This page describes an analysis of Palaeolithic artefacts from a north African site. The purpose of the analysis was to examine the nature of change across the Middle Palaeolithic / Upper Palaeolithic boundary. This project develops further the kind of reduction analysis of the Upper Palaeolithic assemblage previously described in Hiscock, P. 1996 Transformations of Upper Palaeolithic implements in the Dabba industry from Haua Fteah (Libya). Antiquity 70: 657-664.

Haua Fteah is a large cave on the north coast of Libya, formed in Tertiary limestone. The floor area of the cave is approximately 40m by 85m, with a ceiling about 20m high at the entrance. Charles McBurney from Cambridge University located the cave in 1948, when undertaking a geological survey, and returned to excavate in 1951. McBurney excavated down 12.5m without reaching the base of the cultural deposit (see photograph to the right). His extensive analysis of the assemblages was published as: McBurney, C. 1967. The Haua Fteah (Cyrenaica) and the stone age of the south-east Mediterranean. Cambridge: Cambridge University Press.

Artefacts analysed in this project came from the lower levels of the site, described by McBurney as Middle Palaeolithic and Dabba (= Upper Palaeolithic). The Middle Palaeolithic phase was identified from layers XXXV to XX, terminating about 45-40,000 years ago. Assemblages that McBurney assigned to this phase contained handaxes, scrapers, burins and 'points', predominantly made using levallois processes. Dabba begins in layer XX, at a date McBurney thought was 40,000±2,000 years B.P. His depiction of this industry is of collections of implements numerically dominated by backed blades, (making up c.50%), with burins being common (25%), and scrapers and other items (eg. chamfered blades) making up the rest. McBurney described the Dabba as "...industrial assemblages are basically similar to the Upper Palaeolithic and Mesolithic of south-west Asia, Europe and the Maghreb."

The artefact assemblages are stored in metal (post-war chocolate biscuit) boxes in the Cambridge University Museum of Archaeology and Anthropology (see photograph to the right). While the artefacts in most boxes have been extensively studied, and re-packaged, McBurney's original classification is still visible because the specimens illustrated in his book are still clearly labelled.

A sample of artefacts from excavation units representing the Upper Palaeolithic and Middle Palaeolithic levels were selected. Approximately 500-600 artefacts from each of these levels were studied. All of these artefacts are made on a high quality brown or grey chert. Small nodules of the chert are available in the local area. Selected specimens were removed from storage and studied in the Museum's laboratory.

In the lab, (shown above), the artefacts were sorted and measured. Artefacts were inspected under a hand lens, and weighed on digital scales. Dimensions were measured using digital calipers. Additionally artefacts were photographed with a Canon PowerShot 350 digital camera, which has a minimal focal distance of 30mm and provides an image of each specimen at moderate resolution (640 x 480 x 24 bit). These data were entered into a database on a notebook computer.

The software selected for this project was Lotus Approach 97, a flexible but powerful relational database running under Windows 95. The database structure, and a feel for data entry, can be illustrated with a few screen shots of records:

Information about provinence, typology, and the completeness of the specimen are entered on the introductory form. The program automatically assigns a unique identifier (HFSPECNO) to each specimen. Macro-buttons enable new entry screens to be brought up for different categories of data.

Basic information about flake size and shape was collected using the percussion length, width and thickness, plus platform dimensions. Measurements of flake thickness and the thickness of retouch scars discussed by Kuhn (1990 Journal of Archaeological Science 17:583-593) were taken on all retouched flakes. As discussed by Kuhn these measurements can be used to create an index of the extent of retouching. The Approach program automatically calculates this and other indices (see below).

Each object was photographed using the digital camera, and a thumbnail copy of the digital photograph inserted in the database (scale is 1cm units). This image was a handy reference, and was particularly useful in relocating unlabelled specimens. More importantly, the photograph was annotated with information, such as the location of retouch (shown by the red line), or the position of the striking platform. This information was incorporated into the image analysis carried out on the photograph.

As noted above, the program calculated a number of indices during data entry. This information was displayed on a separate form. Indices include typical measures of flake size and shape, the reduction index of Kuhn, and other measures of backed artefact shape.

Data is exported from the database and imported into a number of specialist programs. For example, information from the database is imported into SPSS (Statistical Package for the Social Sciences) for statistical analysis. The result is a spreadsheet-like display of the information shown above.

Using SPSS the patterns in the assemblages are explored and compared. This "box and whiskers" graph is an example of images that are obtained from the database.

For results of the project thus far see: Hiscock, P. 1996 Transformations of Upper Palaeolithic implements in the Dabba industry from Haua Fteah (Libya). Antiquity 70: 657-664.

Acknowledgments:

The examination of the collections was carried out during June and July 1998. Once again I am indebted to Chris Chippindale, Robin Boast and the staff of the Cambridge University Museum of Archaeology and Anthropology for providing access to the Haua Fteah collections.

This research was funded by a "New Starters" research grant from the Australian National University. Equipment was provided by the Department of Archaeology and Anthropology, Australian National University.

Author: Peter Hiscock, Dept. Archaeology and Anthropology
Feedback: peter.hiscock@anu.edu.au .
Date Last Modified: 9-July-98
URL: http://artalpha.anu.edu.au/web/arc/resources/haua/hfanalysis.htm