Review of ‘Palaeonenvironmental Change and the Persistence of Human Occupation in Southwestern Australian Forests’ by Joe Dortch
01st June 2006
Palaeonenvironmental Change and the Persistence of Human Occupation in Southwestern Australian Forests by Joe Dortch. BAR International Series 1288, Archaeopress, Oxford, 2004, xi+226 pp., ISBN 1 84171 638 3 (pbk).
School of Earth and Geographical Sciences, The University of Western Australia, 35 Stirling Highway, Crawley WA 6009, Australia
My first comments address the publisher’s role, rather than the author’s. BAR monographs fulfil a very useful function; many have become archaeological classics. But my copy of this book started falling to pieces before I had finished reading it, which hardly encourages readers to regard these studies as of durable worth. As well as more robust construction, tighter editing would have been helpful. Just as one example of unfortunately common minor inconsistencies, did Dortch measure artefacts ‘to the nearest 0.01 mm’ or ‘to the nearest tenth of a millimetre’ (both on p.89)?
The title of the monograph is distinctly misleading. ‘South-Western Australian Forests’ (Figure 2.1) extend north beyond the latitude of Perth, but this study examines in detail only their extreme southwest corner. Dortch does not always make it clear when he is talking about the southwest in the general sense of the Perth/Leeuwin/Albany triangle; when his remarks refer to the whole southern portion of this triangle (or the wider karri region within it); and when they are limited to his ‘study area’, ‘the Leeuwin-Naturaliste Region, extreme south-western Australia’, the limits of which remain undefined. Statements about the availability of certain plant and animal resources in the southwest do not necessarily apply to every region within it; for instance Dioscorea hastifolia does not occur south of the Murray.
The author could have benefited also from an editorial reminder that mathematical arguments need to be expressed with extreme clarity, particularly when addressed to a readership which, even if adequately numerate, does not put mathematical manoeuvres at the centre of its interests. I found it strange that the main text should retain lengthy and recondite mathematical arguments, whilst straightforward artefact descriptions and drawings were consigned to an appendix. Nor does elaborate statistical treatment necessarily resolve difficult issues. For instance, Dortch puts a great deal of time and effort into tests on changes in flake shape over time, only to deduce ‘more frequent’ visitation of Devil’s Lair at the Last Glacial Maximum (LGM) when there is lack of intensive reduction of chert artefacts (Dortch had previously proposed intensive retouch as the criterion of increased site usage). He then reverts to basing his deduction on the ‘large numbers’ of chert artefacts per unit time and volume, Bill Ferguson’s much simpler criteria, which Dortch had previously rejected.
Dortch does not give sufficient credit to the pioneering work of Ferguson (1985), both in field methodology and in interpretation. Ferguson’s wide and systematic application of road surveys, test-pits and large area excavations in southern forested regions set a new benchmark in field methods (the excavated sediments in Ferguson’s sites were neither ‘poorly stratified’ nor ‘unconsolidated’, as can be seen from the trench depths and sharp outlines in his published photographs).
On the interpretation of the fluctuations in artefact numbers at these sites, both Ferguson and his critics can be faulted. Ferguson’s critics have argued that artefact numbers cannot provide a proxy for demographic trends. The post-4000 increase might be partly explained by changing technology, but this does nothing to account for the earlier decrease in rate of artefact deposition. On the other hand, if Ferguson’s histograms are put side by side, the low points do not coincide between sites, as his ‘depopulation’ interpretation requires. A more economical explanation would be shifts in relative usage, increasing inland in moister, seaward in drier, phases.
I find the data Dortch presents (e.g. Figure 4.1) entirely consonant with continuing overall population growth in the southwest as a whole. His graph shows growth in site numbers to be approximately exponential, which is exactly the mathematical expectation if total site numbers at any one time provide an approximate index of usage and thus of population. Dortch thus provides the best evidence against his own doubts about the validity of using site and artefact numbers as indices of population! Natural demographic increase is an entirely sufficient explanation, without the need to postulate improved technology or efficiency.
The core of Dortch’s book, however, lies not in his interpretations of his own work and of the work of others, but in his description of his own field survey and excavations, and analyses of biotic and lithic remains. Dortch’s lithic analyses (as distinct from their interpretation) set a new benchmark for the meticulous treatment of appallingly difficult stone assemblages. From Tunnel Cave, for instance, the fossiliferous chert ‘tool’ assemblage comprises ‘all utilised pieces … only one has retouch’ (p.100). The quartz is worse. ‘I have examined’, Dortch tells us, ‘every stone artefact from Tunnel Cave and Devil’s Lair under a binocular microscope at magnifications from x6 to x40’. Material, artefact type, dimensions, weight, platform, dorsal scars, ventral scars, use scars, cortex, pot-lids, heat-crazing, all were recorded. This represents a truly massive and admirable labour.
The results are also impressive. This is intractable material. Much of it is quartz, notoriously difficult to analyse. Even among artefacts made from chert, superior in its flaking properties to quartz, a mere handful can be slotted into ‘formal tool types’, mainly pieces that might have been bagged simply as ‘amorphous scraper’, or ‘utilised flake’. Dortch, however, barely concerns himself with formal typology, but concentrates rather on size and ‘intensity of usage’.
In Devil’s Lair (p.97), at the LGM, as chert became more easily accessible from sources now offshore, so people used it more lavishly, making larger tools, and not bothering to squeeze every last drop of use out of any one blank. Later, as sea-levels rose, access to chert sources became limited once more, and people used their products more carefully. Patterns of usage of the more evenly available quartz remained unchanged, Dortch actually concedes (p.97) that human visits to Devil’s Lair may have been ‘more frequent’ at the very time that ‘chert artefacts are not greatly reduced (flaked intensively)’ (my emphasis). This is the opposite of the contemporary situation in Tunnel Cave (p.105), where ‘the high ratio of tools to debitage, and used artefacts to unused artefacts’ leads him to deduce more ‘intensive’ occupation around the LGM. Dortch thus effectively scuttles his own argument that intensity of usage of artefacts measures demographic trends over the whole region. He continues to maintain, however (p.109), that ‘Tunnel Cave provides little evidence that occupation intensity declined in response to major vegetation changes … from 12,000 to 8,000 BP … except that there are few artefacts dated to this period’ (my emphasis). Quantities of bone agree with quantities of artefacts (p.129). Why not concede that this ‘decline in artefact numbers’ is sufficient and acceptable evidence, showing itself more valid than devious and dubious measures of ‘intensity’ of artefact usage?
The data Dortch amasses may not always be able to answer the questions he asks. Bones and charcoal in living deposits were not necessarily garnered right outside the front door. Perhaps when a habitat did change, people just went further to get the same resources, rather than ceasing to visit a really cosy pied a terre. Contrary to Dortch’s statement (p.18) there is no guarantee at all that ‘in such a fine mosaic … it is the immediate surroundings of a site that may be most important for hunter- gatherers’ (my emphasis).
Like other mammals, humans require much more than one sort of vegetation, for access, shelter, foraging and social, mating and ritual activities. A habitat may be intensively used, but have no camp-spot actually within it (e.g. alluvial yam-diggings), or be a good place for a camp, but draw its resources from the surroundings (e.g. many sandblows). The movements of an Aboriginal group over days, weeks, months, years and decades could take them repeatedly through the meagre karri of the Margaret River limestone ridge, and around and beyond it, taking in terrain and resources kilometres and tens of kilometres away. The real questions are about relative usage, of fairly wide ecological zones, which can be answered (if at all) only by wide and intensive survey and excavation.
I found the discussion of burnt bone in hearths fascinating but flawed. Dortch (p.132) equates ‘the proportion of each species burnt specimens in all species burnt specimens’ with ‘the proportions of species present before the hearth was built’. However, the sample from which the first proportion was derived included not only the bones below the hearth, but additionally bones from the hearth, some of which will have been put in later; the equation is therefore untrue. It would be more useful to compare expected numbers of one species burnt with actually occurring numbers, then calculate the difference due to addition bone humanly deposited in the fire.
I am particularly interested in Dortch’s discussion of the role of Aboriginal firing regimes in vegetation change, in partial avoidance of such change and in the (unintended) acceleration of change. I would agree with his deduction that ‘hunter-gatherers need not abandon regions characterised by mosaic vegetation’.
Dortch conveniently sets up an ‘Aunt Sally’ to knock down, in the form of a supposed prior assumption that the southwest karri forests ‘severely limit or even exclude human occupation’. This is a useful expository device, but the Hallam (1975) pages cited do not say this. Hallam nowhere states nor implies that Aborigines avoided all karri forest; nor (despite Dortch p.41) that karri has few foods, necessarily a dense understorey, and is too wet to burn. It remains probable that karri forests in general were ‘burnt only patchily, for example on their western margin’ (Hallam 1975:27). The west is the area of Dortch’s study, a much thinner and more penetrable belt of karri than the extensive block inland from the south coast, from Pemberton across to Walpole. This main block may well have been frequented and burnt more ‘rarely’ (p.27), and be ‘less exposed and liable to fire, though not unaffected on its coastal margin’ (Hallam 1975:55). Hallam (1975:75, 103) points out the diversity within forests; and cites historical and bore data which suggest Aborigines were burning, even if patchily, in the deep south near Broke Inlet and William Bay (if fire return periods of ‘less than a few years’ produced sedgelands in Tasmania, do the sedgelands of the south coast of Western Australia imply anything about Aboriginal burning?).
The work of Pearce (1982) and of Bonner (Anderson 1984) confirmed that the forested triangle of the southwest and the karri areas of the far south, never supported really intensive human activity and population like that on the west coastal plain. Pearce’s forest sites were not ‘somewhat smaller’ than coastal plain sites, but about a tenth of the size, with around a tenth of the artefact density. The density of all sites, at more than 6/km² on the coastal plain, contrasts with around 1/km² in forest; while for major sites, with thousands of artefacts, the figures are 1/km² on coastal plain against between 0 and 0.06/km² in different parts of the forest (Anderson 1984:20-21). Overall decadal usage per unit area on the coastal plain was thus probably several degrees of magnitude greater than in the northern jarrah forest. Dortch gives no comparable site density, size, or artefact density figures for any part of the far southwest.
Dortch provides a very useful summary of ecological work on the varied vegetation and fauna of the southern southwest. He could profitably have discussed how far Aboriginal occupancies related to and depended on the important and widely available resources of non-forested patches of heath, scrub, granite outcrop and their surrounds, swamps, estuaries, and sedgelands, widely scattered (see map p.19) within the generally forested districts (Christensen 1992), rather than to forest pure and simple. Aboriginal occupancies of forests only, or of any one vegetational formation, is not a meaningful concept. Aboriginal groups used the total resources of an entire area, forest and non-forest. Non-forest resources within forested contexts may well have been overall more important for subsistence than the surrounding forest.
Dortch’s hard-won evidence does not enable him to come to incontrovertible conclusions. But to have sharpened the questions posed is a considerable achievement, and should lead to more, and more cooperative, work in a field of multidisciplinary and public significance.
Anderson, J. 1984 Between Plateau and Plain: Flexible Responses to Varied Environments in Southwestern Australia. Occasional Papers in Prehistory 4. Canberra: Department of Prehistory, Research School of Pacific Studies, The Australian National University.
Christensen, P. 1992 The Karri Forest. Como: Department of Conservation and Land Management.
Ferguson, W.C. 1985 A Mid-Holocene Depopulation of the Australian Southwest. Unpublished PhD thesis, Department of Archaeology and Anthropology, The Australian National University, Canberra.
Hallam, S.J. 1975 Fire and Hearth: A Study of Aboriginal Usage and European Usurpation in South-Western Australia. Canberra: Australian Institute of Aboriginal Studies.
Pearce, R.H. 1982 Archaeological sites in the jarrah forest, southwest Australia. Australian Archaeology 14:18–24.Sylvia Hallam
Review of ‘Palaeonenvironmental Change and the Persistence of Human Occupation in Southwestern Australian Forests’ by Joe Dortch
You must be a member to download the attachment ( Login / Sign up )