Settings

The Method

Proverbs 25:2

It is the glory of God to conceal a thing: but the honour of kings is to search out a matter.

The Torah — the five books of Moses — was written in Hebrew, a language of twenty-two consonants and no vowels. The Koren edition of the Torah, the text used in the original 1994 Statistical Science paper and in every search in this book, contains exactly 304,805 letters. This number has been verified by Jewish scribal tradition for over a thousand years: every Torah scroll must contain this exact count, and a scroll with a single letter added or removed is considered invalid.

An Equidistant Letter Sequence (ELS) is a word formed by reading every n-th letter of the text. If you start at position 100 and read every 49th letter, you form a sequence: position 100, 149, 198, 247, and so on. If the letters at those positions happen to spell a Hebrew word, you have found an ELS at skip 49.

This is not mysticism. It is pattern recognition, testable by mathematics. The question is always: could this pattern have arisen by chance?

The Scroll as a Cylinder

The Torah was written on a scroll. A scroll, when rolled, is a cylinder. When the Torah text is wrapped at width n — that is, n letters per row — an ELS at skip n reads as a vertical column on the surface. An ELS at skip n+1 reads as a diagonal. The cylinder transforms one-dimensional letter sequences into a two-dimensional surface where words become visible in all eight directions: horizontal left and right, vertical up and down, and four diagonals.

This is the key insight of cylindrical ELS analysis. A single verse of Torah, wrapped at different widths, produces different grids — and each grid reveals a different set of hidden words. The words that appear depend entirely on the column width, which corresponds to the skip interval. Changing the width by even one letter changes everything visible on the surface.

How We Test: Permutation Analysis

Finding a word in the Torah at some skip is not remarkable by itself. With 304,805 letters and 22 consonants, short Hebrew words will appear at many skips by chance alone. The question is always: is this particular word, at this particular skip, in this particular location, more likely than chance?

We test this with permutation analysis. For a given word at a given skip, we generate ten thousand random Hebrew words of the same length — random combinations of the twenty-two consonants. For each random word, we check: does it land in the same Torah book? Does it pass through the same thematic surface words? The proportion of random words that match is the p-value. A p-value below 0.05 means the result is unlikely by chance. Below 0.01 is highly significant. Below 0.001 is extraordinary.

Throughout this book, every major finding is accompanied by its p-value. Where no p-value is given, the finding is presented as observation, not proof.

How We Test: Proximity

The WRR method measures not just whether words appear, but whether they appear near each other. Two words at the same skip value can be plotted on a grid (the cylinder). The closer they are on that grid — the smaller the bounding rectangle — the more significant the pairing.

We measure proximity across all skip values for every word pair. The tool reports the closest pairs, ranked by distance. If two thematically related words (say, «kiss» and «silver») are among the tightest pairs in the entire Torah at a shared skip, and they land on a verse about the price of a slave — that convergence is not easily explained by chance.

The Tool: Berea Bible Service

Every search in this book was performed using the Berea Bible Service, an open software system developed for biblical research. The ELS engine operates on the Koren Torah text (304,805 letters, SHA256-verified against the WRR source). Its capabilities include:

  • ELS Search: Find any Hebrew word at all skip intervals from 2 to 152,402, with optional book filtering and cylindrical wrapping.
  • ELS Study: Search multiple terms simultaneously and find clusters — skip values where the most terms converge.
  • ELS Proximity: Measure WRR-standard distance between any two Hebrew words across all shared skip values.
  • ELS P-Value: Permutation testing (up to 100,000 iterations) for book-level and surface-word-level significance.
  • ELS Grid: Generate letter matrices at any width, with verse references for every row.
  • ELS Grid Scan: Scan a grid for all Hebrew words from Strong's concordance in all 8 directions, with interactive HTML and 3D cylinder visualizations.
  • ELS Verse Codes: Find all Hebrew words encoded through a specific Torah verse at any skip.
  • Gematria Search: Find all Strong's entries sharing a given numerical value.
  • Word Study: Complete lexical analysis — Strong's, Abbott-Smith, BDB, Thayer's, LSJ, etymology, frequency, sample verses.

Every finding in this book is reproducible. The Berea service is available as both a web API and a command-line tool. The reader is invited to verify every claim.

Verify It Yourself

The Berea command-line tool can be installed on any Linux or macOS machine. Once installed, every finding in this book can be reproduced with a single command. Here are the exact commands for the key findings:

Find Iscariot in the Torah:

berea call els_search term=סכריות max_skip=50000

Test Iscariot on silver (Leviticus) — p-value:

berea call els_pvalue term=סכריות skip=1051 \
book=leviticus permutations=10000

Result: p = 0.0004, significant = true

Test Iscariot on Passover (Exodus) — p-value:

berea call els_pvalue term=סכריות skip=10685 \
book=exodus permutations=10000

Result: p = 0.0008, significant = true

Find all codes hidden in the thirty-silver verse:

berea call els_verse_codes ref="Exodus 21:32" \
max_skip=500

Measure proximity: kiss + Judah root “to throw”:

berea call els_proximity term1=נשק term2=ידה \
min_skip=2 max_skip=50 top_n=5

Result: skip 5, distance 6, Exodus 21:32 — 2nd closest pair in Torah

Scan a cylindrical grid at the thirty-silver verse:

berea call els_grid_image width=5 start_pos=109380 \
max_rows=20 highlight=נשק render=html

Search for the eleven baptism words at skip 49:

berea call els_study \
terms=טבילה,תשובה,כפרה,זרק,אמונה,פסח,משיח,ישועה,ברכה,צדקה,נשמה \
max_skip=50

Run a full cluster study of Judas terms:

berea call els_study \
terms=בגד,נשק,חנק,שדה,דם,שטן,פת,לילה \
max_skip=500

Check the gematria of a number:

berea call search_gematria value=30

Result includes: H3063 יהודה (Judah) and H391 אכזב (treachery)

The tool is deterministic: the same input always produces the same output. The Torah text is the Koren edition, SHA256-verified. No finding in this book requires trust — only a terminal and a willingness to search.

The Berea Bible Service and its ELS engine are documented at berea.publifye.org. For access to the command-line tool or the API, contact us via publifye.org/contact.

What This Book Is Not

This book is not The Bible Code (1997). Michael Drosnin's bestseller used Torah codes to predict assassinations and earthquakes, without statistical controls, without peer review, and without acknowledging that the same method could produce equally dramatic “predictions” from Moby Dick. The academic community rightly criticized his approach.

This book does the opposite. We do not predict the future. We test whether the Torah's hidden letters correspond to events already recorded in Scripture — the baptism of Jesus, the betrayal by Judas, the messianic prophecies of Isaiah. We use the same permutation methodology published in Statistical Science. We report negative results alongside positive ones. And when we tested a claim that did not survive scrutiny — the “blotted out name” hypothesis for Judas Iscariot — we documented the failure and corrected the record.

The question is narrow and testable: did the Author of the Torah encode patterns that correspond to the New Testament narrative? The answer, we believe, is in the letters.