A Site That Keeps Rewriting History
Monte Verde in southern Chile has already upended archaeology once. When Tom Dillehay first reported evidence of human presence there dating to roughly 14,500 years ago, the finding shattered the long-dominant Clovis-First model, which held that humans did not reach the Americas until around 13,000 years ago via a land bridge from Siberia. Now a fresh analysis of the same site is challenging the timeline again — this time by documenting a distinct occupation layer from the much more recent mid-Holocene period.
Published in Science, the new study by Todd Surovell and colleagues used cosmogenic nuclide dating, radiocarbon analysis, and high-resolution stratigraphy to identify a previously underappreciated human occupation spanning roughly 5,000 to 7,000 years ago — the mid-Holocene — at the MV-I component of Monte Verde, a layer above the famous MV-II pre-Clovis deposits.
What the New Dating Reveals
For decades, the MV-I layer was treated as geologically undifferentiated and assigned a broad Holocene age without much scrutiny, overshadowed by the far more sensational pre-Clovis evidence below it. Surovell's team applied cosmogenic exposure dating to stone artifacts and used Bayesian modeling of multiple radiocarbon dates to demonstrate that MV-I represents a genuine cultural horizon rather than a palimpsest of mixed deposits.
The assemblage recovered from MV-I includes bifacially worked stone tools, ground stone implements consistent with plant processing, and faunal remains reflecting a broad-spectrum subsistence economy. This contrasts sharply with the MV-II assemblage, which preserves wooden structures, mastodon bones, and the direct evidence of coastal foraging that made Monte Verde famous.
The mid-Holocene occupation adds a new chapter to the site's story: after the initial colonists arrived during the late Pleistocene, much of southern Chile may have been intermittently abandoned or sparsely populated during the early Holocene climatic optimum, only to be recolonized by culturally distinct groups thousands of years later.
Implications for South American Population History
The finding has significant implications for understanding the tempo and pattern of human dispersal across South America. Genomic studies of ancient and modern Indigenous populations have already revealed multiple distinct migration pulses into the continent, but the archaeological record has lagged behind in documenting those movements at specific localities.
Monte Verde's mid-Holocene layer now provides a concrete anchor point suggesting that the extreme south of the continent — Patagonia and Tierra del Fuego — experienced significant population turnover. The toolkit at MV-I resembles assemblages found at other mid-Holocene sites across the region, indicating that the recolonization was part of a broader southward demographic expansion, possibly driven by population pressure further north or by climatic shifts that made southern environments more productive.
Dating Methods Make the Difference
A key methodological contribution of the paper is its demonstration that cosmogenic nuclide dating of lithic artifacts can successfully resolve occupational history at sites where organic material is scarce. In arid and semi-arid settings across South America, radiocarbon dating is often limited by poor collagen preservation. Surface-exposure dating of stone tools offers a complementary chronometer that does not depend on organic material.
Researchers cautioned that the mid-Holocene age applies specifically to the MV-I horizon and does not alter the much older dates for MV-II. The two occupations are stratigraphically distinct, separated by a sterile layer of sediment representing a hiatus of several thousand years. Monte Verde thus documents two largely independent chapters of human settlement at the same location, separated by a long interval during which the site appears to have been unoccupied.
The study also challenges archaeologists to revisit other South American sites where Holocene layers have been treated as secondary to deeper deposits. Many multi-component sites may harbor discrete occupations that have been conflated into single analytical units because dating was insufficient or interpretive frameworks focused exclusively on the oldest evidence.
This article is based on reporting by Science (AAAS). Read the original article.
