Arroyo del Vizcaíno, near Sauce, Uruguay
Arroyo del Vizcaíno
About 30,000 radiocarbon years (cut marks on sloth bones, or marks the stream made)
In the bed of a stream near Sauce, thousands of bones of extinct animals, most of them the giant sloth Lestodon, lie in a layer about 70 centimetres thick. Twelve radiocarbon dates cluster around 30,000 radiocarbon years ago, more than 33,000 once calibrated. Some grooves on the bones are published as cuts from stone tools, and a 2022 critique publishes them as the ordinary damage of a stream. A handful of stones are published as tools, and as natural pieces.
Why this is extraordinary
The Arroyo del Vizcaíno is a small stream near Sauce, in Canelones. In its bed, packed into a layer about 70 centimetres thick, are the bones of animals that are no longer alive. Most of them are Lestodon armatus, a giant ground sloth. More than 2,000 specimens have been lifted, and the diggers think thousands remain. The sediment changes as you go up, from a muddy sandy gravel to a muddy sand, which is what a stream drops when it slows. Twelve radiocarbon dates, on collagen treated in more than one way and on wood, run in a cluster around 30,000 radiocarbon years ago. Calibrated, that is older than 33,000 years. The coldest part of the last ice age is still ahead of them.
The question is whether anyone was standing in the water. Grooves on the bones have been published as cuts from stone tools. On a set of 20 marks, human experts called 13 cuts and 7 the scratches of trampling. A classifier trained on experimental damage, shown the same 20, called 8 of them cuts. The two lists are not the same list. That matters, because the human case is the grooves, and the dates are on the grooved bones. If a person made the groove, the person is as old as the collagen. If the stream, the gravel, or a tooth made the groove, the collagen is only the age of a sloth. Stones in the bed are the other half. Flakes of silicified sandstone, and a small translucent piece of silcrete described as a scraper with a polish like the polish from working dry hide, are published as tools. A 2022 critique calls those stones equivocal, the maps of the dig too thin, the story of how the bed formed unconvincing, and the natural explanations undertested. It reads four millennia of ordinary stream accumulation. The replies say the critique ignored patterns in which animals and which parts of them are there, and that the classified grooves still match experimental cuts.
There is no sky to add. A bone bed has no published alignment and no star. The drawing is a Lestodon skeleton from a scientific figure, a stand-in for the kind of animal, not a fossil from Sauce and not a picture of a groove. What this changes: if the cuts hold, people were in this landscape before 30,000 radiocarbon years ago, earlier than the baseline most of the field still uses. If the stream made the marks, Uruguay keeps an extraordinary sloth bed and does not gain a campsite. The mounds at Los Ajos are a Holocene village argument, thousands of years younger, and they do not depend on this groove.
Why it matters
People in South America before the last great cold, around or before 30,000 years ago, would sit well before the ages most archaeologists still treat as the start of a widely accepted human presence. A natural bone bed of the same age is a remarkable fossil site and not an archaeological one. The grooves are the difference.
The deposit fines upward, from a muddy sandy gravel into a muddy sand, the sort of change a stream makes as its energy drops. More than 2,000 bones have been taken out, and many more are estimated still in the bed. Lestodon armatus, a giant ground sloth, supplies most of them. A sabre-toothed cat and a gomphothere, an elephant-like animal, are among the other species with marks that have been argued over. Twelve radiocarbon dates on collagen, purified and not, and on wood, from different labs and different years, cluster around 30,000 radiocarbon years, which is over 33,000 calibrated years. The marks and the dates are on the same bones, which is why the excavators say the age and the agent are tied together. Of 20 marks that were then classified by a computer trained on experimental cuts, the ensemble called 8 of them cut marks. Human experts had called 13 of those 20 cuts and 7 trampling. The machine and the experts do not share a list. Stones in the bed include flakes of silicified sandstone and a small piece of translucent silcrete read as a scraper, with a polish compared to wear from dry hide. A 2022 review answers that the “tools” are not clearly cultural, that the spatial record is thin, that the geology does not yet show how the bed formed, and that natural causes for the grooves were not tested hard enough. It reads the site as a stream accumulation spread over about four millennia. Replies defend the marks, the stones, and parts of the body-part pattern the critique set aside. No sky reading exists for a bone bed.
Photo tour
1 photograph, unedited. On a phone, swipe the row. Choose one to look closely.
- No open photograph of the stream bed or of a marked Lestodon bone is used here. The skeleton drawing is a stand-in for the species.
What is still unexplained
These sit beside the consensus account, with the same visual weight. None of them is marked solved.
Location and dating
- Location
- Arroyo del Vizcaíno, near Sauce, Canelones, Uruguay. The pin is the town of Sauce, 34.650 south, 56.065 west. A trench coordinate is not in the sources used here. The drawing is not the site.
- Period label
- Current consensus — open to revision Around 30,000 radiocarbon years ago, over 33,000 calibrated, if the cluster of twelve dates holds for the bed. Before the Last Glacial Maximum. Human presence is the disputed part, not the existence of the animals.
- How this dating was done
- Radiocarbon on purified and unpurified collagen and on wood, twelve dates. The human argument is the shape of grooves and the reading of a few stones, not the carbon.
- What could change it
- A stone tool that refits, or a groove the stream cannot reproduce on a fresh Lestodon bone under published conditions, would move the critique. A map of every bone in place would answer the charge that context was not recorded.
What is actually there
- More than 2,000 bones, mostly Lestodon, in a bed about 70 centimetres thick.
- Twelve radiocarbon dates cluster around 30,000 radiocarbon years ago.
- On 20 marks, experts called 13 cuts. A classifier called 8 cuts.
- Flakes of silicified sandstone and a small silcrete piece are published as tools, and disputed.
- A 2022 critique reads a natural stream deposit over about four millennia.
- The drawing is a Lestodon skeleton, not these bones. No sky reading is published.
Open questions
| Question | Status |
|---|---|
| Did a person cut these bones? | Open. Experts, a classifier, and a geological critique do not return the same answer. |
| Are the stones tools? | Open. A scraper reading with hide polish is published. An equivocal-stone reading is published too. |
The current narrative
Where publication stands today. It is the best-supported reading, and it is open to revision. It is not a verdict.
The current narrative — open to revision
A natural bone bed in a stream
The arroyo collected megafaunal bones over roughly four millennia around 30,000 years ago, and the grooves and the odd stones are what a stream does, not a butchery.
What supports this
- The sediment fines upward, which is stream deposition, and the critique reads the accumulation as time-transgressive.
- Alternative natural origins for the grooves were, in that critique’s view, not excluded.
- The widely used archaeological baseline for people in the Americas remains later than this bed.
What pushes back
- Eight grooves were classified with experimental cut marks, not only with trampling.
- Replies say the critique skipped published patterns in which animals and which body parts are present.
- Dating the bones and finding the marks on those bones is a tighter link than a tool found nearby.
What would settle it
It is the centre because a stream full of sloth bones does not become a campsite without marks that survive a geological reading, and that reading has been published against the site. It moves if the eight classified cuts are reproduced only by stone tools.
Emerging narratives
Other published readings. Each may be right. None is adopted here unless the evidence carries it.
An emerging narrative
People at the stream before the Last Glacial Maximum
Stone tools cut some of the sloth bones about 30,000 radiocarbon years ago, so people were in this part of South America before the coldest part of the last ice age.
What supports this
- A classifier trained on experimental marks called 8 of 20 grooves cut marks, several of them at high probability.
- The dates are on the marked bones, so the age and the groove are the same object.
- Flakes and a silcrete piece with polish compared to hide-working are published from the same bed.
What pushes back
- Experts and the classifier did not agree on the list. Thirteen and eight are not the same count.
- The 2022 critique finds the stones equivocal, the formation of the bed unconvincing, and natural damage undertested.
- A stream gravel is exactly where trampling grooves are expected.
What would settle it
It is not the centre of South American archaeology, which still teaches a later start, and the room does not pretend otherwise. It stays on the table because the marks, the dates, and the replies are a published case, not a rumour.
Should the current view be reevaluated?
Open to revision
The age of the bones is a radiocarbon cluster. The human part is a disagreement among experts, a classifier, and a site-formation critique, all in print. Open is the right weight for a pre-glacial-maximum claim that has not become the baseline.
What drives this
- Twelve dates around 30,000 radiocarbon years on a sloth bone bed.
- Cut-mark counts that do not match, 13 against 8, on the same 20 grooves.
- A 2022 argument for a natural, time-mixed stream deposit.
This is the museum’s own weighing of the evidence. It is not a visitor vote, and it is separate from the slider.
Weigh it yourself
Anonymous. Nothing here is stored except on this browser. Your lean does not change the assessment beside it.
Visitor lean so far — not evidence. On this browser only; a shared tally is not collected.
If the contested reading is right
What would follow if a reading that is not the consensus turned out to be true, and what could settle it. This explains the stakes. It does not adopt the reading.
Stone-tool cuts on sloth bone before the Last Glacial Maximum
People would be in southeastern South America more than 30,000 years ago, earlier than the ages most researchers still use for a secure human presence.
Why it is still worth a test
A published match between the eight classified grooves and experimental cuts that trampling cannot copy, on bones whose places in the bed are mapped.
Sources
- Deep classification of cut marks at Arroyo del Vizcaíno Proceedings of the Royal Society B, 2021, the eight cut marks and the twelve dates
- Does the evidence support human occupation 30,000 years ago? PaleoAmerica, 2022, the natural-bed critique
- Reply on the Vizcaíno bone bed PaleoAmerica, 2022, the excavators’ answer
Image credits
- A skeletal diagram of Lestodon, the giant sloth that dominates the bone bed. This is not a Vizcaíno fossil, and it shows no cut mark. It is a stand-in for the animal. . Rodrigo L. Tomassini, Claudia I. Montalvo, Mariana C. Garrone, Laura Domingo, Jorge Ferigolo, Laura E. Cruz, Dánae Sanz-Pérez, Yolanda Fernández-Jalvo and Ignacio A. Cerda, CC BY 4.0. Shown unedited.