The British Library catalogues Sloane MS 3189 as Liber Mysteriorum Sextus et Sanctus, the Book of Enoch, or Liber Logaeth. It attributes the 1583 transcription to Edward Kelley and describes letters and numbers arranged largely in 49-by-49 grids from folio 10r onward. A grid of 2,401 positions is not ordinary prose laid out decoratively: location, direction, and the size of the written unit become part of the text.
- Evidence type
- Manuscript record
- Date
- 1583
- Anchor record
- British Library · Sloane MS 3189
The work changes register from verbal units to cellular fields
The manuscript is often reduced to “forty-nine tables,” but its internal arrangement is less uniform. The opening material includes extended verbal units, and the first exceptional face uses words where later grids normally use individual characters. Whitby’s manuscript inventory separates an opening prose sequence, a long run of small square tables, and four enlarged tables near the end. The British Library’s foliation likewise includes material before the main grid run. These changes of format change what counts as a readable unit. In prose, order and word division are supplied by the line. In a one-character-per-cell matrix, a reader must choose a path through rows and columns before a word even exists. In an enlarged table, the physical scale signals that the same manuscript cannot be represented faithfully by shrinking every face into one standardized online square. “Table” is therefore a family of page structures within the book, not proof that every face carries the same kind of message.
The table’s logic is positional, and every extraction adds a rule
A 49-by-49 grid gives each mark two coordinates as well as a place in a row and column. Reading straight across, reversing a row, moving vertically, stepping by a number, or following a diagonal produces different strings from the same field. The manuscript preserves the field, while a transcription chooses how to number and display it. Peterson’s edition adds leaf, row, and column labels for navigation and discusses difficult letter forms, corrections, and rows that do not fit an expected count cleanly. Those editorial aids are useful precisely because the original material is resistant. They should not be mistaken for labels written into every 1583 table. The work’s reported instruction to write backward also raises more than one possibility: it may describe the order of reception, the copying sequence, the direction of a row, or the order of leaves. No single modern reversal follows automatically. The tables can be analyzed as constructed spatial texts without claiming that an attractive extracted phrase is their validated translation.
Worked reading: what changes between Leaf 1b and Leaf 2a
Peterson’s transcription begins the exceptional face called Leaf 1b with the words “Oxar varmol pan sampas.” Most of its rows continue with words occupying cells or positions, followed by a smaller alphabetic portion. Leaf 2a instead begins with the heading “alla opnay qviemmah” and then distributes single characters through a square. On Leaf 1b, “varmol” is already a bounded verbal unit: a reader can preserve its spelling without inventing a route through individual cells. On Leaf 2a, any comparable word assembled from the matrix depends on where the extraction starts, how it moves, and when it stops. The material comparison therefore supplies a practical test. A claim about 1b can quote a recorded word sequence; a claim about 2a must preserve coordinates and an extraction rule. Neither face carries a full parallel English translation. Treating them as two pages of continuous “Enochian sentences” erases the manuscript’s change from speech-like sequence to spatial storage. The dedicated Liber Loagaeth reading room continues with the folio inventory, the competing forty-eight and forty-nine counts, and the later relationship proposed between tables and Calls.
Manuscript order and editorial order are separate evidence
Sloane MS 3189 contains later labels and an arrangement note by William Shippen, evidence that the bound object has a history after 1583. A modern transcription adds a second kind of order: it must decide how to name each leaf face, represent blank or corrected cells, and reconcile rows that resist a clean count. Those choices make the material searchable, but they can make every square appear more regular than the written witness. The difference is especially important when “first” means several things. One face can be first in an editor’s navigation, another can carry a historical note calling it the first leaf, and folio 10r can mark the catalogue’s start of the main grid run. The full reading room keeps those counting systems separate so this compact archive guide can focus on the tables’ material and interpretive logic.
Continue to the full Liber Loagaeth reading room →