Neither use nor ornament: a consideration of the evidence for the existence of a system of communication and notation in the European upper Palaeolithic

Abstract

This thesis is concerned with devising an objective means of testing whether or not the people of the European Upper Palaeolithic used systems of notation to create intentional records. I begin with a discussion of the theoretical parameters that I intend to adopt, notably an adaptive rather than a progressive model of change over time, and the need for a controlled use of analogical comparisons between present phenomena and prehistory. By briefly summarising the approach of past researchers in this area, I identified engraved mobiliary artefacts as the primary source of data. In order, to construct an objective model to test the hypothetical existence of notation, I first examine other forms of visual communication, notably iconicity, decoration and the use of signs and symbols. I conclude that any recognisable examples of notation would be structured in order to effectively convey information. As a result I devise a series of criteria based on first principles by which I hope to be able to distinguish between examples of representational, decorative, notational and purely random accumulations of lines. To test the effectiveness of these criteria, I apply then to examples of known systems of visual representation taken from the Ethnographic collections of four museums. These include Australian message sticks, tallies, records of time and calendars. I conclude that my criteria are able to discriminate between examples of notation and other classes of meaningful representation. I also isolate the recurrent use of translational symmetry and rotational symmetry in notational artefacts, and the absence of mirror and slide refection. By discussing in some detail the approach of past researchers to the identification of notation in engraved mobiliary artefacts from the Upper Palaeolithic I isolate certain practical and theo-reucal problems. To correlate my results with those of Marshack I apply my criteria to five of his selected examples and conclude that two out of the five are not notational by my criteria. I also test my criteria against some artefacts from the Piette and Saint-Perier collections in Paris and conclude that the structure of the marks on some examples conform to my criteria of notation. Marshack is criticised for claiming to identify evolving systems and traditions of notation using only a handful of examples from sites all over Europe and the timescale of the Upper Palaeolithic. I counter this defect by looking at a sample of data taken from a single site, Enleue in Ariege. The engraved artefacts date to Perigordian V and the Early and Middle Magdalenian and are backed by independent dating techniques and a wealth of complementary finds. By providing a detailed summary of the characteristics of this data set I was able to identify certain common characteristics of non-figurative engraved mobiliary artefacts. By measuring the physical properties of the lines on the complete data namely their length, distance and orientation, I demonstrate the presence or absence of order and structure using Multivariate Cluster Analysis and Two- and Three-Dimensional plots. The cases are selected from Marshack's data, my Paris data, the complete examples from Enlene, with known examples of notation and randomly generated examples to act as controls. As a result of the identification of orientation and then inter-line distance as the prime indications of artefact variability, the broken bones from Enlene are included. Further statistical tests isolate certain artefacts as examples of low variabilty, which are confirmed by my criteria to be the most likely examples of notation or decoration at Enlene.