The getting of data : a case study from the recent industries of Australia

Two obstacles stand in the way of satisfactory synthesis of Australian prehistory at more than a very general level. In the first place, excavation techniques have grown up to come to grips with the difficult piecemeal and have conditions posed failed by the undifferentiated sandy stratigraphies and low sedimentation rates typical of many Australian sites. In The second place, the piecemeal genesis of excavation techniques is reflected in the lack of established channels for the exchange of excavation data and in incompatibility between the information recorded by different workers. The first part of this thesis examines the prehistory of the Blue Mountains area, starting with a general review and leading to a more detailed study of the Noola and Capertee 3 sites. The study of these two sites suggests that the model of a site as a chronologically continuous sample may not be applicable even to sites with rich assemblages and that horizontal patterning is an important feature which must be taken into account when designing site sampling strategies. New dating of Capertee Site 3 confirms that backed implements appear in the site at about 3000BP. Following on from the Capertee 3 dating, a critical review of the literature relating to the appearance of backed implements and of other traits associated with the Australian Small Tool Tradition demonstrates that no dates older than about 5000BP have yet been substantiated for Small Tool Tradition assemblages, however it has not proved possible to tie the dating down accurately owing to the lack of data and imprecision of associations between dated samples and artefact assemblages. The concluding chapters are aimed at resolving some of the problems identified in the preceding discussion. Attention is focussed on methods of data recording, since standardisation of this aspect of excavation not only helps establish a framework for excavation onto which individual strategies can be grafted, but represents a first step in the direction of greater comparability between data collected by different workers. The discussion is completed by the documentation of a computerised system for the recording and analysis of excavation data. The use of such a system facilitates the analysis of the material and also generates a catalogue which can be duplicated for storage with the collection. It is suggested that future developments should be aimed at standardisation of basic classificatory and attribute systems in order to promote comparability between collections and thus provide a better information base for the synthesis of Australian prehistory.