Technological organization and settlement in southwest Tasmania after the glacial maximum.

McNiven, Ian J.

A growing quantity of data about the late Pleistocene sequence in Tasmania has not been matched by an equivalent clear understanding of just what are the patterns of its lithic record. A new model is developed.

Introduction

The discovery over the last decade of numerous sites dating within the period c. 35,000-11,000 b.p. in the densely-forested limestone valleys of southwest Tasmania has heralded a new era in late Pleistocene archaeological research in Australia. Many of these sites exhibit rich stone artefact assemblages unparalleled in density and volume by their chronological counterparts on the mainland (Cosgrove 1989; 1991; Cosgrove et al. 1990; Jones 1990; McNiven et al. in press). Pioneering research at Kutikina Cave identified an important change in stone artefact assemblages commencing at the time of the last glacial maximum around 18,000 b.p. (Jones 1988; 1989; 1990). However, more recent insights drawing on a larger sample of assemblages have down-played this pattern of dramatic change. This more conservative and static view sees the stone industries as 'largely amorphous' (Cosgrove et al. 1990: 70) and limited for analysis by theoretical problems of temporal scale and resolution in the archaeological record (Cosgrove 1991).

This paper develops Jones' model of chronological change at the time of the last glacial maximum. A new model is outlined which discusses chronological change in southwest Tasmanian stone assemblages in terms of technological organization and settlement restructuring. It is based on a synthesis of recent data on chronological changes in the use of raw materials (quartz and Darwin Glass) and thumbnail scrapers, and how these may reflect broader-based changes in mobility and the relative employment of curated components to the toolkit.

Quartz and Darwin Glass

The main raw material change after the glacial maximum is quartz and Darwin Glass use. In the lower part of the Kutikina sequence 'some 60% to 70%' of artefacts were quartzite, while in the upper levels 'over 90%' were quartz (Jones 1990: 279-80). Jones (1990: 280) notes that this changeover occurred 'between 16,000 and 18,000 b.p.'. The increased use of quartz has also been documented at Bone Cave, where the relative proportion of quartz artefacts changes from 5% before the glacial maximum to 14% after the glacial maximum (McNiven et al. in press).

Darwin Glass is a distinctive-looking raw material which occurs naturally as small, marble-sized pebbles. It was formed around 0.7 mya by meteorite impact and can be found within topsoil across its strewnfield which extends up to 20 km west of Darwin Crater (Fudali & Ford 1979; Jones 1990: 282). Artefacts of Darwin Glass were restricted to the upper section of Kutikina, with most of the top 14 levels containing 'a few glass flakes' (Jones 1988: 37; 1989: 769; 1990: 282). At Nunamira, 80% of all excavated Darwin Glass artefacts came from the upper levels of the site dated from c. 16,000 to 12,000 b.p. (Cosgrove 1991: 77-8, 153-4). A little further to the south, the single fragment of Darwin Glass recovered from Bone Cave is dated to after c. 18,000 b.p. (Allen et al. 1988: 82).

Thumbnail scrapers

The only distinctive and recurring implement type identified from the southwest Tasmanian assemblages is the thumbnail scraper (Cosgrove et al. 1990: 70). Characteristically, thumbnail scrapers tend to be small (maximum dimension |is less than~2.5 cm), discoid or semi-discoid in outline with a sub-tabular cross section. Retouching tends to be steep, finely executed and located on the dorsal surface of the distal half of the flake (cf. end scrapers). Of all the assemblages examined to date, Kutikina provides the best example of change in the use of thumbnail scrapers after the glacial maximum. Jones notes that the tools from the upper assemblage 'consisted almost entirely' (1990: 280) of thumbnail scrapers with 'some 80 examples' recovered from less than one cubic metre of deposit (1988: 36). No thumbnail scrapers are reported for the lower |is greater than~18,000 b.p. levels of the site. A similar dramatic increase in thumbnail scrapers occurs in Bone Cave, where over 95% of all specimens were recovered from levels dating to after 18,000 b.p. (Allen et al. 1988; McNiven et al. in press). Nunamira thumbnail scrapers register a slight increase in discard in the upper parts of the sequence with 63% occurring in levels dated between c. 16,000 b.p. and c. 12,000 b.p. (Cosgrove 1991: 140).

Establishment of new sites

Recent excavations at Mackintosh 90/1 Shelter (Mack 90/1), Condominium Cliffs 2 Rockshelter (CC2) and Maneena Langatick Tattana Emita Cave (MLTE) reveal stone assemblages with important similarities to the upper occupation levels of Bone, Kutikina and Nunamira Caves. In terms of chronology, all three sites show remarkable overlap with occupation commencing after the glacial maximum. Pleistocene cultural remains at Mack 90/1 are bracketed between c. 17,000 b.p. and c. 14,800 b.p. (Stern & Marshall 1993: 12), while the main in situ cultural sequence at MLTE is dated between c. 17,200 b.p. and c. 15,500 b.p. (Pocock 1992: 39-41). The sequence at CC2 produced a comparable basal date of c. 17,500 b.p. and a near-surface date of c. 13,800 b.p. (Brown et al. 1991: 33).

Stone assemblages from these later sites show important similarities to stone assemblages from the post-18,000 b.p. levels of the long duration sites. For example, Darwin Glass flakes and thumbnail scrapers were recovered from Mack 90/1 (Stern & Marshall 1993: 14), while 40 Darwin Glass artefacts were documented from MLTE (Pocock 1992: 48). Although only a single Darwin Glass flake was recovered from the small sample of 28 artefacts excavated from CC2 (Brown et al. 1991: 33), further excavations may reveal more examples. Thumbnail scrapers have been recorded at Mack 90/1 (Stern & Marshall 1993) and MLTE (Pocock 1992: 28).

Increased mobility and use of curated toolkits

From the above data it is clear that cultural systems operating across southwest Tasmania during the late Pleistocene underwent a change after the glacial maximum involving re-organizations of both technology and settlement. In technology, distinctive changes included the relative increased use of quartz, Darwin Glass and thumbnail scrapers. Although in some cases these changes were subtle, the presence of quartz, Darwin Glass and thumbnail scrapers in other sites established after the glacial maximum corroborates the notion of a region-ally consistent pattern. In terms of settlement, the relative synchronous establishment of new sites (e.g. Mack 90/1, CC2 and MLTE) identifies both a re-organization and possible areal expansion of land-use activities (cf. Lourandos 1993: 74).

It appears that the post-18,000 b.p. re-organization was associated also with an increase in both the magnitude and frequency of mobility arrangements (see Shott 1986; 1989). In terms of mobility magnitude, the increased use and movement of Darwin Glass across southwest Tasmania during this period suggests an increase in the area over which people were moving. In this sense, it is not necessary to view Darwin Glass as either 'highly prized' (Jones 1990: 282) or a 'central trade item' (Pocock 1992: 85). Alternatively, the increased use and spread of this raw material with suggested 'bad flaking qualities' (Cosgrove 1991: 168), at least in terms of brittleness and small nodule size (Fullagar 1986: 339; Jones 1990: 282), may only reflect changing settlement and mobility patterns which increased people's relative access to Darwin Crater and its splash zone (Cosgrove 1991: 168; Cosgrove et al. 1990: 71; Fudali & Ford 1979; Jones 1990: 281-2).

With respect to the number of camp moves per year or mobility frequency, the establishment of new sites immediately after the glacial maximum may reflect an increase in both the number of sites used by group(s) operating in the region and the number of residential moves per unit time. It should be noted, however, that in some cases the establishment of new sites may have been complementary to the abandonment of other sites in the region.

Is it possible that increases in mobility frequency after the glacial maximum were causally related to synchronous increases in the use of thumbnail scrapers? Shott's (1986) analysis of ethnographic data found that a strong positive correlation exists between mobility frequency and the curated component of tool kits. Put simply, 'the more you move, the less you carry and the more you conserve what you carry' (Shott 1989: 221). This observation may have some explanatory value for southwest Tasmania assemblages if it can be demonstrated that many thumbnail scrapers used after 18,000 b.p. represent curated or long-term maintainable tools. Although data are sketchy at present, three lines of evidence are consistent with this hypothesis.

First, the manufacture of many thumbnail scrapers from exotic raw materials is consistent with the inter-site transportation aspect of the curation hypothesis. For example, at Kutikina, Jones (1990: 282) identified thumbnail scrapers made from Darwin Glass and chert sourced 26 km and 50 km respectively from the site, while the Darwin Glass thumbnail scrapers at Mack 90/1 are located 70 km from Darwin Crater (Stern & Marshall 1993: 14). At Nunamira, Cosgrove notes 20% of thumbnail scrapers were made from silcrete which outcrops 'approximately 20 to 30 kilometres' from the site (1991: 135, 144). Although he argues that the 'most parsimonious explanation for the source of the chert' from which most (54%) thumbnail scrapers are manufactured is local within the Florentine Valley, exotic sources cannot be ruled out given that his own surveys failed to locate sources with the same high-quality properties exhibited by the archaeological samples (Cosgrove 1991: 130-34, 144, 167). Similarly, over 80% of thumbnail scrapers at Bone Cave are manufactured from either high-quality chert or silcrete for which no local sources are known.

Second, the manufacture of a disproportionately high number of thumbnail scrapers from chert, a high-quality raw material with excellent flaking properties, in eastern southwest Tasmanian sites (e.g. Bone and Nunamira Caves) is consistent with the design requirements of reliability, maintainability and flexibility associated with curated tools (see Nelson 1991: 6673). By definition, curated tools are 'curated and transported to and from locations in direct relationship to the anticipated performance of differing activities' (Binford 1973: 242). As such, a bias should exist towards high quality raw materials which allow both reliability and flexibility during tool use/maintenance/reshaping activities over the long term. As Goodyear (1989: 4) states, 'such materials can be said to have a high degree of "plasticity" because they can be transformed from design to design with the greatest ease and success'. It defeats the purpose to manufacture curated tools from poor-quality raw materials with limited potential for long-term use and manipulation (Gould & Saggers 1985). In this light, evidence for thumbnail scraper resharpening (Cosgrove 1991: 145; Fullagar 1986: 342; Stern & Marshall 1993: 14) is consistent with the curation hypothesis. Similarly, thumbnail scraper flexibility is demonstrated further by the wide range of task applications (e.g. butchering, bone-working and woodworking) exhibited by thumbnail scrapers from Kutikina (Fullagar 1986: 325-51; Jones 1990: 281).

Increases in both the use of quartz and the manufacture of quartz thumbnail scrapers after the glacial maximum seen at some sites (e.g. Kutikina and Bone Caves) may also reflect the need for a micro-crystalline raw material with excellent edge-holding qualities. However, it is likely that quartz was ranked below chert in terms of overall raw material quality, given that much quartz in southwest Tasmania exhibits flaws. Quartz may only have been used instead of chert to manufacture thumbnail scrapers if local chert sources were unavailable.

Third, the possibility of thumbnail scraper hafting is consistent with observations that 'curated tools are likelier to be halted than expedient tools' (Shott 1986: 39; see also Binford 1979: 269; Fullagar 1986: 350; Jeske 1989: 36; Keeley 1982: 799). Many thumbnail scrapers exhibit a small, extra section of marginal retouch on the ventral surface near the platform. Studies on small end-scrapers similar to thumbnail scrapers from other parts of the world suggest that this ventral retouch may be to shape and taper the tool to accommodate a haft (cf. Clark 1959: 202; Deacon & Deacon 1980; Gallagher 1977: 410; Holmes 1912: 140; Lothrop 1989: 114; Rule & Evans 1985: 216). Similarly, Fullagar (1986: 348-9) found that many of the 'small rounded convex tools' (thumbnail scrapers) from Kutikina exhibited signs of wear consistent with having been hafted. Diagnostic wear patterns included

1 'abrasion, slight rounding and scarring on the edge opposite to that with extensive distinctive use-wear (working edge)', and

2 'restricted abrasion, rounding and short striations on ridges and some surfaces away from the edge with distinctive use-wear (working edge)'.

The mobility/curation hypothesis may be tested further by examining the association between thumbnail scrapers and exotic raw materials in a larger sample of sites. In this context, the absence of thumbnail scrapers from late Pleistocene levels of ORS7 Rockshelter located in southeast Tasmania is consistent with the near-exclusive use of locally available hornfels and quartzite (Cosgrove 1991: 180, 206, 310; Cosgrove et al. 1990: 70). Such patterning suggests differing settlement/mobility arrangements compared to southwest Tasmania, a finding consistent with Cosgrove's model of systemic differences between southwest and southeast Tasmania during the late Pleistocene (see Cosgrove 1991; Cosgrove et al. 1990).

An apparent inconsistency with the thumbnail scraper curation hypothesis is their recovery from sites which exhibit local supplies of alternative raw materials. If thumbnail scrapers were developed for long-term use, why bother investing in their production when local stone is readily available at each site? Two possible scenarios may contribute to this apparent paradox. First, it is possible that the functional requirements of stone artefacts at these sites could not be met by local raw materials, necessitating the use of exotic raw materials whose expensive replacement costs were offset by long use-lives achieved through curatorial activities (cf. conservation/rationing models of Bamforth 1986; Hiscock 1986; 1988). Second, while thumbnail scrapers may have been discarded at sites with ready access to local stone, they may have been used in different site or off-site contexts where tool manufacture was limited by either lack of raw materials or time-stress constraints associated with food procurement (Bamforth 1991; Torrence 1983).

A further dimension to the thumbnail scraper curation hypothesis is provided by some of the western southwest Tasmanian sites (e.g. Kutikina and Mack 90/1) which exhibit thumbnail scrapers manufactured from quartz which is locally available. These artefacts represent a potential anomaly in terms of the curation hypothesis if they were both manufactured from local quartz (via-a-vis exotic quartz) and used locally. In this situation, however, it is likely that these thumbnail scrapers were used as temporary, low-cost, expedient replacements for more expensive thumbnail scrapers manufactured from exotic raw materials. People may have saved high-cost, imported thumbnail scrapers for use at sites where alternative local raw materials for thumbnail scraper manufacture were either rare or non-existent. The demonstration of local use of local thumbnail scrapers would provide compelling evidence that thumbnail scraper form was also constrained by the functional requirements of the tasks they performed on-site.

Summary and conclusion

This paper presents a model of chronological change for late Pleistocene southwest Tasmania which emphasizes the organizational aspects of stone artefact technology, particularly in terms of changes in mobility arrangements, raw material procurement, and site usage. In summary, the model posits that increased use of thumbnail scrapers across southwest Tasmania after the glacial maximum was largely a response to increasing mobility which placed extra demands for highly portable, long-term maintainable components of the tool-kit. These extra functional demands necessitated increased use of high-quality raw materials such as chert and quartz. Special cases of local manufacture and use of thumbnail scrapers are seen as a situationally expedient strategy for conserving expensive, exotic thumbnail scrapers. These expedient examples suggest that the manufacture of thumbnail scrapers was conditioned also by the tasks they performed. Thumbnail scrapers recovered from contexts pre-dating 18,000 b.p. suggest that these functional constraints on tool shape may have been operating for a long time.

Although this model reinforces Jones' conception of pre- and post-18,000 b.p. stone assemblages, it differs fundamentally to his social replacement explanation proposed for Kutikina. Jones states (1988: 37):

The typological differences between the two assemblages must be put down to different 'fashions' in tool design, reflecting different technological traditions. Although such views are no longer fashionable in contemporary archaeology, these would be two different 'cultures' in Gordon Childe's terminology, and they might possibly reflect two different ethnic groups or 'tribes in a general sense, the change occurring due to the abrupt intrusion of members of a second group into territory previously occupied by the first.

The alternative model of technological organization proposed in this paper not only is more testable than Jones' model, but also is flexible enough to cope with chrono-spatial variations in the use of diagnostic technological/typological markers such as the use of Darwin Glass and the manufacture of thumbnail scrapers. This flexibility results directly from an emphasis on changes in mobility arrangements which can occur at both intra- and inter-regional levels (Mc-Niven et al. in press). However, despite our recourse to differing theoretical/explanatory frameworks, my work has reaffirmed Jones' pioneering work in isolating the last glacial maximum as a turning point in stone artefact technology in the late Pleistocene of southwest Tasmania.

Acknowledgements. This study was made possible by funds supplied by the Australian Research Council and La Trobe University. Special thanks to Jim Allen for his support during all stages of this research. Critical and helpful comments on earlier drafts of this paper were provided by Jim Allen, Richard Cosgrove, Joanna Freslov, Rhys Jones, Brendan Marshall, Tom Richards, Nikki Stern, Robin Torrence, Cathie Webb and an anonymous reviewer. All errors in fact or interpretation are my responsibility.

References

ALLEN, J., R. COSGROVE & S. BROWN. 1988. New archaeological data from the Southern Forests region, Tasmania: a preliminary statement, Australian Archaeology 27: 75-88.

BAMFORTH, D.B. 1986. Technological efficiency and tool curation, American Antiquity 51(1): 38-50.

1991. Technological organization and hunter-gatherer land-use: a California example, American Antiquity 56(2): 216-34.

BINFORD, L.R. 1973. Interassemblage variability -- the Mousterian and the 'functional' argument, in C. Renfrew (ed.), The explanation of culture change: models in prehistory: 227-54. London: Duckworth.

1979. Organization and formation processes: looking at curated technologies. Journal of Anthropological Research 35(3): 255-72.

BROWN, S., S. KEE, A. MCGOWAN, G. MIDDLETON, M. NASH, B. PRINCE, N. RICKETTS & D. WEST. 1991. A preliminary survey for Aboriginal sites in the Denison River valley, March 1989, Australian Archaelogy 32: 26-37.

CLARK, J.D. 1959. The prehistory of southern Africa. Harmondsworth: Penguin.

COSGROVE, R. 1989. Thirty thousand years of human colonization in Tasmania: new Pleistocene dates, Science 243: 1706-8.

1991. The illusion of riches: issues of scale, resolution and explanation of Pleistocene human behaviour. Unpublished Ph.D thesis, La Trobe University.

COSGROVE, R., J. ALLEN & B. MARSHALL. 1990. Palaeo-ecology and Pleistocene human occupation in south central Tasmania, Antiquity 64: 59-78.

DEACON, H.J. & J. DEACON. 1980. The hafting, function and distribution of small convex scrapers with an example from Bloomplaas Cave, South African Archaeological Bulletin 35: 31-7.

ELLIS, C.J. & J.C. LOTHROP (ed.). 1989. Eastern Paleoindian lithic resource use. Boulder (CO): Westview Press.

FUDALI, R.F. & R.J. FORD. 1979. Darwin Glass and Darwin Crater: a progress report, Meteoritics 14: 283-96.

FULLAGAR, R. 1986. Use-wear and residues on stone tools: functional analysis and its application to two southeastern Australian archaeological assemblages. Unpublished Ph.D thesis, La Trobe University.

GALLAGHER, J.P. 1977. Contemporary stone tools in Ethiopia: implications for archaeology, Journal of Field Archaeology 4: 407-14.

GOODYEAR, A.C. 1989. A hypothesis for the use of crytocrystalline raw materials among Paleoindian groups of North America, in Ellis & Lothrop (1989): 1-9.

GOULD, R.A. & S. SAGGERS. 1985. Lithic procurement in central Australia: a closer look at Binford's idea of embeddedness in archaeology, American Antiquity 50(1): 117-36.

HISCOCK, P. 1986. Raw material rationing as an explanation of assemblage differences: a case study of Lawn Hill, northwest Queensland, in G. Ward (ed.), Archaeology at ANZAAS Canberra: 176-90. Canberra Archaeological Society, Department of Prehistory and Anthropology, Australian National University.

1988. Prehistoric settlement patterns and artefact manufacture at Lawn Hill, northwest Queensland. Unpublished Ph.D thesis, University of Queensland.

HOLMES, W.H. 1912. Peculiar stone industries of the Argentine coast -- continued, in A. Hrdlicka (ed.), Early man in South America: 125-51. Washington (DC): Smithsonian Institution. Bureau of American Ethnology Bulletin 52.

JESKE, R. 1989. Economies in raw material use by prehistoric hunter-gatherers, in R. Torrence (ed.), Time, energy and stone tools: 34-45. Cambridge: Cambridge University Press.

JONES, R. 1988. Hunting forbears, in M. Roe (ed.), The flow of culture: Tasmanian studies: 14-49. Canberra: Australian Academy of the Humanities.

1989. East of Wallace's Line: issues and problems in the colonization of the Australian continent, in P. Mellars & C. Stringer (ed.), The human revolution: behavioural and biological perspectives on the origins of modern humans: 743-82. Edinburgh: Edinburgh University Press.

1990. From Kakadu to Kutikina: the southern continent at 18000 years ago, in C. Gamble & O. Soffer (ed.), The world at 18000 BP 2: low latitudes: 284-95. London: Unwin Hyman.

KEELEY, L.H. 1982. Hafting and retooling: effects on the archaeological record, American Antiquity 47(4): 798-809.

LOTHROP, J.C. 1989. The organization of Paleoindian lithic technology at the Ports Site, in Ellis & Lothrop (1989): 99-137.

LOURANDOS, H. 1993. Hunter-gatherer cultural dynamics: long- and short-term trends in Australian prehistory, Journal of Archaeological Research 1(1): 67-87.

MCNIVEN, I., B. MARSHALL, J. ALLEN, N. STERN & R. COSGROVE. In press. The Southern Forests Archaeological Project: an overview, in M.A. Smith, M. Spriggs & B. Fankhauser (ed,), Sahul in review: Pleistocene archaeology in Australia. New Guinea and Island Melanesia. Canberra: Department of Prehistory, Research School of Pacific Studies, Australian National University.

NELSON, M.C. 1991. The study of technological organization, Archaeological Method and Theory 3: 57-100.

POCOCK, C. 1992. A report on the excavation of Maneena Langatick Tattana Emita (TASI 2867). A report to the Hydro-Electric Commission of Tasmania.

RULE, P. & J. EVANS. 1985. The relationship of morphological variation to hafting techniques among Paleoindian endscrapers at the Shawnee Minisink site, in C.W.W. McNett, Jr, Shawnee Minisink: a stratified Paleoindian-Archaic site in the upper Delaware Valley of Pennsylvania: 211-20. Orlando (FL): Academic Press.

SHOTT, M.J. 1986. Technological organization and settlement mobility: an ethnographic examination, Journal of Anthropological Research 42: 15-51.

1989. Technological organization in Great Lakes Paleoindian assemblages, in Ellis & Lothrop (1989): 221-37.

STERN, N. & B. MARSHALL. 1993. Excavations at Mackintosh 90/1 in western Tasmania: a discussion of stratigraphy, chronology and site formation, Archaeology in Oceania 28(1): 9-17.

TORRENCE, R. 1983. Time budgeting and hunter-gatherer technology, in G. Bailey (ed.), Hunter-gatherer economy in prehistory: a European perspective: 11-22. Cambridge: Cambridge University Press.