New Gorhambury·Baconian Ciphers

One-Day Report · Biliteral Cipher · September 4, 2026

Today's claim

A page nobody claims is enciphered failed to reproduce the split in the letter w — the first time a test designed to kill the claim did not kill it — while on the same day the Baconians' own books turned out to claim this same passage in three different printings of Shakespeare, which at most one of them can be reading.

findings may be superseded by later reports

The short version

In 1899 an American cryptanalyst named Elizabeth Wells Gallup said she had found a hidden message inside the First Folio of Shakespeare — not in the words, but in the shapes of the letters. Two slightly different italic typefaces, alternating, spelling out a binary code devised by Sir Francis Bacon.

Almost everyone has considered this as nonsense for a century. Almost nobody has measured it. That is the whole of what this project does.

Today's result in one sentence. A page nobody claims is enciphered failed to reproduce the split in the letter w — the first time a test designed to kill the claim did not kill it — while on the same day the Baconians' own books turned out to claim this same passage in three different printings of Shakespeare, which at most one of them can be reading.
The Troilus Prologue with the 1916 marking laid over every letter
drag sideways on a phone · tap or click for full size
Everything the 1916 manual claims about this page, laid on the page itself: a blue or red bar under every letter for the form Riverbank assigns it, a tick at each five-letter group, gold rings on the eight w that could be measured, magenta on the two glyphs Lesson 10 orders marked against their appearance, and the message reading down the right margin.
Can I just see the claim, on the page?

The figure above is the whole of it in one picture. Every letter of the Prologue carries a bar in the colour of the form Riverbank assigns it. Every fifth letter starts a new group, and each group is one letter of the message running down the right-hand margin — FRANCIS ST ALBAN at the top, ELIZABETH R and the signature at the foot.

Two things are worth looking at closely. The gold rings mark the eight w clean enough to measure; the two the 1916 index names by word and line are ringed heavily, and the strip at the bottom shows all eight cut from the page — five with a small angular counter, three with a large round bowl, and nothing in between. That much is visible without any statistics.

The magenta rings are the other half of the story. Those two glyphs are the ones a Riverbank lesson instructs the student to mark against what the glyph looks like. A rule that needs exceptions supplied by the teacher is not being read off the page.

Questions

What is the biliteral cipher?

Sir Francis Bacon described it in 1623. Every letter of the alphabet becomes a group of five symbols, each symbol either an a or a b — binary code, three centuries before binary code.

The two symbols need not be letters at all. They can be two founts of type so similar that a reader notices nothing. Print an innocent passage in an alternating mixture of the two, and the shapes of the letters carry a second message underneath the first.

So what is actually being tested here?

Not whether Bacon wrote Shakespeare. Something much narrower: is the italic type on these pages one fount, or two? That is a question about metal, ink and paper, settled in 1623. The photographs are free and public, and anybody with the same images can check the working or overturn it.

It is the necessary condition. If there is only one fount, the claim dies where no argument can reach it. If there are two, nothing about Bacon follows — but a fact about the book would be established that nobody currently knows.

What was the test, and why does it matter that it failed to kill anything?

Yesterday's measurement found that the sixteen lowercase w on our page fall into two clean groups by height, with an empty gap between them and nothing near the boundary.

The obvious objection — and it was ours, not a critic's — is that this might be ordinary printing-house history. The letter w does not exist in Latin. Continental type-founders under-supplied it, and English printers made up the shortfall from wherever they could. A case holding ws of two different origins would produce exactly this split, with no cipher anywhere near it.

That objection makes a prediction: the split should appear wherever italic w appears in this book. So we went and looked at another page.

The Prologue to Henry V — same book, same fount, same shop, an italic Prologue in verse, and not one of the five passages Riverbank ever claimed. Fourteen ws. They are all the same height. One single group, no gap, and every one of them sits in the range of the short form from our page. Not one is tall.

So the split does not appear wherever w appears. Something about our page still needs explaining.

Does that mean the cipher is real?

No — and today also produced the worst evidence against it yet.

Tracking down an old newspaper story led to a book from 1913 by Gertrude Horsford Fiske, containing alphabets drawn by Gallup herself. Its plate of our passage is taken from the Second Folio of 1632 — a completely different printing, set nine years later by different workmen from different type.

And a folder of 1916 worksheets in the New York Public Library shows a Riverbank student deciphering the same passage from the Third Folio of 1664, and getting a different message, signed with different initials.

A cipher carried by the shapes of letters cannot survive being re-set in new type. Three printings, three readings, and at most one of them can be reading the type. That is not our argument — it is theirs, assembled from their own books.

Then why keep measuring?

Because the two findings are about different things, and conflating them would be the easy mistake.

The three-printings problem is about what Gallup and Riverbank were doing. It is close to fatal to their decipherments.

The height measurement is about what is on one leaf of one book. Two groups of w with an empty gap, on a page where a control page shows one group, is a fact about that leaf whatever anyone in 1899 or 1916 believed. It might have a thoroughly boring explanation — and there is a good candidate, below — but it is not disposed of by showing that the people who first pointed at the page were unreliable.

What is the boring explanation you have not ruled out?

Our page is a cancel — a leaf printed separately and later, to replace one already set. Troilus and Cressida gave the Folio's printers more trouble than any other play: begun, abandoned over a rights dispute, resumed, its opening pages reprinted, and much of it set by an apprentice whose work was so error-prone that his superiors had almost everything he touched proof-read.

A leaf produced in those circumstances might easily have been set from a case holding sorts of two origins. That would give a split on this page and no split on a normally-produced page — which is exactly what we observe.

It is testable. The rest of that cancel is three more pages, and nobody has ever claimed a cipher on any of them. They are next.

What would change your mind?

Written down in advance and unchanged:

  • If the rest of the cancel shows the same split, it is a printing accident and the cipher reading of it is finished.
  • If more control pages turn out to split after all, the Henry V result was a fluke of small numbers.
  • If the other letters never separate, the claim is dead regardless — Riverbank's own key gives paired forms for the whole alphabet, and today fourteen letters were tested and none of them separated.

A negative is published in the same terms as a positive. That was decided before any data was looked at.

Isn't a fourteen-letter failure already fatal?

It is the strongest thing against the cipher on the measurement side, and it is stated in full in the technical section rather than buried.

Every letter with enough examples to test was tested — fifty-six separate tests on shape, height, width and position. Nothing survived. The letter e, with ninety-nine examples, came back further from significant than a coin-toss would predict.

So whatever is happening with w, it is not happening across the alphabet. A cipher needs the whole alphabet. A type-case accident needs only one letter — and w is precisely the letter an English printer was most likely to have in mixed sorts.

Who is doing this and why?

Walter Wilkinson, Worland, Wyoming. The overall project encompasses Orville Owen's Word Cipher; this is its sister strand.

The reason it has merit is that the question has stayed arguable for 127 years purely because nobody measured it — and until about six years ago, nobody easily could. The photographs went online free. The statistical machinery got built for another purpose entirely. Both halves now exist, and no one had put them together.

The wider project lives at New Gorhambury, which collects the sources, the experiments and the Word Cipher work. Experiment One is this report's direct ancestor. Worth knowing before you read it: that page argues a conclusion. These reports deliberately do not. The difference is the point of publishing them daily.