First Discussion — What Do We Know and What Next?

Participants: Dirk Pitt, Hesiod, Heinrich Schliemann, Alan Turing, Richard Feynman

Pitt: You know what keeps nagging me? Everyone treats this disc like a locked safe — figure out the combination and the treasure pours out. But I've spent my career recovering things from the sea floor, and the first lesson is: before you open the box, understand the box.

Feynman: That's exactly the point I keep trying to push. We have one object and a century of people projecting their favourite theory onto it. Godart was honest enough to say it: with one text, you can produce a plausible translation in almost any language. That should be embarrassing.

Turing: It is embarrassing — statistically. Forty-five sign types, 241 positions. The combinatorial space of possible phonetic assignments is enormous. Without a second text to cross-validate, any mapping that produces grammatical output in some language is unfalsifiable. Godart's position is mathematically correct.

Hesiod: And yet the object is real. It was made for a reason by a person who understood what they were doing. The question is whether we can reach that reason without reading the signs.

Feynman: I think we can, and more than people give it credit for. Look — someone carved forty-five precision stamps. Each one is a miniature relief, detailed enough to show individual fingers and feather barbs. That's a serious investment of skill and time. Then they prepared a disc of fine clay, scored spiral guidelines, and stamped it on both sides. One correction on Side A — smoothed over and re-stamped. Then fired. This is careful, deliberate work.

Schliemann: Which means it mattered. Nobody expends that effort on a triviality. When I found the gold in the shaft graves at Mycenae, the craftsmanship told me immediately: this was made for someone important, for an occasion that demanded the best. The disc has the same character.

Hesiod: Let me push on the craftsmanship point. The stamps were not made for this disc alone — that much is obvious from the existence of reusable punches. But consider the other direction: the stamps themselves represent a complete system. Forty-five signs, each distinct, each covering a specific category — people, animals, plants, tools, geometry. That is not a random collection. It is a curated inventory, like the categories I used to order the Theogony. Whoever designed those forty-five signs was classifying their world.

Feynman: That's a beautiful observation, and it has a testable consequence. If the sign inventory was designed as a system, there should be internal logic to it — signs that form natural groupings, gaps where you'd expect a sign but don't find one, clusters that correspond to domains of experience. Has anyone mapped the forty-five signs by semantic category and looked for structure in the inventory itself, rather than just in the text?

Turing: Schwartz catalogued the signs by visual type, and others have grouped them loosely — human figures, animals, tools, abstract shapes. But I'm not aware of a rigorous analysis of the sign inventory as a designed system. Most research jumps straight to the text — what order the signs appear in, what combinations recur. The inventory design is treated as given.

Feynman: And that's a mistake. If I were designing a writing system from scratch — say, a syllabary for a language with forty-five distinct syllables — I wouldn't necessarily make each sign a picture of something meaningful. Some signs might be arbitrary. But on this disc, almost every sign is recognisable: a person walking, a fish, a shield, a flower. That's pictographic, not arbitrary. It suggests the signs weren't chosen purely for phonetic convenience. They might carry meaning through what they depict, not just through what sound they represent.

Hesiod: Or both. In my own compositions, a word can function simultaneously as a name and a description. When I call the dawn "rosy-fingered," the epithet identifies her and paints her. These signs may work the same way — each one a sound and a picture at once, with the image reinforcing or modifying the sound's meaning. That would make the system richer than a pure syllabary, and harder to crack, because you would need to recover both layers.

Turing: That's an important qualification. If some signs carry logographic weight on top of phonetic value, the frequency distribution won't match a clean syllabary model. There'd be noise — signs appearing less often than their phonetic frequency predicts, because they're only used when the pictographic meaning is also relevant. I'd want to model that: a mixed system where some signs are purely phonetic and others are phonetic-plus-semantic. The statistical signature would be distinct from either a pure syllabary or a pure logography.

Schliemann: Gentlemen, this is fine theorising, but I want to return to something Hesiod said — that the stamp inventory is a curated classification of the world. If that's true, then the choice of what to include is as telling as any text. Why a bee but not a bull? The Minoans were obsessed with bulls — they painted them, leapt over them, sacrificed them. If the disc is Minoan, the absence of a bull sign is strange. Evans noticed this too, and it fed his argument for a foreign origin.

Feynman: That's actually a strong point, Heinrich. Absence of expected elements can be more informative than presence. In physics we call it a missing resonance — when a particle that should exist by symmetry doesn't show up, it tells you your model is wrong. If the disc's sign inventory doesn't match the Minoan cultural vocabulary we'd expect, either the disc isn't Minoan, or our assumptions about what Minoans would choose as symbols are wrong. Either way, it constrains the problem.

Pitt: Or the signs depict things specific to the context of use, not the culture as a whole. If this is a document for a particular purpose — say, maritime trade or a specific ritual — you'd see fish, vessels, tools of that trade. Not bulls. The sign inventory might tell us what the disc is about rather than where it's from.

Hesiod: Pitt's suggestion is worth dwelling on. When I composed Works and Days, I included the tools and tasks of farming — the plough, the sickle, the threshing floor — not because all of Greek life was farming, but because the poem was about farming. If the disc's signs cluster around a particular domain of activity, the inventory is not a universal alphabet but a specialised notation. That would explain both its uniqueness and the absence of certain expected symbols.

Turing: And it might explain why we have only one specimen. If the system was designed for a narrow purpose — a specific ritual, a particular kind of record — the total number of documents produced would have been small to begin with. The Poles' work on Enigma taught me something I've never forgotten: the size of the system constrains the volume of traffic it can handle. A forty-five-sign stamped-clay system is slow and laborious compared to a stylus on wet clay. You wouldn't use it for routine administration. You'd use it when the effort was justified — when the occasion demanded permanence, or portability, or both.

Feynman: Which brings us full circle. Maybe we've been wrong to assume the disc is a page torn from a lost library. What if it's a rare or even singular document — something made for a specific occasion, using a specialised system that wasn't intended for everyday use? If that's right, the whole strategy of looking for more discs may be misguided. The right move might be to squeeze everything possible out of this one object — the clay, the stamps, the spiral, the correction, the sign inventory as a designed system — before assuming more evidence is out there waiting.

Schliemann: I don't like that conclusion, Richard. My instinct says dig. But I can't dismiss it. If Turing's right about the system being purpose-built and limited in scope, the odds of finding a second specimen drop sharply. I still think examining the Phaistos archive context more carefully — what was stored alongside the disc, whether any of the Linear A tablets share sign-group patterns — is worth doing. Even without a second disc, its neighbours in the archive might tell us what category of document it belonged to.

Pitt: That's the approach I'd take. We don't need to decode the text to learn from its surroundings. The disc wasn't found floating in space — it was in a building, in a room, next to other objects. Those associations are data. And if we pair that with a proper material analysis of the clay and the sign inventory's internal logic, we start building a picture of this object's life before we ever touch the inscription. The disc isn't a cryptogram. It's an artifact. We should treat it like one.