Monte Verde, Revisited (Again): What the Rebuttals Say

No archaeological fact is truly ever a fact; those were the words I used to introduce you to a new study on the dates of Monte Verde II on Medium a few months ago. At the time, the study was circulating and making headlines all over the place.

The study by Todd Surovell re-examined the Monte Verde II site in Chile, long considered the best and among the oldest pieces of evidence that humans had reached South America over 14,500 years ago; Surovell's team claimed it was actually probably only around 8,200 years old.

Reading the study, it seemed legit: the team revisited the site, re-dated the layers, identified a volcanic ash layer that supported their hypothesis, and provided explanations for why the site wasn't really that old.

Well, since I wrote that piece, the original excavation team of Monte Verde II, led by Tom Dillehay and his colleagues, as well as geologists led by Michael Waters and geneticists led by David Meltzer, fired back. As it turns out, Surovell's work was perhaps not as rigorous as we (myself included) had all thought. The rebuttals are a little old by now, as they flew under the radar, but that doesn't mean they shouldn't be covered. So I want to walk you through what Dillehay and his colleagues had to say.

They Never Even Dug There

Before we get into it, just a quick recap: Monte Verde II is located in the Chinchihuapi Creek area of Chile. It was first excavated between 1977 and 1985 by Dillehay and colleagues, yielding a whole plethora of remains, including lithics (stone tools), plant materials such as tools and wild potatoes, and animal remains, including extinct megafauna such as mastodons.

The original team dated the site to around 14,500 years old, providing one of the best pieces of archaeological evidence that humans reached South America over 14,000 years ago.

When Surovell's team returned to the site and published their study in 2026, they argued that the date was likely wrong. They based their argument on new radiocarbon and luminescence dating as well as the discovery of a volcanic ash layer known as Lepué Tephra. This was a volcanic eruption that occurred around 11,000 years ago, resulting in many South American sites having a characteristic ash layer, which basically acts as a dating marker, where things above the layer are younger than 11,000 years and things below are older.

The dates and volcanic ash layer suggested that the site was actually much younger, around 8,200 years ago. How did this happen, according to Surovell? A flood of the nearby Chinchihuapi Creek had washed and mixed wood and animal vertebrae from a higher terrace into Monte Verde II, making them seem artificially older

But Dillehay and his team strongly dispute these claims. While Dillehay and his colleagues spent well over a decade excavating the site, Surovell and his team spent no more than a few hours at the site sampling an area that wasn't even the location of the original excavation (a terrace they called T4).

The argument is simple: imagine you need to figure out the age of your home's foundation, so you hire a team, and they decide to sample your neighbor's backyard. They come back and tell you, "Well, based on your neighbor's garden, your foundation is actually only X years old."

That, Dillehay's team contends, is basically what Surovell did. Which means they couldn't possibly study the site's true stratigraphy.

Instead, Surovell's team is said to have cherry-picked disturbed, messy, exposed creek-bank samples nearby, which aren't comparable to the original excavation.

And what about the volcanic layer, the smoking gun of the original study?

Was the Volcanic Layer Ever There?

If a Lepué Tephra layer did exist at Monte Verde II, the logic should still hold that things above that layer are younger, and things below are older. But not if that layer was never there to begin with. According to Dillehay's team, decades of their research show a clear, well-established layering at the site, with the archaeological materials sealed under a peat layer (MV-5) securely dated to ~14,500 years ago.

Then why or how did Surovell think there was a volcanic ash layer? Basically, Surovell's team compared similar-looking soil layers from T4 to the site's peat layer and treated them as if they were the same layer formed at the same time, but they formed at different times in different locations. Crucially, two of the three layers Surovell sampled weren't even ash layers; they were a fungal deposit and an iron oxide mineral layer, while the third layer is a younger volcanic deposit rather than the Lepué Tephra.

That means the stratigraphic link Surovell's team used to argue for a younger site date is, according to Dillehay's team, mistaken.

And if that wasn't enough, the final nail in the coffin, for them, is that Surovell's team invented a new layering scheme (called "SU2") that was never part of the actual recognized site sequence, forced it into the site's stratigraphy, and implied their dated samples came from above and below the archaeological layer. Surovell's dated sequence, they say, doesn't even make logical sense, with numbers jumping around: 4,200 years here, 7,200 years just below it, then back to 4,900 years, then suddenly 13,100 years right above the human occupation layer.

Now stratigraphies can get messed up, with older layers and younger layers all over the place. This can happen if an animal burrows into the layers, or a flood disrupts a sequence. Which brings me to Surovell's redeposition theory. According to Surovell, artifacts and wood washed downstream from the T4 terrace and were redeposited in Monte Verde, making it seem older than it is.

But that argument, they say, doesn't hold up; for one, the wood on T4 doesn't match the type or size found at Monte Verde II. Second, the T4 terrace shows that the environment was cold and barren, not a forested landscape as it is today, making it unlikely that there was ever enough large wood available there to be washed downstream. While Waters' team argues that Surovell's team never proved wood or animal bones were washed into Monte Verde II, they just assumed it. They reported no fossils in T4 (so what exactly was being washed in), and on top of that, the Patagonian Forest formed ~15,100–15,700 years ago in the area, and could have been the source of the wood found in Monte Verde II.

On top of that, of the two other archaeological sites found nearby, neither shows any signs of flooding.

Finally, Dillehay's team directly challenges Surovell's claim about radiocarbon-dated soil cores taken near the site, saying Surovell's team wrongly assumed these cores came from the T4 terrace and that they corroborate their results when they actually came from the T2 terrace (where the Monte Verde II site itself sits),  meaning (if correct) they support the original late Pleistocene dating of the site.

What May Have Happened at Monte Verde II

To support the site's age, Dillehay and his team point to various lines of evidence, including bones from extinct species that showed signs of human manipulation, which wouldn't have been around 8,200 years ago. This means either humans were intentionally working on bones already thousands of years old (unlikely), or they were around at the same time as these extinct animals and deliberately shaped their bones into tools after butchery.

As for the flood hypothesis, have you ever seen pebbles in rivers and on beaches? What do many of them have in common? They have been smoothed by constant water action. This smoothing, rounding, and polishing wasn't evident on any of the bones at Monte Verde II, thus suggesting no flood had ever occurred there.

Similarly, the site contains various wooden tools, storage pits, hearths, plant remains, and even human footprints, all in situ (in their original depositional context), which would have been swept away had there been a flood.

And finally, among the originally radiocarbon-dated material are fragile chewed plant material, including seaweed and aromatic leaf, all dated to 14,200-14,600 years ago, which would not have survived a flood event.

Genetic Proof beyond Monte Verde II

I only briefly glossed over this at the time, but technically even if Surovell had been correct and Monte Verde II was not as old as assumed, this would not have meant there was no possibility of other lines of evidence, or sites existing that were as old or older in South America. Which brings me to Meltzer's genetics team.

Genetic models have been used to show when humans arrived in various parts of the Americas, providing evidence completely independent of Monte Verde II. Based on these lines of evidence, mitochondrial DNA (inherited through the maternal line) suggests that Native American ancestors split off from Asian populations about 24,900 years ago. Distinct founding lineages were established by around 18,400 years ago, and major population expansion is evident by around 16,000 years ago.

Y-chromosome DNA (inherited through the paternal line) points to similar divergence around 18,900 years ago and expansion by 15,000 years ago. These studies have shown that there are three deep genetic Native American branches: the ancient Beringians (who split off around 20,900 years ago) and two branches that split into northern and southern Native Americans around 15,700 years ago. For that split to occur, a southern population had to exist pre-Clovis, regardless of whether Monte Verde II was occupied at the time or not.

Meltzer's team also points to direct genetic evidence south of the ice sheets: ancient human DNA recovered from ~14,600-year-old fossilized feces at Paisley Caves, Oregon, and early American dog lineages that split off and expanded around the same time, both independent of Monte Verde entirely.

So, what does this mean? It means when Surovell's study first came out, it looked rigorous, detailed, and compelling, and I wanted to cover it. But if the rebuttals hold up, there may be serious problems with that study.

Archaeology is all about self-correction and being willing to challenge even our own facts. It's kind of how a hypothesis works: you set out to challenge your hypothesis, not to prove it.

It's happened many times before; for example, the dim-witted Neanderthals, a long-held scientific fact that we now know was wrong, after better excavation techniques, dating methods, and systematic analysis revealed they were probably just as intelligent as us.

Of course, sometimes claims change, such as the case of the ‘Beachy Head Woman’ being of African ancestry based on her skull features when DNA revealed she was a local Briton.

I'll be keeping the original Medium story up too, complete with a link to this piece and and a correction of the facts that didn’t hold up, because quietly burying a mistake isn't any more honest than making one in the first place.

Surovell's team has published replies of their own and this exchange is clearly not over.

Which side do you find more convincing after reading both arguments, Surovell's re-dating or Dillehay's defense? And have you ever changed your mind about something you thought was "settled" archaeology?

Let me know what you think, and if you'd like to support this blog and the person writing it, you can chip in once, monthly, or yearly. Only if you'd like to.

References

Dillehay, T. D., Pino, M., Lara, L. E., Abarzúa, A. M., Pérez-Balarezo, A., Davis, L., ... & Sierpe, R. (2026). Geomorphological and archaeological evidence at Monte Verde II, Chile supports the claim of human occupation 14,500 years ago. Science. e-Letter. https://www.science.org/doi/10.1126/science.adw9217#elettersSection

Meltzer, D. J., Moreno-Mayar, J. V., Pinotti, T., Heintzman, P. D., Pedersen, M. W., & Willerslev, E. Genetic evidence and the peopling of the Americas: reply to Surovell et al. 2026. Science. https://fundacionmonteverde.cl/wp-content/uploads/2026/05/Meltzer-et-al-Rebuttal-to-Surovell-et-al.-2026.pdf

Todd A. Surovell et al., A mid-Holocene age for Monte Verde challenges the timeline of human colonization of South America. Science 391,1283–1288(2026).DOI:10.1126/science.adw9217

Waters, M. R., Halligan, J. J., Mandel, R. D., Holcomb, J. A., Davis, L. G., & Holliday, V. T. (2026). Geoarchaeological Assessment of the Suggested Middle Holocene Age for Monte Verde II, Chile. https://liberalarts.tamu.edu/csfa/wp-content/uploads/sites/14/2026/05/Three-Critiques.pdf