Current Methodological Considerations in the Application of Two-Dimensional Geometric Morphometrics within Handaxe Assemblages

Summary

In the construction of past hominin behaviour, frameworks emphasise the role of morphological variation within a particular artefact class, as deviations and commonalities in shape exemplify conscious/unconscious decisions by past populations and individuals. For early prehistory this is best illustrated through discussions of shape variance, and its role in understanding cognition and aspects of cultural transmission. In the formal categorisation of handaxe shape variance, within a statistical framework, recent methodologies have utilised two-dimensional geometric morphometric frameworks through Fourier-based transformations or semi-landmark (sensu Bookstein) analyses.

We examine an essential methodological protocol, under-looked in literature, and a potential source of shape-error, specifically the orthogonal projection of a three-dimensional handaxe into a two-dimensional shape-space. Various technological and morphological rationale have been adopted in determining which planform profile should be digitised and analysed, however the effect each of these different orthogonal projection methods have on a control dataset is unknown. To what extent do rationale influence the main sources of shape variation, the underlying statistical output (e.g. PCA scores or Procrustes distances), and the ability to discriminate between different groups?

Using an experimental dataset of known-sided handaxes, we conclude that issues of orthogonal projection are essential within a priori considerations of handaxe shape analysis.

This Resource is Part of the Following Collections

• Society for American Archaeology 82nd Annual Meeting, Vancouver, BC (2017)

Keywords

Record Identifiers

Abstract Id(s): 16200