The taming of dating precision and accuracy in 14 C-chronologies

This report suggests that ongoing debates over Levantine Early Iron Age chronology and the dating of the Santorini eruption largely arise from Bayesian overprecision, as defined by Peter Steier and Werner Rom (2000). In both cases, a chronological offset of about 60 years is traced to inappropriate calibration curve shape correction. We further show that distortions in Bayesian summed probability distributions (SPDs) – including spikes, artificial peaks, and systematic shifts – share the same cause. These issues cast serious doubt on the suitability of Bayesian SPDs for dates-as-data demographic analyses, supporting Timothy J. Heaton et al. (2025), whose concerns we share, though noting that such distortions do not occur in CalPal-generated SPDs. The SPD-based barcode seriation (BS) method avoids overprecision-caused distortions and automatically corrects for archaeological sampling bias.

Archaeomagnetic dating · Bayesian probability · Chronology · Offset (computer science) · Resampling · Seriation (archaeology) · Archaeological Research and Protection · Archaeology and ancient environmental studies · Geology and Paleoclimatology Research