A Working Book, Not a Composition

The Paris codex is best understood as a professional's shelf collapsed into one volume. It runs to several thousand lines and holds dozens of independent items with no argument connecting them: a procedure for a dream oracle sits beside a text for driving out a spirit, which sits beside an ascent narrative of real literary ambition. Some items carry their own titles and attributions, naming an earlier authority the way a cook's notebook names whose recipe it copied. Others break off. The hand is consistent enough to indicate one copyist working through material of very different origins and dates.

That structure is the reason the book matters. A single spell tells you what one person wanted. A handbook tells you what a practitioner thought was worth keeping, and in what order, and how he classified it. The Paris codex sets Greek divine names beside Egyptian ones and beside Hebrew forms such as Iao and Sabaoth, and it does so without visible embarrassment or explanation, which is itself the evidence: for whoever compiled it, these were not competing religious systems but a common stock of effective names.

The strings of vowels and unetymological words that fill the margins of this material, the voces magicae, are not decoration and not nonsense in the writer's terms. They are treated as language belonging to the powers rather than to people, and their consistency across separate papyri shows they were copied faithfully rather than improvised. What they meant to the copyist is a different question from what they meant to whoever first wrote them down, and neither is recoverable from the codex alone.