The Gallo-Roman Dodecahedron and Hollow Icosahedron as Non-Verbal, Tactile Mnemonic Trade Counters: A Unified Framework

Version 3 proposes that the Gallo-Roman dodecahedron should be investigated as a portable tactile numerical interface: a possible tactile commercial counting register or portable bulk-counting device. The central question is deliberately changed from the direct identification attempted in earlier work to a necessity-based question: why would a specialised object of this form have been needed, and what function could that necessity have produced? The hypothesis begins from the physical architecture of the object: twelve pentagonal faces, twenty vertices or knobs, thirty edges and twelve apertures. The revised operational proposal associates twenty counting positions with each of the twelve faces, producing 12 × 20 = 240. A further proposed aperture sequence groups twelve completed 240-unit cycles, producing 240 × 12 = 2,880. These relationships are mathematical observations joined to proposed operating conventions. They do not establish ancient use, a demonstrated physical capacity or a maximum count. The numerical interpretation requires archaeological and experimental testing. The proposed function is therefore not primarily measurement of an external property. The object may instead have served to register and retain a numerical state through physical manipulation. Counting produces a quantity; calculation manipulates quantities; registration maintains a numerical state; recording preserves it for later reference. The device would not replace language. It would make the numerically critical part of a transaction—how many?—less dependent on spoken language. Version 3 integrates evidence concerning Roman material counting practices, professional bulk-goods handling, multilingual commerce, warehouses and supply chains, physical information technologies, and the archaeological distribution and variation of the polyhedra. The hypothesis is explicitly falsifiable through reconstruction, learning experiments, use-wear analysis, corpus comparison and contextual analysis. The Change of Direction: From Identification to Necessity Version 2 attempted to identify the dodecahedron through classical textual terminology and passages. That approach proved methodologically too narrow. Surviving texts can establish that Romans counted, calculated, traded, used interpreters, recorded quantities and protected information without providing a name or description that can securely be matched to the archaeological object. Version 3 therefore announces a deliberate change of direction. The question is no longer primarily “what ancient word describes this object?” but “what historical problem would make an object with these properties worth producing and using?” This is an evidential correction, not a dismissal of classical evidence. Ambiguous, uncertain or unsupported textual translations and terminology have been removed from the evidential foundation. In particular, “calculi acidai” is not used as evidence because it has not been established as an attested term in the relevant classical context. Where genuine ancient passages have only contextual relevance, they are presented as contextual evidence rather than as direct identification. This distinction is essential for peer review. The thesis separates attested evidence, contextual evidence, comparative analogy and interpretation, so that the proposed function does not depend on a textual identification that the surviving evidence cannot securely support.