The tomb owner at the offering table: a pilot study on applying statistical analysis to dating Old Kingdom Memphite tombs

Theoretical thesis.Bibliography: pages [73]-86 of volume 1..Volume I. Text -- Volume II. Appendices.The (re-)construction of Old Kingdom Egypt requires a reasonable relative chronological framework, which has proven to be an elusive and often controversial endeavour. The analysis that has been done in tracing the evolution of tombs has in part been hampered by the necessity to use established dates, often controversial dates, to interpret and determine the stylistic development of tomb construction and decoration. While new technology has enhanced the study of ancient cultures through such applications as satellite archaeology, radio carbon dating, 3D imaging of mummies and other artefacts, the use of statistics and mathematical techniques has remained at a basic level. More recently new techniques have appeared, such as using cladistics for Nubian pottery analysis, decision tree analysis for Middle Kingdom Egyptian stele dating, and correlation analysis for the use of royal regalia at Medinet Habu.The study aims to determine if advanced mathematical methods, like cluster analysis, can refine the methods employed to date Old Kingdom tombs. IBM's SPSS package using Kohonen Network and TwoStep cluster analysis are applied to 113 Memphite examples of one scene type, the tomb owner seated before an offering table, using approximately three hundred features associated with the scenes on all walls of the offering chapels. The analysis, without reference to any currently considered dates, intends to determine if natural clusters exist and whether these correlate to currently accepted dates of individual Old Kingdom tombs. The initial results indicate that approximately 40-50% of the clusters fall within the "ranges" that had been proposed by various scholars. This study has indicated that using cluster analysis has some validity in dating tombs and results could be improved by selectively analysing smaller groups of criteria.Mode of access: World wide web1 online resource (2 volumes) black & white illustration