replying to L. Carmignani et al. Nature Ecology & Evolution https://doi.org/10.1038/s41559-024-02548-9 (2024)

Carmignani and colleagues¹ contend that our identification of the oldest and easternmost Initial Upper Palaeolithic (IUP) attribution at Shiyu, northern China² is based on a misuse of technological definitions and biased artefact sampling. However, this criticism is hampered by restrictive Eurocentric definitions and methodological misunderstandings. Carmignani et al.¹ apply a narrow definition of the Asian IUP based on the presence of points and blades reduced from sub-volumetric, non-Levallois-system burin cores. Shiyu does not perfectly conform to this definition, so they exclude its lithic industry from the IUP. We argue that this definition hampers understanding of modern human dispersals across Eurasia. Like any other discipline where classification is central, archaeologists define categories of objects and assemblages based on similarities in sets of traits. These categories are not inherently meaningful; rather, their significance arises from assumptions about underlying biological and cultural processes. The assumption that lends meaning to the IUP is that it reflects a combination of demic and cultural diffusion processes associated with the spread of Homo sapiens throughout Eurasia³. To understand this evolutionary history, we need to adopt a flexible comparative approach that uses the original IUP definition as a baseline rather than a set of strict, restrictive criteria. A more flexible approach may help to disentangle cultural innovations and diffusion occurring during or after demic diffusion, as well as cultural blending with local populations on a regional scale, thereby illuminating the complex interactions that certainly occurred, with possible implications for gene exchange. Success in this endeavour depends on being able to distinguish diffusion of innovation from cultural convergence. Opening up our definitions admittedly creates greater uncertainty about the drivers of similarity and the meaning that underlies cultural constructs^4,5, but there are good reasons for taking this risk. Can we reasonably believe that in a territory of 44 million km², stretching from the Arctic to the Equator, the expansion of H. sapiens over thousands of years, encountering different hominin species, was associated with a single, uniform technology? It is clear in this context that the strict application by Carmignani et al.¹ of a points, blades and volumetric burin core definition, possibly useful in characterizing the IUP in one region, will fail to capture the overall complexity of H. sapiens’ expansion in others. This strict Eurocentric definition and interpretive framework does not pay attention to regional traditions in which cultural novelties occur, possibly signalling the dispersal of H. sapiens populations. Carmignani et al.¹ suggest that Shiyu lithics share features of the European Mousterian. Implying relationship to lithic industries produced by Neanderthals discounts any element of novelty at Shiyu that could link to H. sapiens dispersal, as was suggested by previously discovered human remains at the site⁶.