Pansani et al. (2023) recently published an interesting update on some giant ground sloth osteoderms possibly modified by humans about 27ky ago at the Santa Elina site in central Brazil. If these osteoderms really were perforated and used as pendants by humans at Santa Elina twenty seven thousand years ago, it would make them the oldest securely dated archaeological objects in the Americas by about twelve or thirteen thousand years.
This paper is full of great new data about the osteoderms in question, and some very thought provoking discussion about how to determine whether they were modified by humans. I am not ready to conclude on the basis of this paper that there are 27 000 year old archaeological remains at Santa Elina, but there are some good starting points in the discussion for further work to test the hypothesis more fully.
I discussed an earlier report on the Santa Elina osteoderms back in 2017. I’m glad to see that work is ongoing. One of the problems with Santa Elina and other similar very early claimed sites is that they are usually claimed to be archaeological (i.e. show evidence of human activity) on the basis of stone tools that may or may not be real.
The reality is that humans are not the only force in nature that can fracture rocks. In a big pile of fractured rocks, especially if they are rocks that fracture conchoidally, there is a certain probability that frost, fire, water erosion, landslides etc, could produce a few rocks that look very similar to either stone tools made by humans, or like what’s left after humans have produced a tool (i.e. lithic debitage).
We’ve known for a long time that when a few potential stone tools among a mass of fractured rocks are the only evidence of human activity in a particular place at a particular time, we have to be very cautious, and we usually need other lines of evidence to support the conclusion that the location is evidence of human presence and activity. This was notably a problem with the Calico Hills claim in the 1970s.
This is the case of Santa Elina and other places like it. They show very clear evidence of human activity until perhaps twelve or fourteen thousand years ago, along many converging lines of evidence, including fireplaces, dietary remains, stone tools, storage pits, ancient DNA etc, but they only show a few potential stone tools in older deposits.
So far, this has been normal for the Americas, where the human signal tends to disappear around sixteen thousand years ago. This doesn’t mean no one was around, but it does mean that if they were, we haven’t been able to find and identify their material traces.
That’s why the osteoderms of Santa Elina are so intriguing. In addition to the claimed stone tools, they would represent a second line of evidence in support of human activity at the site almost thirty thousand years ago.
This new article doesn’t deal with the claimed stone tools in older deposits. It takes them for granted. It focuses entirely on evaluating whether the osteoderms in question were perforated by humans and used as pendants.
The claim
Three osteoderms recovered in a 27000 year old layer show evidence of having been perforated by drilling with a stone tool. They also show signs of wear by suspension with some sort of string.
The authors argue that under close examination by optical and electron microscope, multispectral imaging, and microtomography, the modifications made by humans are different from modifications made by other agents such as insects and bacteria (bioerosion), and that they are similar to those made experimentally with a stone drill on modern armadillo osteoderms.
Evaluating the claim
First, there is little doubt that the geological find context is 27000 years old. The only real question is whether the osteoderms were modified by humans, or by other forces. This is a tough one, because seeing those perforations, the tendency is very much to think that someone made them, and for a purpose. They are visually impressive. The reaction is visceral.
The suspected anthropic osteoderms, from Pansani et al. 2023
The problem here is very similar to the issue with potential stone tools found in a very big pile of broken rocks. It’s an question of probabilities. The three osteoderms identified as human-modified are part of a collection of 7069 osteoderms recovered at the site. The question then becomes: How likely is it that 3 of 7069 osteoderms (1 in about 2356) would look like this after 27000 years, in the absence of human activity?
Through comparative imaging of osteoderms, the authors do a good job of ruling out at least two non-human forces. The osteoderms don’t look like they were modified through rodent gnawing or through passing through a digestive tract. Using multispectral imagery, they also establish that the perforations are old. They were not made recently, and were probably made around the time the osteoderms were deposited in the context in which they were found.
The experimental work, which consisted of perforating modern osteoderms with a stone drill, is interesting but necessarily inconclusive for this kind of study. Yes, one can produce modifications that look like the ones seen on the target osteoderms, but that doesn’t mean that they were modified by humans using those methods. The experimental work doesn’t rule out that the osteoderms in question are archaeological (which is good, because it could have), but it can’t confirm it either. Still, the experimental work is interesting and valuable.
Close ups of one of the perforations, from Pansani et al. 2023
The ruling out of bioerosion by fleas and/or other insects or bacteria is a lot less conclusive. The evidence presented in the paper shows that bioerosion by fleas and bacteria can produce perforations on osteoderms, and can also produce the kinds of grooves that the authors identify as usewear from suspension by a string.
The authors’ main argument here is that the perforations and grooves on the target osteoderms are larger than the ones on the ones identified as unmodified by humans. “The mean value of the area of the natural foramina is 0.19 ± 0.36 mm, while the mean values of the bioerosions is 0.47±0.72 mm, and the mean value of the perforated holes is 5.56 ± 2.53 mm.”
That is a big difference, and it could be significant. On the other hand, putting the very large perforations in one class because they could be made by humans, and putting the smaller ones in another class because they are not, naturally creates a big difference between the classes. So this is something to watch out for.
Some of the sources the authors use to argue that there is a significant difference between the human-modified osteoderms of Santa Elina and other ones modified by bioerosion, actually create more questions for me. Moura et al. (2021), in particular, show lesions on extinct quaternary armadillo osteoderms in their supplementary materials that are, visually at least, very much like the ones on the claimed Santa Elina ornaments.
Lesions and density plots from Moura et al. 2021
Also, Moura et al.’s density plots of lesion sizes show that while lesions as large as the perforations on the Santa Elina objects are rare, they do occur, and perhaps often enough not to be surprising in a sample of 7000.
There remains the imaging of the sides of the perforations. Here, the authors provide very useful images and data, but the focus is on the objects suspected of being human-modified, and there isn’t much in the way of comparison with others. This had already been noted by Richard Fariña of the Universidad de la República de Uruguay in the first round of reviews, and alluded to again by John Hutchinson of University of London in round four. It still hasn’t been completely addressed, in my opinion.
Yes, drill marks are recognizable, but how different are they from flea marks after 27000 years? That question is not clearly answered, or if it is, I missed it in the paper and the accompanying materials.
Bottom line
The three osteoderms from Santa Elina that have large perforations could indeed be the result of human activity, but after reading the paper carefully, including the supplementary materials, and after going to some of the sources the authors cite in support, I still have questions.
I don’t think the authors have shown convincingly that the perforations are highly likely to be human related. In particular, I would like to see more attention on comparison of the features of the walls of the perforations on the target objects, and those on objects from contexts that are definitely not human related. That’s where the answer probably lies, and it is an underexplored vein in the paper’s rich data deposit.
There may be a lot that I am missing here, but as a generalist reader, I don’t find convincing answers to my main questions. I hope the work on the Santa Elina osteoderms continues.
References
Moura et al. 2021. Damaged armour: Ichnotaxonomy and paleoparasitology of bioerosion lesions in osteoderms of Quaternary extinct armadillos, Journal of South American Earth Sciences https://doi.org/10.1016/j.jsames.2021.103255
Pansani et al. 2023. Evidence of artefacts made of giant sloth bones in central Brazil around the last glacial maximum, Proc. R. Soc. B. https://doi.org/10.1098/rspb.2023.0316
ArcheoThoughts 