PUBLIC RESEARCH PREVIEW Research in progress · uncertainties and source access are recorded throughout.

SITE-018 · 2026-10-29

Santa Elina

About 25,000–28,000 years ago · contextual overview

South AmericaBrazilearly human activitydatingstone tools
Working research. Assessments and source-access limitations are retained from the research record. This entry has not completed its public-release review.

Three objects, two separate arguments

SEI6059, SEI6557 and SEI6386 are giant-sloth osteoderms—small bones formed in the skin—attributed to Unit III(4). The first question is whether people modified them; the second is how their context constrains when that happened. Open the object and dating ledger.

Pansani and colleagues combine optical microscopy, electron microscopy, micro-CT and photoluminescence. Perforations, surface working and localized wear support their interpretation of intentional modification and probable ornament use. Natural foramina, bioerosion and rodent damage are explicit comparators; a hole alone is insufficient. Chemical and surface relationships support modification before fossilization. They do not provide an absolute manufacturing date. [S65 · Pansani et al. (2023)]

The experiment on fossil specimen SEI8004 is a modern control, not a fourth ancient ornament. Similarly, FR3A, FR6 and the rodent-marked FR10 are comparison specimens. Keep their identities distinct from the three proposed artifacts. [S65 · Pansani et al. (2023); S77 · Pansani et al. (2023)]

The dating chain

The 2017 report places microcharcoal GIFA-99177 at 3.03 m in Unit III(4): 23,120 ± 260 radiocarbon BP, reported as 27,818–26,887 cal BP (95.4%; IntCal13). Quartz UW465, sampled at 296–300 cm, gives 25,100 ± 2,500 years by OSL. Osteoderms SE9402, around 3 m in square 26C, give 27,000 ± 2,000 years by U/Th. SE9402 is not an identified assay of one of the three ornaments. [S64 · Vialou et al. (2017)]

Wood in III(2) and quartz in underlying Unit IV provide additional depositional constraints. Neither a natural wood deposit nor charcoal automatically demonstrates a human action. The ledger preserves raw ages separately from calibration and the later Bayesian model. [S64 · Vialou et al. (2017); S77 · Pansani et al. (2023)]

The 2023 model places the beginning of Unit III at 28,743–26,536 cal BP (95.4%). This boundary must not become a date stamped on each object. The supplementary model summaries and layer labels contain differences that remain explicitly flagged. [S65 · Pansani et al. (2023); S77 · Pansani et al. (2023)]

What this can support

The combined evidence supports a published case for human working of sloth bone in an early archaeological context. Probable ornament use is a functional inference. It does not demonstrate sloth hunting, a particular necklace, ritual meaning, a named community or a migration route. The authors cannot establish that the objects were manufactured at the shelter. [S65 · Pansani et al. (2023)]

A glimpse of lives, not a complete society

Santa Elina adds a question to our comparison of life before the Younger Dryas: did people invest time in objects carried or worn on the body? If the ornament interpretation holds, these finds would preserve something of personal expression alongside practical skills. They cannot tell us who wore them or what they meant.

Daily-life questionEvidence and reasonable inferenceWhat remains unknown
Where did people spend time?Finds beneath a natural rock overhang support investigating shelter use.Built dwellings, seasonality and length of residence.
How did they obtain materials?Excavators distinguish toolstone from the shelter rubble and identify nearby outside sources. Procurement suggests knowledge of the surrounding landscape.Exact journeys, exchange or migration routes.
What did they eat?Sloth remains occur in the early deposits.An association is not a meal: the proposed ornaments do not establish killing, butchery or dietary proportions.
What skills were used?The proposed manufacturing sequence includes shaping, perforating and finishing.Who made or taught these skills; whether manufacture happened at the shelter.
What did objects communicate?Localized wear is interpreted as attachment or suspension.A necklace, ritual, rank or named cultural identity cannot be reconstructed.

The material and sourcing observations come from the excavation and object studies; the limits above are our evidential assessment. [S64 · Vialou et al. (2017); S65 · Pansani et al. (2023); S77 · Pansani et al. (2023)]

Later hearths, food remains and wall paintings must keep their own contexts. The 2017 excavation account associates much of the painted sequence with Holocene occupation. Neither the paintings nor the richer later domestic record can be assigned the age of the earliest deposits. [S64 · Vialou et al. (2017)]

The competing explanation deserves a specific test

The published case: the authors identify a combination of working traces and wear, not simply unusual holes. Their supplement includes natural and bioeroded comparison specimens and an 80-osteoderm survey. [S65 · Pansani et al. (2023); S77 · Pansani et al. (2023)]

The objection: archaeologist André Costopoulos asks whether rare natural lesions could account for three selected objects among thousands. He wants stronger comparisons of perforation walls and warns that reproducing a similar object experimentally does not exclude other causes. This is a scholarly blog critique, not an independent specimen reanalysis or a demonstrated natural explanation. [S185 · Costopoulos, André (2023-07-26)]

Our assessment: the combination of traces makes this a substantive archaeological case worth including, with attribution. Its strength depends on how reliably those traces distinguish manufacture from natural alteration. A blind comparison with a broader control collection would help test that distinction; the low proportion of proposed ornaments alone neither proves nor disproves human activity.

There is also a useful protocol correction. The critique describes drilling modern osteoderms, but Text S6 reports polishing, without drilling, the modern dry armadillo control UFCAT-MAM 1/6. Drilling and polishing were performed on fossil sloth SEI8004. These controls answer different questions. This correction does not resolve the larger objection. [S77 · Pansani et al. (2023); S185 · Costopoulos, André (2023-07-26)]

Two additional research leads, kept separate

Tooth dates: a paper assigned to a 2024 issue, first published online in July 2023, reports tooth bioapatite ages and a conversion to estimated collagen-equivalent ages. For Unit III4, the abstract gives a minimum of 22,339–22,534 cal BP and a converted result of 25,994–26,396 cal BP. These are different treatments of a tooth result, not independent dates on an ornament. The ledger review note retains both units’ published values; full laboratory and conversion methods remain to be audited. [S186 · Pansani et al. (2024 issue; first online 2023-07-10)]

Possible burning: an October 2026 conference abstract reports heat indicators in a sloth osteoderm and rib from Unit II, provisionally described as about 13 kya BP. The authors suggest possible human-caused burning. This is a preliminary report concerning a later context; it does not establish cooking or fire use in the earliest ornament layer. [S187 · Pansani et al. (2026)]

Focused next checks

  1. Recover the object-ID-to-position register and relate each ornament to dated samples, without inventing a one-to-one mapping.
  2. Compare perforation walls and working traces across the proposed artifacts, natural controls and experimental specimens, retaining the distinction between publication review and physical inspection.
  3. Obtain the full tooth-dating methods: original specimen and laboratory IDs, pretreatment, conversion equation, uncertainty propagation and calibration.
  4. Locate the full Unit II burning report and its specimen contexts; verify its age convention before plotting it.
  5. Retain the unresolved chronology-model discrepancies and obtain executable inputs before attempting reproduction.

The present ledger is a source audit, not a new dating analysis or independent inspection of the specimens. Atlas/globe coverage checked: the existing early contextual span is retained with clearer interpretive limits; possible Unit II burning is explicitly unplotted. No new traditions entry is warranted by these bone objects.

After this Santa Elina pass, the comparison queue moves to Ohalo II, then Paisley Caves and Page-Ladson. The original Monte Verde algae documentation remains an open, separate dating task.

Inspect the measurements and review notes · Compare early South America · Locate on the atlas