or natural decay in datable material. I conclude that taphonomic bias is the most important variable in the distribution of the radiocarbon date sample from the Hamersley Plateau. That said, further accumulation of dates and data may show archaeological changes in the Hamersley Plateau that represent local expressions of broader trends in the Australian semi-arid and arid zones.

Keywords: Pilbara, Holocene, chronology, cultural change


Studies of regional archaeology in Australia often describe a late Holocene (i.e. after 4000 BP) phenomenon of increases in the diversity and quantity of archaeological evidence (e.g. David and Chant 1995, Flood et al. 1987, Lourandos 1983, Morwood 1987). Despite the abundance of archaeological resources in the Hamersley Plateau, it is one of the few regions of the Australian continent that has been excluded from discussions of late Holocene change (Figure 1). This is mostly because data from the area are hidden in a grey literature of consultant reports that has accumulated over the last four decades. This aim of this paper is to evaluate the implications of evidence on the dating of backed artefacts, seed grinding technology, rock art and fluctuations in occupation intensity on the Hamersley Plateau.

The dating of backed artefacts and adzes

Backed artefacts and adzes have long been a focus of debates about cultural and technological change in Australian archaeology (Bowdler 1981; Bowdler and O'Connor 1991; Hiscock 2001; 2002; Hiscock and Attenbrow 1998; 2004; Layton 1996). The first evidence of these new types on the Hamersley Plateau is a backed stone artefact with resin on the backed section at Newman Rockshelter. The artefact is associated with a date of 3740 [+ or -] 100 BP (Brown 1987:27). Other specimens at this site include five backed artefacts and one adze all deposited after 3700 BP (Brown 1987:29). Backed artefacts first appear at Marillana A at 3000 BP and at 1700-1100 BP at Cleft Rock Shelter (Marwick 2002). A single backed geometric microlith occurs at Site P5313 immediately below a 2400 BP date (Brown 1987:43). At RR3--O there are two adzes representing the new Holocene types in association with a 310 [+ or -] 50 BP (Wk 8364) date, suggesting that the new types were continuously present in the Hamersley Plateau until the very late Holocene (Harris 2000:17-25; cf. Hiscock and Veth 1991).

Hiscock (1993, 2001) and James and Davidson (1994) have suggested that a coincidence between the appearance of backed artefacts and increases in artefact discard may represent a sample size effect rather than an important cultural change. This is true here with increases in discard rates of cultural material coinciding with the first appearance of backed artefacts at 3000 BP at Marillana A and 1700-1100 BP at Cleft Rock Shelter (Marwick 2002). Table 1 shows that the strength of correlation between numbers of backed artefacts and excavation unit assemblage size is very low at the five Hamersley Plateau sites with backed artefacts. However, the number of backed artefacts is very small and none of the correlations are significant at the p([H.sub.0]) = 0.05 level, so sample size effects on the presence of rare items cannot be ruled out.

The dating of rock art

Like backed artefacts, increases rock art production are often considered to represent important cultural changes. At Skew Valley on the Burrup Peninsula five buried engraved panels were recovered by Lorblanchet (1983) from a stratified midden deposit associated with charcoal dated to 3770 [+ or -] 80 BP (ANU 1837), 3410 [+ or -] 80 BP (ANU 1839) and 2770 [+ or -] 70 BP (ANU 1838). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.