Schedule by Session

Archaeoastronomers have been primarily interested in alignments between structures and celestial objects such as the sun, moon and stars (Ruggles 2011). In cases where there is textual or ethnographic evidence that complements the archaeological record, these can support or falsify any observations, hypotheses or speculations regarding celestial alignments (Ruggles 2015). In the case of prehistory, however, archaeoastronomer’s methodologies have focused on two pillars: field measurements of orientation and the statistical analyses of orientation data for a large number of structures (Ruggles 1999). The reasoning is that if a statistically significant pattern in orientation matching a celestial object is found, then one is on stronger ground to claim that it was intentional and not due to chance alone. However, even though measurements of orientation are the primary form of data used by archaeoastronomers, very little measurement theory has been taken on-board.

This paper will consider the principles and theory of measurement (Krantz et al 1971) as well as how measurement in handled by the experimental sciences (Taylor 1967). In particular, it will highlight the role of measurement uncertainty for inferential purposes and the extreme dangers of underestimating or ignoring it. Finally, a likelihood-based approach to inference (Edwards 1992), that uses full uncertainty estimation, will be presented, applied to a couple of case studies, and its merits discussed.