When multiple archaeomagnetic samples are collected from the same site or component, it is tempting to look for evidence of temporal differences between the collected features. This paper addresses the issue of the practical limits to confidently identifying temporal differences based on archaeomagnetic sample information. Three controlled data sets suggest that when the specimen polar projections of the samples compared have substantially overlapping ranges and when sample mean projections are less than five degrees apart, statistical comparisons do not support conclusions about probable age differences. These results suggest that temporal differences less than 50 to 100 years (depending on the rate of directional change in the geomagnetic field) between features within a site or component will be difficult to confidently identify. Additional analysis leads to the hypothesis, however, that the averaged results from sets of archaeomagnetic samples might provide the basis for making fine scale temporal distinctions between sites or components from which multiple archaeomagnetic samples have been collected.
INTRODUCTIONWhen several archaeomagnetic samples are collected from the same site or component, archaeologists are tempted to look for evidence of temporal differences among the collected features. Since it can be shown that, theoretically, accurate dating can be as precise as 2 5 years (Tarling, 1985:248; Wolfman, 1984:395 -396), the search for temporal differences is all the more tempting. Clearly, the practical limits on accurate dating will be different from the theoretical limits on precision (Tarling, 1985:285), but few attempts have been made as yet to describe these practical limits. Based on the limits to the precision with which archaeomagnetic secular variation curves can be derived and calibrated mathematically, Sternberg (1982: 105) feels that minimum date ranges will be 50 years in length. This paper focuses primarily on archaeomagnetic sample data, rather than theory, to see whether temporal differences less than 100 years can currently be expected.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.