Hieroglyphs before the stone

The Rosetta Stone gets the credit, and the credit is mostly deserved; but the stone is the end of this story, not the middle of it. Before it was pulled from a fort wall in the Nile Delta, Europe had spent fourteen centuries reading Egyptian hieroglyphs fluently and entirely wrongly. The history of that long mistake is as instructive as the decipherment, because the mistake was not ignorance. It was a theory.

The theory came from an Egyptian. Around the fifth century of our era, when hieroglyphic literacy was dying or already dead, a writer known as Horapollo produced a Greek treatise explaining the signs one by one. His explanations are allegories: the goose stands for "son" because no creature loves its offspring so much; the vulture means "mother" because all vultures are female, conceived by the wind. Here is the strange part, the part that kept the book alive. The readings themselves are often correct. The goose really can write "son", and the vulture really does write "mother"; not for Horapollo's reasons, but because in Egyptian the words happened to share their sounds. The signs were puns, rebuses, sound-tricks. Horapollo, or his sources, remembered the answers and had lost the method; he wrapped true readings in symbolic explanations, and it was the explanations that Europe would swallow whole.

His manuscript surfaced on the Greek island of Andros in 1419 and reached Florence at the exact moment the Renaissance was hungry for ancient wisdom. Printed by Aldus in 1505, it taught two centuries of scholars that hieroglyphs were not writing at all but pure philosophy, ideas drawn as pictures, a language of thought with no sounds in it. The emblem books grew out of it; so did an entire European habit of regarding Egypt as the home of sealed, symbolic knowledge. It is worth pausing on how reasonable this was. The signs are pictures; the picture-theory explained them; the theory was ancient, and it was Egyptian. It was also precisely upside down, and it made real progress impossible, because no one deciphers a sound-system while convinced it has no sounds.

There were earlier dissenters, further east. An Arabic treatise attributed to Ibn Wahshiyya, compiled around the tenth or eleventh century, tabulated ancient scripts and floated the idea that hieroglyphs could carry sounds as well as meanings; a genuinely early appearance of the right principle, though its actual sign-readings are mostly wrong, and modern claims that it "half-deciphered" the script do not survive inspection. The principle without the readings, as it turned out, was not enough either.

The seventeenth century produced the mistake's most magnificent monument. Athanasius Kircher, the Jesuit polymath, did one thing of permanent value: he championed Coptic, the language of Egypt's church, and argued correctly that it descended from the language of the pharaohs. He even identified one sign's sound correctly, reading the water-ripple as the Coptic word for water; the first true phonetic value ever assigned by a European. Everything else was fantasy on the grand scale. Confronted with a Roman obelisk whose signs actually record the titles of Ramesses II, Kircher translated page upon page about supramundane Osiris flowing down into the world of Genies. He was not a fraud; he was a man reasoning inside Horapollo's theory, and the theory could generate any meaning at all, which is the surest sign a theory of writing is wrong.

By the late eighteenth century the frame was quietly cracking. Barthélemy suggested in 1762 that the oval rings in the inscriptions, the cartouches, enclosed royal names. Zoëga catalogued the signs and noticed there were too few distinct ones for a symbol-per-word system, and said, carefully, that some signs might be phonetic and that decipherment would need more evidence. The evidence arrived in July 1799, in the wall of a fort near Rashid, as a broken grey stela carrying one decree of 196 BCE in three scripts: hieroglyphic, demotic, and Greek. French soldiers found it; the British took it under the surrender terms of 1801; and copies travelled faster than the stone did.

Two minds closed the distance, and they were built differently. Thomas Young, physicist and physician, treated the stone as a cryptogram. Working from 1814, he matched the demotic text against the Greek, saw that demotic mixed sound-signs with word-signs, and broke the cartouche of Ptolemy into sound values, six of them right. He published anonymously, in an encyclopaedia supplement, and stopped short; he could crack names but had no way into the language. Jean-François Champollion had spent his youth acquiring exactly that way in: Coptic, learned so deeply he wrote letters in it. The cartouches of Ptolemy and Cleopatra, cross-checked between the stone and an obelisk shipped to England by William Bankes, gave him an alphabet of foreign names; the question was whether native Egyptian words used sound-signs too, or whether the phonetics were a special apparatus for spelling Greek rulers.

Thirteen days later Champollion read his Lettre à M. Dacier to the Académie in Paris, with Young sitting in the audience; the priority quarrel between the two men began almost at once and has never fully ended. The fair division is roughly the one Richard Parkinson gives: Young found parts of a key; Champollion opened the language. The real statement of what he had understood came two years later, in the Précis of 1824, in a sentence that answers Horapollo across thirteen centuries: it is a complex system, writing figurative, symbolic and phonetic all at once, in the same text, the same phrase, almost in the same word. Sound-signs, word-signs, and silent classifier signs, braided together. Not a language of pure ideas; a working script, with puns at the bottom of it, just as the goose and the vulture had been trying to say all along.

Champollion got nine years to use the language he had opened, and died at forty-one. The lesson of the long failure before him is not that his predecessors were fools; Kircher was among the most learned men alive. It is that a wrong frame does not feel wrong from inside. Fourteen centuries of scholars looked at phonetic writing and saw philosophy, because a fifth-century book told them what kind of thing they were looking at, and evidence cannot correct a category. The stone did not just supply a translation; it supplied a reason to change the category, and even then it took two decades and a man who had rebuilt the dead language's living descendant inside his own head.