On the Entity and Antiquity of the Aurignacian at Willendorf (Austria): Implications for Modern Human Emergence in Europe

The time of the Aurignacian’s first appearance in the archeological record lies at the heart of debates on the emergence of European anatomically modern humans. Based on a re-study of Archeological Horizon (AH) 3 of Willendorf II, it has been claimed that the Early Aurignacian was present in the loess plains of Lower Austria by 43.5 ka (thousands of calendar years ago), several millennia earlier than in western Europe. The claim rests on the argument that a refit set linking an area excavated 2006–2011 with another excavated 1908–1909 implies that the two stone tool collections derive from a single, homogeneous assemblage. Therefore, the dating of the 2006–2011 context would also date the Early Aurignacian diagnostics found in the 1908–1909 collection. Based on the published evidence, this argument cannot be supported. The 1908–1909 excavation extended way beyond the boundaries of the stratigraphic level in which AH3 was identified in 2006–2011: lens C8-3 of subunit C8. Northward, subunit C8 becomes internally undifferentiated and merges with subunit C7 above, within which a new horizon, AH3ab, was first formally recognized in 2006–2011. This evidence implies that “AH3” of 1908–1909 was a stratigraphically heterogeneous unit over at least half of the area then excavated and that the stone tools it yielded must be treated as a multi-component assemblage that conflates material derived from at least two different occupation horizons. In line with the chrono-stratigraphy of Europe’s Early Upper Paleolithic sequence, the few Early Aurignacian diagnostics found in the “AH3” collection of 1908–1909 must date to ca.39.1 ka, the calendar age of AH3ab. The single refit set linking the two collections shows that an earlier component, dated to ca.43.5 ka by the 2006–2011 work, is also represented in the collection from 1908 to 1909. However, in the absence of diagnostics, the technocomplex affinities of that earlier component cannot be ascertained. The association of the Early Aurignacian with modern humans in Lower Austria remains a legitimate inference but is valid for the ca.38–40 ka time range, not before.