The ninety-five percent that was painted on skin

Claim (verbatim)

The oldest uncontested symbolic objects from our species are durable by accident: the cross-hatched ochre plaques from the Middle Stone Age levels of Blombos Cave (roughly 73,000-100,000 years old) and the geometrically engraved ostrich-eggshell fragments from Diepkloof Rock Shelter (around 60,000 years). They are read, rightly, as evidence of an early symbolic capacity. But the same sites shout a second fact more quietly: engraving on a hard, imperishable surface is a rare act against a background of ochre used overwhelmingly for something else - grinding, powder production, presumably colouring bodies, hides and wooden objects that could not survive. If deliberate geometric marking were a common mode of expression, the durable engraved pieces would be a modest fraction of worked ochre; if the engraved objects are the visible tip of a symbolic economy that ran mostly on perishable media, that fraction will be tiny, and the durable record will be a systematic under-sample of the repertoire. The missing knowledge is the perishable majority - the marks on skin and bark and wood. It cannot be recovered, but its scale can be bounded from below by measuring how small a share of all worked ochre ever received an engraving at all.

Prediction clause (verbatim)

Prediction: across the published Middle Stone Age ochre assemblages that have been use-wear coded (Blombos M1/M2/M3, Klasies River, Pinnacle Point, Sibudu, Klein Kliphuis), the fraction of modified/utilised ochre pieces bearing deliberate geometric engraving will be under 5 percent (primary clause), establishing engraving as a rare durable trace and the durable symbolic record as a lower bound on a mostly perishable repertoire. Disambiguation: 'deliberate engraving' follows each excavation team's own distinction between intentional incised motifs and utilisation striations from grinding; ambiguous pieces are counted as non-engraved (a conservative choice that only raises the fraction). Coverage guard: assemblages without published counts of total worked ochre are excluded so the denominator is real, not estimated.

Kill-dataset (verbatim)

Kill: the published ochre inventories - Henshilwood, d'Errico et al. on the engraved Blombos ochres (Journal of Human Evolution, 2009) and the site's total worked-ochre counts, Watts' Middle Stone Age ochre studies, and Texier et al. (PNAS, 2010) on the Diepkloof engraved eggshell with its assemblage totals - divide engraved pieces by total modified pieces per site and test the pooled fraction against 5 percent.

Novelty / leakage triage

already answered in the literature

The primary clause is guaranteed by the published inventories. At Blombos the deliberately engraved ochres number in the low tens against thousands of pieces of worked/utilised ochre (Henshilwood, d'Errico et al. 2009, Journal of Human Evolution), and the other Middle Stone Age assemblages carry essentially no engraved ochre, so the pooled engraved-vs-worked fraction is far below 5% on the published counts alone. The interpretive point too - that most ochre was ground for powder/colouring while deliberate engraving is an exceptional durable trace over a mostly perishable symbolic economy - is a stated conclusion of the ochre literature (Watts; d'Errico). Only the cross-site pooling arithmetic is left; the threshold itself is a published near-certainty.

• C.S. Henshilwood, F. d'Errico & I. Watts, 'Engraved ochres from the Middle Stone Age levels at Blombos Cave, South Africa,' Journal of Human Evolution 57 (2009)
• I. Watts, 'Red ochre, body painting, and language: interpreting the Blombos ochre,' in R. Botha & C. Knight (eds.), The Cradle of Language (Oxford: Oxford University Press, 2009), and Watts' Middle Stone Age ochre studies
• P.-J. Texier et al., 'A Howiesons Poort tradition of engraving ostrich eggshell containers dated to 60,000 years ago at Diepkloof Rock Shelter, South Africa,' PNAS 107 (2010)

Predictions

No prediction registered yet.