The Perks and Perils of Replicability

Abstract Following a trend across the sciences, recent studies in lithic analysis have embraced the ideal of replicability. Recent large-scale studies have demonstrated that high replicability is achievable under controlled conditions and have proposed strategies to improve it in lithic data recording. Although this focus has yielded important methodological advances, we argue that an overemphasis on replicability risks narrowing the scope of archaeological inquiry. More specifically, we show (1) that replicability alone does not guarantee reliability, interpretive value, or cost effectiveness, and (2) that archaeological data often involve unavoidable ambiguity due to preservation, analyst background, and the nature of lithic variability itself. Instead of allowing replicability to dictate research priorities, we advocate for a problem-driven, pluralistic approach that tailors methods to research questions and balances replicable measures with interpretive depth. This has practical implications for training, publishing, and funding policy. We conclude that Paleolithic archaeology must engage with the replicability movement on its own terms—preserving methodological diversity while maintaining scientific credibility.