3.3 million years of stone tool complexity suggests that cumulative culture began during the Middle Pleistocene

Abstract

Cumulative culture, the accumulation of modifications, innovations, and improvements over generations through social learning, is a key determinant of the behavioral diversity across Homo sapiens populations and their ability to adapt to varied ecological habitats. Generations of improvements, modifications, and lucky errors allow humans to use technologies and know-how well beyond what a single naive individual could invent independently within their lifetime. The human dependence on cumulative culture may have shaped the evolution of biological and behavioral traits in the hominin lineage, including brain size, body size, life history, sociality, subsistence, and ecological niche expansion. Yet, we do not know when, in the human career, our ancestors began to depend on cumulative culture. Here, we show that hominins likely relied on a derived form of cumulative culture by at least ~600 kya, a result in line with a growing body of existing evidence. We analyzed the complexity of stone tool manufacturing sequences over the last 3.3 My of the archaeological record. We then compare these to the achievable complexity without cumulative culture, which we estimate using nonhuman primate technologies and stone tool manufacturing experiments. We find that archaeological technologies become significantly more complex than expected in the absence of cumulative culture only after ~600 kya.

Keywords: cultural evolution; cumulative culture; human evolution; imitation; stone tools.

Fig. 1.

Count of procedural units (n = 64) as a function of time. The line represents the mean of the statistical model posterior predictions, and the shaded ribbon is the 80th percentile interval. Posteriors are sampled from the global grand mean with no site-specific deviations. The horizontal dash line at PU = 6 marks the most complex nonhuman primate technology, i.e., the brush-tipped termite probe production reported among chimpanzee groups (29). Other nonhuman primates and human experiment values fall below this line (SI Appendix, Table S2).

Fig. 2.

Top panel. The total number of tool and core types present across assemblages is summarized using the technological Modes A-I system (, –75) (N = 1,192). Middle panel: The efficiency of stone tool technology is measured as the ratio between edge length and flake size. Data represent the median ratio among N = 81 assemblages (76). Bottom panel: Hominin behavioral timeline (, , –79).