When Marcelino Sanz de Sautuola reported painted bison on the ceiling of Altamira in northern Spain in 1880, the archaeological establishment ruled it a forgery. Engraved bones and stones from Ice Age deposits were already accepted; a polychrome ceiling produced by a working group of painters was not, because “primitive men” were not supposed to have conceived such a thing. The dispute ran more than twenty years, closing only when the eminent French prehistorian Émile Cartailhac published his “mea culpa of a sceptic” in 1902.
Paolo Pecere opens his essay in the journal Mechane with that episode and adds a stranger one. In the cave of Niaux in the French Pyrenees, painted bison in the Salon noir stood within reach of graffiti left by modern visitors, and for years nobody remarked on them. The paintings were not hidden. They were unseeable, because no one yet had a category that would let them count as anything.
That is the essay’s real subject: not what cave art means, but what kind of knowledge about it is available to us at all.
A Hundred and Fifty Years, No Consensus
Once Paleolithic art was admitted, explanations arrived in waves. Was it hunting magic, meant to secure game and control the weather? Solomon Reinach argued so in 1903, drawing on James Frazer and on reports of Aboriginal Australian painting, and the theory carried the authority of Henri Breuil for decades. Was it religion? A badge of sexual or social groups? The record of shamanic visions?
Nearly all these hypotheses leaned on the same move: comparing prehistoric images with the practices of living peoples documented by ethnographers. And nearly all ran into the same trouble. The painted species in European caves frequently do not match the species people were eating. Some scholars now hold that the meaning of rock art is simply lost.
Pecere’s claim is that this is not only archaeology’s difficulty. Tens of thousands of years of image-making are a subject for aesthetics, philosophy of mind, philosophy of language and environmental philosophy — and philosophy has more to contribute here than commentary.
Two Warnings from Naples
The first resource is Giambattista Vico (1668–1744), the Neapolitan professor whose Scienza Nuova set out to make a science of human history. Vico faced the same predicament: he wanted to understand peoples who left no writing.
He drew a line between “philology” — the empirical study of language, literature and antiquities — and “philosophy,” whose task was to interpret what philology gathered. And he argued that history is knowable in a way nature is not, because humans made it: we can trace human deeds back to human capacities. To explain why unconnected peoples independently developed religion and the burial of the dead, Vico appealed to senso comune, common sense, a judgment felt without reflection across the whole race. Against that he set two warnings — the “conceit of nations” and the “conceit of scholars,” twin habits of reading the present into the past.
Pecere takes something specific here. He declines the familiar move of hailing Vico as a precursor of later science; the providential scheme, the three fixed ages and much of the linguistic history are not salvageable. What survives is a posture: that prehistoric art and religion are open to free rational inquiry at all, and that the route to the past runs through capacities rather than through borrowed content.
Dewey, and Art That Has to Be Remade
The second resource is John Dewey (1859–1952) and Art as Experience (1934). Dewey denies that a work of art is a physical object. A parchment or a slab of marble persists; the work of art is recreated each time someone experiences it, as a score is recreated in performance. The very idea of segregating “fine art” from feathers, utensils, bows and spears is modern.
This gives Pecere a two-layer structure. Underneath any artwork lies a biological stratum — rhythm, the organisation of energy, the enhancement of living — reasonably stable across the species. Above it sit cultural codes that assign meanings and change constantly. Aby Warburg’s Pathosformeln, formulas of gesture that resurface across ages without conscious revival, name the motifs that recur precisely because they sit low in that stack: handprints, sexual organs, animals.
Hence the method. Stage one: reconstruct the cognitive and affective capacities involved in making and seeing pictures, using cognitive archaeology, neuroscience and phenomenology. Stage two: build careful interpretive conjectures from historical and ethnological analogy. Stage two is irreducibly conjectural, and Pecere attaches a proviso with real force — the cave of El Castillo was decorated at intervals of up to ten thousand years. Prehistories predated prehistories. No tradition survives that long without transformation.
Hands, and Hybrids
Two cases test the model. At Gargas in the High Pyrenees — another site where paintings were known but unseen until 1906 — more than 220 hand stencils, dated to some 27,000 years ago, were made by blowing red and black pigment around adult and children’s hands. Many are missing phalanges.
Stage one asks what the gesture required. Aristotle called the hand “the instrument of instruments.” Giordano Bruno imagined a beast with twice our wits and feet where its hands should be, and asked where its institutions would then be. André Leroi-Gourhan grounded human technique in the hand freed by upright posture and named the planned sequence of toolmaking the chaîne opératoire. Pecere’s inference from the incomplete hands is the sharp one: once mutilation and disease are set aside, a printed hand short a finger is no longer bound to any hand of flesh. Its maker could modify what was perceived — could imagine alternatives.
The second case is San rock art in southern Africa, where therianthropes, human-animal hybrids, appear among floating eland at Game Pass in the Drakensberg. David Lewis-Williams built the shamanic theory from nineteenth-century sources (Joseph Orpen’s conversations with the San survivor Qing; Wilhelm Bleek and Lucy Lloyd’s archive of ǀXam testimony), twentieth-century Kalahari ethnography, and the neuroscience of entoptic imagery. Critics have pressed hard on the strength of the analogy and on the circularity of reading geometric signs as hallucinations and then explaining them by hallucination.
Pecere’s own move here is the most productive thing in the paper. The biological ground of rock art need not be an extraordinary state at all. Eland were eaten. They became part of the humans who ate them. A felt continuity of substance between hunter and prey is a more ordinary and more widely available basis for imagining hybrids than trance — and it sits well with Mark McGranaghan and Sam Challis’s rereading of Qing’s “spoilt” as something nearer “tamed,” which folds hunting magic and shamanic vision into a single concern with staying on good terms with animals.
What This Opens
The essay presents its case studies as sketches offered for the sake of the method, which is the right scale for a piece this length — and it means the method is now ready to be run at full length on a single problem. Three projects follow directly.
The eating hypothesis is testable in a way its rivals are not. A study could set therianthropic species against zooarchaeological remains, region by region: do hybrids track dietary staples, dangerous predators, or neither? The known mismatch between painted and eaten species in European caves gives the test real stakes, and a negative result would be as informative as a positive one.
Second, “prehistories predated prehistories” deserves to become a working instrument rather than a caution. What the field lacks is a graded scale of analogical warrant — an explicit account of how fast interpretive authority decays with elapsed time and broken transmission. Arnhem Land, where meanings have been carried forward continuously, and Gargas, where they have not, are the two endpoints. A paper calibrating the scale between them would give stage two something it currently supplies by intuition.
Third, Vico has more to give than senso comune. His sapienza poetica, poetic wisdom, and his imaginative universals — concrete particulars that do the work concepts do later — describe rather precisely what a hand missing a finger might be: not a sign standing in for something else, but a vivid particular functioning as a category. A focused study asking whether the imaginative universal outperforms “symbol” as a description of the Gargas hands would put Vico to work on the evidence, which is where the essay’s own argument points. The Tassili–Predynastic Egypt comparison, where hunting drifts from subsistence to a symbol of pharaonic power over chaos, is likewise the seed of a longitudinal study: one valley, dated sequences, motif change tracked against subsistence change.
The Line Worth Drawing
What the essay finally offers is a way of holding two things at once: that we can know a great deal about the people who painted at Gargas, and that we may never know what they meant. Philosophy’s job, on this account, is to draw that line accurately rather than argue it away. Drawn well, the caves stop being a closed case and become a research programme.
Pecere, Paolo. “Caves, Hands and Therianthropes: Methodological Notes for a Philosophy of Prehistoric Art.” Mechane, no. 10 (2025): 99–122. https://doi.org/10.7413/2784mchn0026.
Leave a Reply