Scientific Reports (Mar 2024)

Automatic analysis of the continuous edges of stone tools reveals fundamental handaxe variability

  • Antoine Muller,
  • Gonen Sharon,
  • Leore Grosman

DOI
https://doi.org/10.1038/s41598-024-57450-y
Journal volume & issue
Vol. 14, no. 1
pp. 1 – 18

Abstract

Read online

Abstract The edges of stone tools have significant technological and functional implications. The nature of these edges–their sharpness, whether they are concave or convex, and their asymmetry–reflect how they were made and how they could be used. Similarly, blunt portions of a tool’s perimeter hint at how they could have been grasped or hafted and in which directions force could be applied. However, due to the difficulty in accurately measuring the complex 3D geometry of tool edges with traditional methods, their attributes are often overlooked. When they are analyzed, they have traditionally been assessed with visual qualitative categories or unreliable physical measurements. We introduce new computational 3D methods for automatically and repeatably measuring key attributes of stone tool edges. These methods allow us to automatically identify the 3D perimeter of tools, segment this perimeter according to changes in edge angles, and measure these discrete edge segments with a range of metrics. We test this new computational toolkit on a large sample of 3D models of handaxes from the later Acheulean of the southern Levant. Despite these handaxes being otherwise technologically and morphologically similar, we find marked differences in the amount of knapped outline, edge angle, and the concavity of their edges. We find many handaxes possess blunt portions of perimeter, suitable for grasping, and some handaxes even possess more than one discrete sharp edge. Among our sample, sites with longer occupations and more diverse toolkits possessed handaxes with more diverse edges. Above all, this paper offers new methods for computing the complex 3D geometry of stone tool edges that could be applied to any number of artifact types. These methods are fully automated, allowing the analysis and visualization of entire assemblages.

Keywords