Statistical Inference of Prehistoric Demography from Frequency Distributions of Radiocarbon Dates: A Review and a Guide for the Perplexed

Abstract

The last decade saw a rapid increase in the number of studies where time–frequency changes of radiocarbon dates have been used as a proxy for inferring past population dynamics. Although its universal and straightforward premise is appealing and undoubtedly offers some unique opportunities for research on long-term comparative demography, practical applications are far from trivial and riddled with issues pertaining to the very nature of the proxy under examination. Here I review the most common criticisms concerning the nature of radiocarbon time–frequency data as a demographic proxy, focusing on key statistical and inferential challenges. I then examine and compare recent methodological advances in the field by grouping them into three approaches: reconstructive, null-hypothesis significance testing, and model fitting. I will then conclude with some general recommendations for applying these techniques in archaeological and paleo-demographic research.

Human Paleodemography and Paleoecology of the North Pacific Rim from the Mid to Late Holocene

“…Over the last two decades, archaeologists have invested considerable time, resources, and effort in aggregating large radiocarbon databases, allowing us to better address the problem of random sampling error (Rick, 1987; Williams, 2012; Brown, 2015, 2017; Fitzhugh et al, 2016; Crema et al, 2016, 2017; Robinson et al, 2019; Kelly et al, 2021; Price et al, 2021; Shennan and Sear, 2021; Crema, 2022). These efforts to aggregate and archive data in digital repositories and targeted attempts to compile archaeological radiocarbon databases at up-to-continental scales (e.g., Canadian Archaeological Radiocarbon Database /CARD: ; Kelly et al, 2021; Palmisano et al, 2022) have made it possible to compile TFDs of dates for many regions of interest.…”

Bayesian Inference of Prehistoric Population Dynamics from Multiple Proxies: A Case Study from the North of the Swiss Alps

“…While the methodology continues to evolve and address these critiques (Crema, 2022), it remains subject to fundamental problems common to all approaches relying on a single proxy Schmidt et al (2021). We believe that these problems cannot be overcome by methodological refinements in this area alone.…”

“…In particular, the 'dates as data' technique (Rick, 1987), using the frequency of radiocarbon dates as a proxy for population dynamics, has been significantly developed in the last decade (e.g. Shennan et al, 2013) and widely applied to archaeological contexts worldwide (Crema, 2022). This approach has contributed greatly to our understanding of prehistoric demography, but is not without its critics Carleton and Groucutt (2021).…”

Soil fertility depletion is not a credible mechanism for population boom/bust cycles in early agricultural societies

“…More comprehensively (but also more controversially) population booms and busts are evidenced in summed probability distributions of human presence P a g e | 2 aggregated from radiocarbon databases (Shennan et al, 2013;Timpson et al, 2014;Palmisano et al, 2021a). While the reality of population declines suggested by this data source is a matter of an ongoing scientific controversy (Torfing, 2015;Crema, 2022), there is a substantial degree of agreement with more traditional archaeological indicators (Downey et al, 2014;Palmisano et al, 2017;Hinz et al, 2022). Overall, it is becoming increasingly clear that, even if population declines following initial population increases in early farmers are not universal (we currently don't have enough data to make such a case), they are certainly a frequent occurrence, documented in different parts of the world (Kim et al, 2021;Kohler et al, 2020;Palmisano et al, 2021b).…”