Where it fits — the master reference: the spine of the whole audit, which the other five documents each illustrate one part of.
The Eleven-Step Model · the master reference
The Eleven-Step Model
The complete, named algorithm of Owen’s Word Cipher — every component the primary sources name, each marked by how far it has been proven, with the callable function and the metrics that say how close we stand to proof
Riddle: How is an encryption algorithm like a Flight Recorder?

The algorithm · every named component, promoted to a step
The eleven steps
ENTRY — the five guide words, Fortune first
Fortune, Nature, Honour, Reputation, Pan. Guides are not keys — they are “guides whereby to find the key words,” located by colour-marking every occurrence. Fortune is named “the first great guide”: begin where Fortune clusters.
Verified
DELIMIT — the guide’s sentence is the unit
“Pencil around every sentence containing the guide word, thus enclosing the keys as well.” The delimiter is the sentence boundary; each is typed onto its own sheet.
Mechanizable
KEY WORDS — the content-selectors that head each page
Flagged by capitals, parentheses, tilde marks, “frequent and unnecessary iteration,” and above all RARITY: “if some of the words are but rarely used, it doth even more conceal a Cipher mystery.”
Verified
CONCORDANT · single — a synonym of the key
Widens the catch: love → devotion, adore, adoration; king → majesty, highness, kingdom, court. So every related sentence is swept in, not just literal matches.
Open
CONCORDANT · double — a compound name built by “consort”
Two words coupled by similarity so a name is never set whole: Jude + ass → Jud-as. The least-explained device in the corpus, and the one most likely to carry a determinate constraint.
Open
SHIFT-CUES — the cues for moving between works
Owen: the key words, “repeated by count 10,641 times… and about them revolved the cues for shifting to the different works.” How the decipherer knows to jump from one play to another.
Open
THE SYSTEM — sort and collect by shared key
“Placing in piles all pages containing the same key words… unite and collect the dispersed and distributed matter.” The governing collection rule — stated in the sources, never explained.
Open
JOIN — like must be joined to like
Fit fragments of like key, meaning, or origin; add joining tissue. Because each fragment is self-addressing — the guide and key ride inside it — the coarse assembly (which pieces belong together) is carried by the keys and is determinate. Only the fine last-mile sequence among same-addressed pieces is left unforced by the text.
Open · last mile
TRANSPOSE — inversions and transpositions
Any joined sentence that “reads in a vague or unnatural manner” is transposed until the meaning appears — Owen’s “guides to the inversions and transpositions.” Rule-governed reordering, not read-straight-off.
Open
PARALLEL SENTENCE — weld parallel passages across works
A line built from matching passages in different works describing the same matter — “the parallels, concordances and similar matter.” Sherman’s example welds Winter’s Tale IV.4 with Bacon’s Essay on Gardens.
Partial
TYPOGRAPHIC MARKER — capital letters (claimed)
Sherman: Bacon “enclosed our name in such capital letters that he that runneth by may read.” Tested on the faithful Bodleian Folio, “Bacon” is only an ordinary initial capital. The real typographic cipher is italic-vs-roman — that is the biliteral cipher, not this one.
Falsified
The measurements · how close to proof
Key metrics, computed on the 29-work corpus
Corpus: 29 first-edition works · 923,217 Folio word-tokens · 1,369 guide-word occurrences. Every number below is machine-computed and re-checkable.

The algorithm, resolved to code
One callable function, both directions
The objective core of the eleven steps reduces to a short function over the corpus index. It runs both ways: give it an Owen fragment and it finds the Folio site; give it a Folio site and it returns the metrics.
# df(w) = number of the 29 works containing w ; folio_count(w) = times in the Foliodef locate(fragment): # DIRECTION A: Owen fragment -> Folio site keys = [(w, df(w), folio_count(w)) for w in set(words(fragment)) if folio_count(w) > 0 and df(w) <= 3] keys.sort(key=lambda t: (t[1], t[2])) # rarest first w, d, fc = keys[0] return dict(key=w, sites=folio_positions(w), pinpoint=(fc == 1), confidence=1/fc/d) def score_site(center, r=60): # DIRECTION B: Folio site -> metrics gd = sum(1for g in guide_positions if center-r <= g <= center+r) rare = [w for w in window(center, r) if df(w) <= 2] return dict(guide_density=gd, rarest_keys=rare, has_pinpoint=any(folio_count(w) == 1for w in rare))
→ {key:’ariachne’, sites:[1], pinpoint:True, confidence:1.0}
score_site( As-You-Like-It · I.ii )
→ {guide_density:10, rarest_keys:[‘fauouredly’,’reignes’], has_pinpoint:False}
This is the machine-checkable half of the model. It can be pointed at any of the five Owen volumes’ fragments (Direction A) or swept across every scene of the Folio (Direction B).
The decision the metrics force
Does scanning advance the case? Yes — and the numbers say where
The black box separates cleanly. The beacon (entry) and the serial (key) are text-solved — sharp, objective, re-checkable, and they land on Owen’s own founding sites. That much of the story survives on paper alone, four centuries on.
The final sequence does not — but far less is missing than it first appears. Because the fragments are self-addressing, the surface order is irrelevant by design and the coarse assembly is determinate: the keys tell you which pieces belong together (step 7). What the words leave unforced is only the last-mile order among pieces that share an address — the seams there admit several orderings, so Owen’s fine sequence is coherent but not compelled by the Folio. Steps 6, 7, 9 (shift-cues, “the system,” transposition) feed that last-mile decision.
The Eleven-Step Model · reverse-engineered from Sherman (Baconiana, April 1895) and Owen (Volume I), tested against the Millett known-plaintext and the Bodleian First Folio TEI, and measured on a 29-work first-edition corpus (First Folio + 28 sources). Guide-density, rarity, and locational metrics computed 2026-08-14; every figure re-checkable via the function above. Research by Walter H. Wilkinson · New Gorhambury · gorhambury.org
