Unicode-Level Detection
The complete structured text was examined for rare code points, punctuation spacing, quotation balance, parenthesis balance, elision marks, numeral signs, and verse-final patterns.
A Documentary Restoration of Scrivener's 1887 Greek New Testament
The Printed Text, Returned to the Page.
Restored through repeated human collation, page-by-page image preparation, independent AI-assisted extraction, controlled difference detection, image-based adjudication, and whole-New-Testament reverification.
Hero images were not found in the Praxis_Bible_Society folder.
Final Audit Certification
Whole-New-Testament alphabetic integrity: verified.
The four independently prepared sections were joined into one master dataset and reprocessed from Matthew through Revelation. These figures are generated from the final audit record.
Punctuation and Editorial-Marks Audit
Every detected finding was adjudicated against the printed source.
This supplementary audit examined quotation marks, parentheses, spacing, elision marks, detached diacritics, Greek numeral signs, and verse-final punctuation patterns throughout the structured New Testament text. Automated detection identified candidates; the 1887 printed page determined every final decision.
Verse records7,957
Original findings67
Verified as printed63
Confirmed corrections4
Pending review0
Mapped image references227
Printed-Page Adjudication
No software finding was treated as self-authenticating. Each decision was made by directly consulting the corresponding image of Scrivener’s 1887 printed edition.
Correction and Reverification
Four confirmed transcription defects were corrected in the controlled dataset. A subsequent run verified that all four disappeared while the 63 source-faithful findings remained fully documented.
Read-Only Public Record
The public report preserves the official decisions, image provenance, review mapping, release metadata, and checksums without allowing visitors to alter the audit record.
Audit class: Punctuation and editorial marks
Source authority: Scrivener 1887 printed page
Verse records examined: 7,957
Original findings reviewed: 67
Verified as faithful to the source: 63
Confirmed and corrected: 4
Unresolved findings: 0
Mapped source-image references: 227
Shared source-image repository: /merged_images/
The Restoration Journey
Four sections became one controlled New Testament dataset.
Duplicate local page names were preserved within section namespaces, registered under unique global identifiers, and reverified as a single corpus.
Matthew-Mark
132 pages and 30,425 Greek words.
Luke-John
148 pages and 36,031 Greek words.
Acts-2 Corinthians
158 pages and 37,493 Greek words.
Galatians-Revelation
162 pages and 37,095 Greek words.
600 Unique Page IDs
Stable global identifiers prevented collisions among the four original page sets.
Source JSON Audited
Metadata, empty placeholders, page continuity, JSON structure, and dataset identity were checked before comparison.
The Page Corrected the Data
Acts 26:20 and Revelation 9:3 showed that even the structured Truth dataset remained subordinate to the printed page.
600 Pages Passed
The integrated master dataset produced 600 PASS results and a 1.000000 match for every page.
Documentary Evidence
The decisions can be inspected.
These cases show why neither AI nor structured reference data was allowed to overrule the printed source.
Truth Dataset CorrectedActs 26:20
Initial Capital RestoredRevelation 9:3
Open the searchable 600-page verification ledger.
Whole-NT Evidence600 PASS Records
Manual Collation Archive
The labor behind the certification remains visible.
Representative photographs from the manual and digital collation process
are retained here. The complete archive preserves every image currently
stored in the project's png folder.
Error RecordActs 2[Acts 2] Red-ink annotations isolating micro-typographical errors, targeting breathings, accents, and punctuation marks with absolute precision.
Word-Level ReviewActs 5[Acts 5] The meticulous manual audit phase, parsing the text word-by-word to ascertain orthographic integrity.
Orthographic ReviewGalatians 5[Galatians 5] Orthographic and accentual correction of key theological vocabulary, such as 'πνεύματι'.
Final Visual AuditOCR Confirmation[Final Visual Audit] The genesis of a zero-defect text: manual confirmation of OCR data through hundreds of subsequent visual inspections.
Visual AuditMark 14[Mark 14] The Visual Audit process: a strict 1:1 digital comparison between the 1887 original scan and the Scrivener text via dual-monitor workstation.
Manual CollationMatthew 6[Matthew 6] Rigorous manual cross-examination of the Sermon on the Mount, utilizing color-coded highlighting for textual verification.
Methodology
Detection was automated. Decisions were documentary.
Software located candidate differences and preserved context. The printed image determined the reading, and only approved corrections entered the controlled dataset.
Independent Extraction
Page images produced a second word dataset capable of exposing differences.
Controlled Pass 1
Diacritics and punctuation were removed while spelling, order, omission, addition, and capitalization remained significant.
Human Approval
Every discrepancy remained unresolved until the corresponding printed page was examined.
Whole-NT Recheck
After integration, the complete 600-page master dataset was processed again.
Printed Page Image
Independent Extraction
Difference Candidate
Human Page Decision
Approved Text and Recheck
Release ID: TR1887-NT-PASS1-20260716
Status: VERIFIED
Pages checked: 600
Greek words checked: 141,044
Errors: 0
Warnings: 0
Integrity failures: 0
Generated from audit record: 2026-07-16T20:27:01
This panel reports the whole-New-Testament Pass 1 result. The page-by-page detection method that produced it — including the handful of Greek runs supplied from the structured Truth dataset rather than a mapped source page — is documented in full in the Truth-Assisted Coverage Register.
The Completed Edition
A restored text prepared for enduring scholarly use.
The completed edition joins documentary accuracy with professional book design. Its Greek text, marginal references, textual notes, typography, and two-color visual system form a unified scholarly publication.
- Greek alphabetic text verified across all 600 printed source pages
- Professional polytonic Greek typography and spacing
- Marginal cross-references and differentiated textual notes
- Publicly inspectable alphabetic, punctuation, and page-ledger audit releases
Ἐν ἀρχῇ ἦν ὁ Λόγος, καὶ ὁ Λόγος ἦν πρὸς τὸν Θεόν, καὶ Θεὸς ἦν ὁ Λόγος.