Liber II Regum

Translated 2026-08-23, one translator, 8 chunks (6,815 Latin words), on Opus. verify-english.mjs clean: columns, notes and sections aligned, no duplicate paragraphs.

Apparatus fired: 6 [var:] · 1 [cj:] · 0 [sic:] · 0 [d:] · 0 [ed:] · 0 [nt:] · 115 [n:] (body count; the frontmatter noteCount fields sum to 111 — runbook false positive 3, count the body) (exact parity with the Latin twins, contents verbatim). Zero guillemets in either language. Zero em-dashes in the English. Structure: CAPUT PRIMUM–CAPUT XXIV.

Conventions. Ruth (8968) §1 governs and nothing here re-opens it: VERS. n.-- passes through verbatim; the lemma is Englished; a fragmentary lemma is rendered as the phrase it stands for; the attribution sigla ride as ordinary [n: …] notes with Latin unanglicized. usque ad → as far as in its own italic span (8978/8970 precedent), no verb ever supplied to complete an abridgment. Chapter heads follow Ruth's shipped English forms — CAPUT PRIMUM. → CHAPTER ONE., CAPUT II. → CHAPTER II. (Ruth ships the roman numeral, not "CHAPTER TWO"; where the launch brief and Ruth's plate differ, the brief's own tiebreaker is "match Ruth's shipped forms").

§1 · MIGNE'S OWN IRREGULARITIES, REPRODUCED NOT TIDIED

• 0571D · VERS. 15-- — the only verse address in the work that lacks the period after the numeral (every other reads VERS. n.--). Reproduced exactly.
• 0566C · …et sub Joab calcandus subditur — the sentence ends with no full stop. The English ends without one (Pattern 8: never supply).
• 0576C · …id est Christum Filium Dei rediit — likewise, and it is also the last line of its chunk. No stop supplied.
• 0576D · …Armoni scilicet et Miphiboseth — likewise. No stop supplied.
• 0576B · cur non servas legem Israelitis, quae olim servata est alienigenis. — Migne closes a direct question with a full stop, four words after closing the previous one with ?. Both marks ride as printed.
• 0569A, 0570B, 0570C, 0573C, 0574C, 0579C, 0580D — the anchor is set hard against the verse address with no space ([0569A]VERS. 16.--). Reproduced without a space.
• 0564C · (qui interpretatur patris lucerna ) — Migne's space before the closing parenthesis is kept.
• 0576A · [n: (HIERON., quaest. Hebr. in Reg. )] — the note's own internal spacing is kept verbatim.
• 0565D · Duo autem viri. (RAB. in lib. Reg., [0565D] tom. II.) — this siglum is not an [n: …] note; Migne sets it inside the italic lemma span as running text, and a column anchor falls inside it. Carried verbatim, unanglicized, anchor in place.

§2 · MARKERS FIRED, AND WHY

2.1 [cj:] — one, at 0563B

…quomodo Saul recedens a timore Domini per obedientiam periit.

Rendered "perished through obedience", with [cj: obedientiam; read inobedientiam, "disobedience"].

Why it clears Pattern 18's bar. The printed word is a perfectly good Latin noun, so Pattern 12 has nothing to wrap and Pattern 7 renders it — and the English then asserts the opposite of the author's own argument. The control is in this work, eight lines below, in the same voice: 0563C reads o Saul, inobedientia tua quasi idololatria in excelsis reputata est, and 0563B's own preceding clause has Saul recedens a timore Domini. Saul perishing "through obedience" is the thing the passage is written to deny.

The rival, and why it lost. The alternative was to leave it as a crux only. Declined: 18a's extension covers the case where the faithful English asserts something false, and this one does, on a reading page, with no marker anywhere to warn a reader. The gloss is additive — Migne's word keeps its English and the conjecture stands beside it — and the verbatim-in-the- Latin-twin guard is satisfied by obedientiam, exactly as 7561 @0648 is satisfied by solum.

2.2 [var:] — six, all resting on sources/vulgate/clementine-flat.txt

Not fired, and the reasoning is the point: every [var:] above is a claim about the Clementine, cited by book chapter:verse from the file on disk. Where the divergence is a single letter, a tense, an enclitic, a connective, or a name-spelling, no marker — see §3. sources/vulgate/README.md warns that this copy is not trustworthy to single-letter precision, and a [var:] is a public claim.

2.3 [sic:] — NONE, deliberately

Three non-word candidates were found (§4.1). All three are raised as plate-read candidates and none carries a marker, because the 2026-08-23 addendum measured non-words at 10 for 10 as OUR digitization's corruption, not Migne's — and a [sic:] publicly blames Migne's plate. The ship test governs: the crux is always safe, the marker is a claim.

§3 · THE LEMMA COLLATION — all 275 italic spans walked

Every span in data/briefs/8953-lemmata.txt was read against the Clementine. The agreements are listed as well as the divergences, because a findings-only list cannot be told from a list by a reader who never looked.

3.1 Checked and AGREEING with the Clementine exactly (clip at either end excepted)

CAPUT I–IV (0563A–0565D): Apparuit homo veniens de castris Saul, (1:2, opening clipped from In die autem tertia) · Accedens, (1:15) · Filios. · Arcum. · In libro. · scriptum est. · Considera, Israel. (all 1:18) · Geth. (1:20) · Montes Gelboe, · Montes Gelboe, nec ros, (1:21) · A sanguine interfectorum. · Sagitta. (1:22) · Igitur post. (2:1) · Veneruntque. (2:4) · Dixitque. (2:5) · Isboseth. · Circumduxit. (2:8) · Et fuit. (2:11) · Egressus. (2:12) · Porro Asael cursor velocissimus fuit quasi unus de capreis. (2:18, clipped before quae morantur in silvis) · An ignoras quod periculosa sit desperatio? (2:26, question mark included, as Migne and the Clementine both print it) · Facta est ergo. (3:1) · Natique. (3:2) · De Egla. (3:5) · Sermonem. (3:17) · Fluxum seminis. · Tenens fusum. · Cadens. (3:29) · Delicatus. (3:39) · Duo autem viri. (4:2) · Et ostiaria. (4:5) · transfigurat se in angelum lucis (2 Cor 11:14).

CAPUT V–VI (0566A–0568C): Et venerunt. (5:1) · Os tuum et caro tua sumus: (5:1, clipped before Ecce nos) · Seniores. · Unxeruntque. (5:3) · Et quadraginta. (5:4) · Caecos et claudos. (5:6) · Cepit autem. (5:7) · Domatum. · Fistulas. (5:8) · Et aedificavit. (5:9) · Misit quoque. (5:11) · Natique sunt. (5:13) · Baal Pharasim. (5:20) · Pyrorum. (5:23) · In cacumine. (5:24) · Congregavit. (6:1) · Sedentis. (6:2) · Qui mortuus. (6:7) · Tribus mensibus. (6:11) · In organis. (6:5, in omnibus lignis fabrefactis context) · Egressa Michol filia Saul in occursum David, (6:20) · Gloriosus. · Et nudatus. (6:20) · Factus est in pace locus ejus, (Ps 75:3, clipped before Et) · Mihi autem absit gloriari nisi in cruce Domini (Gal 6:14, clipped before nostri Jesu Christi).

CAPUT VII–X (0568D–0571A): Nunquid. (7:5) · Et ponam locum. (7:10) · Et fidelis. (7:16) · Et sedit coram. (7:18) · Lex Adam. (7:19) · Quae est. · Ex Aegypto. (7:23) · Invenit servus. (7:27) · Frenum. ×2 (8:1) · Coaequans. (8:2) · Sanctificavit rex. (8:11) · Caesis. (8:13) · A commentariis. (8:16) · Sacerdotes. (8:17) · Et dixit David, (9:1) · Factum est autem post haec, ut moreretur rex filiorum Ammon, (10:1, verbatim) · Faciam, (10:2) · Dominus celavit a me, et non indicavit mihi (4 Rg 4:27, verbatim) · Constitui te Deum Pharaonis (Ex 7:1, clipped before Ecce) · et sorte divisit eis terram in funiculo distributionis (Ps 77:54, verbatim; Migne's (Psal. LXXVII) is right).

CAPUT XI–XII (0571B–0573C): Factum est autem vertente anno, eo tempore quo, (11:1) · Dum haec. (11:2) · Ponite Uriam, (11:15) · Dixitque Nathan ad David, (12:13) · Misitque Joab nuntios, (12:27) · Urbs aquarum, (12:27) · Et tulit, (12:30) · Solliciti servare unitatem spiritus in vinculo pacis (Eph 4:3, verbatim but for lower-case spiritus) · In sole posuit tabernaculum suum (Ps 18:6, verbatim).

CAPUT XIII–XX (0573D–0576C): Sorbitiunculas. (13:6) · Oppressit. (13:14) · Et exosam eam habuit, (13:15) · Tholomai. (13:37) · Cessavitque. (13:39) · Et ancillae, (14:6) · In me, Domine mi. (14:9, clipped before rex) · Igitur post. (15:1) · Post quadraginta autem annos. (15:7) · Et deposuerunt arcam, (15:24) · Sadoc. (15:24) · Chusai. ×3 (15:32) · Quid mihi? ×2 (16:10) · Vadam et amputabo. (16:9) · Dimittite. (16:10) · Semei autem. (16:13) · Achitophel. ×2 (16:15) · Et fortissimus. · Scit enim. (17:10) · Absalom. (17:14) · Accidit autem ut occurreret Absalom servis David sedens mulo. (18:9, the Clementine's comma before sedens dropped) · Quid mihi. (19:22) · Illotis. (19:24) · Tu et siba. (19:29) · Est autem. (19:37) · Et tenuit, (20:9) · Nonne. ×2 (20:19) · Qui abscissum. (20:22) · Et insidiis suis capientur iniqui (Prov 11:6, verbatim).

CAPUT XXI–XXII (0576D–0577A): Facta est. (21:1) · Tulit itaque rex duos. (21:8) · Filios Michol filiae, (21:8) · Gob. (21:18) · Tertium. (21:19) · Quartum, (21:20) · Bethlehemites, (21:19) · Locutus est, (22:1).

CAPUT XXIII (0577B–0579D): Egregius. (23:1) · In timore. (23:3) · Nec est. (23:5) · Et si quis. (23:7) · Fortium. · Ipse. · Tenerrimus ligni. (23:8) · Semma filius. · Age de, · Arari. (23:11) · Stetit. (23:12) · Descenderant tres. (23:13) · Abisai quoque frater Joab filius Sarviae, (23:18, verbatim) · Sed libavit. (23:16) · Tres primos. (23:19) · Ipse percussit (23:20) · In media. (23:20) · Virum Aegyptium. · Dignum. · Hastam. (23:21) · Qui erant. (23:23) · Urias. (23:39) · Et addidit furor Domini, (24:1) · Dei aedificatio estis. (1 Cor 3:9, verbatim) · Lavabis me, · et super nivem dealbabor, (Ps 50:9, verbatim) · diebus nivis. (23:20).

CAPUT XXIV (0580A–0580D): Percussit autem, (24:10) · Sermo Domini. (24:11) · Aut. (24:13) · Erat autem. (24:16) · Areuna. (24:18) · Et obtulit holocausta. (24:25) · Sacrificium laudis honorificabit me, (Ps 49:23, verbatim) · Sacrificium Deo spiritus contribulatus, (Ps 50:19, verbatim) · Haec est domus Dei, (1 Par 22:1).

3.2 DIVERGENT but taking no marker — every one named, with the reason

• 0563A · Qui enim fodit foveam, incidit in eam (Prov XXVI). Clementine Prov 26:27 Qui fodit foveam incidet in eam. Two differences: an added connective enim, and present for future. No marker: the connective is the gloss's own stitching, and the tense turns on a single letter, which sources/vulgate/README.md says this copy should not be trusted to adjudicate. Rendered "falleth into it", present, as printed. The same quotation recurs at 0575C in the same form and is rendered identically.
• 0563A · Os quod mentitur occidit animam (Sap I). Sap 1:11 has os autem quod…. Clipping at the head of a quotation. Not a divergence.
• 0565B · Igitur Michol filiae Saul non est natus filius, usque ad diem mortis suae (2 Rg 6:23 usque in diem). ⚑ This usque ad is the verse's own words, not the abridgment formula — hazard D of the brief, and it is why the span is not italicised as usque ad on its own. A preposition swap with identical sense; no marker.
• 0565D · Erat autem Jonathae filius pedibus debilis (2 Rg 4:4 Erat autem Jonathae filio Saul filius debilis pedibus). Migne drops Saul, turns the dative into a genitive and reverses debilis pedibus. The assertion is unchanged ("Jonathan's son was weak in his feet"); no marker.
• 0566A · In Hierusalem. — Migne's spelling of the lemma against the Clementine's Jerusalem. Orthography, and Migne uses both forms in this work (Hierusalem at 0566B and 0574C, Jerusalem everywhere else). Kept as printed at each site.
• 0567C · Nisi circumcidamini secundum morem Moysi, non potestis salvi fieri. Act 15:1 has non potestis salvari. A periphrasis with the identical sense and no English exponent — both render "you cannot be saved". Recorded here rather than marked, and the English is not contorted to make a Latin-internal difference visible.
• 0569C · Percussit David Philisthiim, et humiliavit eos, et tulit Geth et filias ejus de manu eorum, against 1 Par 18:1 ut percuteret…humiliaret…tolleret…de manu Philisthiim. The mood shift follows from clipping the governing ut; eorum is a pronoun for the noun it replaces, same referent. No marker.
• 0570C · Hic ne ingredietur domum meam? against 1 Rg 21:15 hiccine ingredietur domum meam? ⚠ A negation trap, and it is not a negation. Migne's ne is the interrogative enclitic detached from hiccine, not the negative particle: rendered "Shall this man enter my house?", with no negative anywhere in the English. Reading it as non would invent a negative the plate does not carry and reverse Achis's taunt. No marker: the divergence is one orthographic split of a single word, which is exactly the class the README says not to adjudicate from this file. The tail Abiit inde David et fugit in speluncam Odollam is 1 Rg 22:1 with inde David for ergo David inde — same class.
• 0571B · fons aquae salientis in vitam aeternam, quam qui biberit non sitiet in aeternum (Joan IV). A conflation of John 4:14 with 4:13, both printed accurately, joined by quam. Same family as the usque ad abridgment (8970 precedent: an abridgment is not a variant). No marker.
• 0572C · Si crederetis Mosi… — John 5:46 has Si enim crederetis Moysi. A dropped connective and a name-spelling.
• 0573A · Dominus transtulit peccatum tuum, — 2 Rg 12:13 has Dominus quoque transtulit. Dropped particle.
• 0573A · quia multi venient ab Oriente et Occidente… — Matt 8:11 reads multi ab oriente et occidente venient, et recumbent cum Abraham, et Isaac, et Jacob. Word order and two dropped et. Free quotation, nothing asserted differently.
• 0573C · Tulit David coronam Melchon… — 1 Par 20:2 Tulit autem David coronam Melchom. Dropped particle, name spelling. The gloss then argues Melchon is not a proper name at all, so the form is load-bearing for the gloss and is kept exactly.
• 0576A · Dixi, Tu et Siba dividite agrum. — Migne labels this In Hebraeo, not as a Vulgate quotation (2 Rg 19:29 has possessiones). Pattern 14 does not reach a reported Hebrew reading; the same holds for cessavit rex exire post Absalom (0573D), An ignoras quod amarus erit finis… (0564D), Traduxit eum in Mahanaiim (0564D), Fortissimus ipse cujus est cor quasi leo pavore solvetur (0575B), infectis (0575D) and Ecce servus tuus (0575D). Seven reported-Hebrew readings, none marked, all rendered as printed.
• 0576A · usque ad: Ephraim et Manasses tanquam Ruben et Simeon erunt mihi. Gen 48:5 reads sicut Ruben et Simeon reputabuntur mihi. Two synonym swaps inside an abridgment tail, and erunt mihi echoes the same verse's own mei erunt. No marker.
• 0576D · Vana salus hominum. In Deo faciemus virtutem, et ipse ad nihilum reducit tribulantes nos cited (Psal. CVII). ⚑ The words are Ps 59:13–14, not Ps 107 — Ps 107:14 ends deducet inimicos nostros, while tribulantes nos occurs only at Ps 59:14. Against its real source the divergences are hominum for hominis and reducit for deducet. No [var:]: measured against the verse Migne actually quotes, what is left is a number and a tense. The wrong reference is a citation misprint and belongs in data/citation-corrections.json at index time, never in the running text (Ruth precedent).
• 0575C · [n: (Prov. XI; XXV)] on Qui enim fodit foveam…Et insidiis suis capientur iniqui. The first half is Prov 26:27, the second Prov 11:6. Same class: citation repair, not text repair. Left as printed.
• 0577B · Haec sunt autem. — 2 Rg 23:1 reads Haec autem sunt verba David novissima. Word order inside a two-word lemma; and 0577C Haec sunt nomina. adds sunt to 23:8's Haec nomina fortium David. Neither changes an assertion.
• 0577C · Usque ad tres primos non pervenit. (roman, not italic) — 2 Rg 23:19 has non pervenerat. ⚑ Another usque ad look-alike: these are the verse's own words and are rendered as ordinary prose, not as the abridgment formula. Tense only; no marker. The same holds for usque ad excellentiam divinae Trinitatis non attingunt (0577C) and Praecidit tunicas usque ad inguina (0570D). 48 occurrences of usque ad in the work; four are not the formula, and all four were decided from what stands on either side.
• 0577C · Quis enim in nubibus aequabitur Domino in filiis Dei? (roman) — Ps 88:7 reads Quoniam quis in nubibus aequabitur Domino; similis erit Deo in filiis Dei? Migne drops the middle member. Abridgment family, no marker.
• 0579A · Et Bana. — 2 Rg 23:20 has Et Banaias filius Jojadae, and the gloss below spells Banaias in full four times. The lemma prints the shortened form with a stop, which reads as Migne's abbreviation of the name. Rendered as printed ("And Bana.") — the Ruth ruling on Liam/Lia governs: Migne's name-forms are kept everywhere. Logged, not marked.
• 0579A · de Capseel, for 23:20 de Cabseel; 0578B Semma filius Age / Semeiam filium Heli / Achimea beside 0577D's Achimaea; 0579D Semmaa filius Agge. Name-form variance inside one work. Each kept as printed at its own site — the Latin column carries the variance and normalizing it would be our hand.
• 0580B · Porro angelus Domini stabat…proni in terram, juxta aream. — a conflation of 1 Par 21:15b with 21:16, plus vidit angelum Domini for vidit angelum Domini stantem and juxta aream carried down from 21:15 to close 21:16. Abridgment family; the dropped participle changes nothing. No marker.

3.3 The Hebrew-etymology italics — rendered as the Latin says, italics 1:1

These are glossed Latin equivalents of Hebrew names, not scripture lemmata, and 7a″ does not apply to them; they are rendered as the Latin says and never conformed to a lexicon: conjugium → wedlock (0564B) · filius confusionis → son of confusion · patris lucerna → the father's lamp (0564C) · castra → camps (0564D) · inimicorum divisio → the enemies' division (0567B, with the column anchor kept between the two words by using the possessive) · flentium → of the weepers (0567B) · ex patre → from the father · sanguinis poculum → the cup of blood · sublimis / humecta → lofty / moist (0569D) · interficientes / vivificantes → slaying / making alive (0570B) · pater pacis / patris pax → father of peace / the father's peace · donans → giving · amaritudo → bitterness (0573D) · festinans → hastening (0574D) · ruina fratris → the ruin of a brother (0575B) · pax pacis → the peace of peace (0575C) · columbae domum → the dove's house (0575C) · suspirans → sighing (0575D) · lacus → a pit (0576D) · occidentes / vivificantes → killing / making alive (0579A) · leo Dei → the lion of God (0579A) · aedificator Dominus → the Lord is builder (0579B) · lumen nobis → a light to us (0580C) · puteus septem → a well of seven (0572A) · lux mea Dei → my light of God (0572B).

⚠ Two pairs are deliberately not harmonized with each other: interficientes/vivificantes (0570B) reads "slaying / making alive" while occidentes/vivificantes (0579A) reads "killing / making alive" — Migne prints two different Latin words for the Cerethi, and one English for both would erase that. Within a work: one voice; but two words are two words.

§4 · CRUCES — the plate, and what was declined

4.1 Non-word candidates — three, PLATE READ RUN AND SETTLED (orchestrator, 2026-08-24)

⭐⭐ VERDICTS FIRST — the plate was read the same session, so nothing below is outstanding. PL 113, archive.org patrologiaecurs04migngoog on disk, page map re-calibrated against the printed corner numbers (p. 287 = cols 563/564) before any crop, and every crop taken deliberately wider than its target word.

⚑ The non-word rule holds and is now 12 for 12: both non-words were ours, and the one form that was a name rather than a non-word was Migne's — the same split the addendum records. Firing no [sic:] was right at all three sites: two of them would have blamed Migne's plate for Corpus Corporum's letters, and the third would have asserted a defect in a word the plate prints exactly as we hold it.

⚑ The TEI newline agreed at 0563C and disagreed at 0567A (it falls after precaepit, exactly where Migne breaks the line; at the second site it falls after Samua, while the plate breaks Sa-/mua). Corroboration after the plate has spoken, never a substitute for it.

⚑ The work has been re-chunked so the patched Latin is what the page prints; verify-english passes unchanged, chunk boundaries unmoved, and the English needed no edit at either site — it had rendered both words for their evident sense ("commanded", "of the names"), which is what 7a⁗ requires and what left these two repairs free to be made in the Latin alone.

The candidates as the translator filed them

Per the 2026-08-23 addendum, a form that is no Latin word at all has been our digitization's corruption 10 times out of 10, never Migne's. Each is rendered for its evident sense, logged here, and carries no [sic:], because that marker would blame Migne's plate for what is probably Corpus Corporum's.

• 0563C · precaepit for praecepit (et ob hoc precaepit verus David). A vowel transposition; praecepit stands correctly at 0563B, eleven lines above, in the same gloss. Rendered "commanded". Non-word — worth the plate.
• 0567A · interpretationes neminum. for nominum (quibus congruunt interpretationes neminum: Samua, audacia; …), followed immediately by nine name-etymologies. neminum is no form of any Latin word. Rendered "the interpretations of the names". Non-word — worth the plate.
• 0578B · Achimea beside 0577D's Achimaea for the same man in the same argument. This one is a name-form, not a non-word, and is kept as printed at both sites; named here only because it sits four words from the next candidate and §C of the addendum says to look at the neighbours when you look at anything. ⚑ Deliberately NOT plate-read (orchestrator): a name-form is the real-word class, where the plate has refuted the suspicion four times out of four and where corpus frequency — including the work's own inconsistency between the two spellings — is worthless by the addendum's own rule. It fires no marker, so shipping asserts nothing about it. The read stands available at p. 294 (cols 577/578) if it is ever wanted.

⭐ Recording the negatives too, as the addendum asks. Three real words that read oddly were suspected and refused a marker on the addendum's rule that a real word is Migne's until a plate says otherwise: habeat (0563A), rectorum (0566C), donum (0567B) — see 4.2.

4.2 Real words that read oddly — rendered as printed, [cj:] DECLINED, reasons given

• 0563A · Diadema autem et armillam a patre commendatam habeat. A jussive/potential subjunctive in a run of perfect indicatives (fuisse, fuit). Rendered literally, "But let him have the diadem and the bracelet entrusted to him by his father." Declined: 18a's bar is that the English asserts something false; an odd mood asserts nothing false, and the ruling explicitly leaves mood and person anomalies that merely fail to construe declined (11064 @1021A, @1154C).
• 0566C · Domatum. Id est rectorum. tectorum ("of the roofs") is the obvious suspect for a gloss on domata, and the sentence that follows is flatly literal (lead pipes, aqueducts), which argues for it. Declined, and the rival is named because it is strong: the Glossa allegorizes exactly this way three lines later (per Joab, id est, praedicatores*), so "the housetops, that is, the rulers" is a live reading in this author's own idiom, not a defect. A real word plus a good argument is precisely the combination the addendum measured as refuted four times out of four. Rendered "That is, of the rulers."
• 0567B · Qui est gradientis, id est donum in carne advenientis — sonum is the obvious suspect, since sonitus stands twice in the same paragraph and the lemma is sonitum gradientis (5:24). Declined: "the gift of him coming in the flesh" is not a false claim in this author's mouth; and neither donum (neuter) nor sonum (masculine accusative) agrees with Qui, so the conjecture does not even repair the grammar. Rendered as printed.
• 0571C · sic nonnunquam concepta intrinsecus compellit sibi extrinsecus oculus deservire. oculum is required by compellit … deservire; oculus leaves two nominatives. Rendered on the structure the sentence's own sicut … sic pivot demands — temptation drawn IN through the eyes, then temptation conceived within driving the eye outward — i.e. "compels the eye without to serve it". Declined: a case slip with no English exponent (Pattern 9 family); reading oculus as the subject would reverse Gregory's pivot, which is the one thing the structure forbids.
• 0572D · istae per vitae innocentiam in prophetia exprimit culpam — feminine plural with a singular verb, for iste (Urias, answering ille = David). Rendered "and this man…". Declined: Pattern 9 — Latin grammatical gender and number agreement have no English exponent here, and the antecedent is fixed by the ille … iste pair.
• 0579C · Neminem moverat… where moveat ("let it move no one") is expected. Rendered "It had moved no one that…". Declined: a tense, not a false assertion.

4.3 The one genuinely undecided passage

0575B · In Hebraeo: «Fortissimus ipse cujus est cor quasi leo pavore solvetur,» subaudis non fortissimus iste David intelligitur. Unde: «Scit enim.»

Two parses, and the whole sense turns on where non attaches:

1. taken: subaudis "non"; fortissimus iste David intelligitur — "supply 'not'; that most valiant one is understood to be David." Rendered understand "not", that most valiant one is understood to be David.
2. rejected: subaudis: non fortissimus iste David intelligitur — "understand: that most valiant one is not understood to be David."

Why (2) lost. (a) subaudi(s) takes the word to be supplied as its object throughout this work — 0566D prints subaudi murum and subaudi alium murum, the same construction two columns earlier, and that is a control of the same class. (b) The clause Migne quotes next, Scit enim (2 Rg 17:10, scit enim omnis populus Israel fortem esse patrem tuum), identifies David as the valiant man, which reading (2) denies four words earlier. Reading (1) makes the Hebrew say that David, the valiant, will not melt with fear, and Scit enim is then adduced as its proof.

⚠ Recorded as a crux and not as a marker: the printed word is non, the English carries a negative in the position the printed word occupies, and nothing here is a claim about the plate. The straight quotes around "not" are ours (Pattern 6) and are the only added punctuation in the work besides §5's one licensed colon.

4.4 Rough syntax rendered as it stands, nothing supplied

• 0566D · Haereticorum ecclesias quae … gloriantur. An accusative with no governing verb — the head of a sentence the abridgment has decapitated. Rendered as the noun phrase it is; no verb supplied (the standing usque ad rule, and it applies to every abridged fragment).
• 0570D · qui Ammonitarum, id est, malignorum spirituum, id est, rector populi moeroris… A genitive chain with rector stranded in the nominative. Rendered clause for clause. No word supplied.
• 0575D · ad explorandum factum Absalom et David renuntiandum decreti fugientes Absalom — a gerundive chain the abridgment has cut. Rendered as it stands.
• 0576C · The chunk, and CAPUT XX, end mid-clause (id est Christum Filium Dei rediit, no stop). Left mid-clause.
• 0564A · ita illi fortitudinem dabant subjectis, a quibus Saul et Jonathas interfecti sunt quasi non a rusticis. The referent of a quibus is genuinely open (the fortes of the lemma, or the subjecti). Rendered word for word so that both remain open, as they are in the Latin. Nothing was resolved silently.
• 0563C · Alii excelsi — an elliptical "others [render it] the high places". Rendered "Others, high places", with no bracketed supplement.
• 0565C · Antiqui delicata dicebant diis consecrata, quae nunc dicimus dedicata. Migne sets no italics on delicata/dedicata, and the gloss IS the etymology, so both Latin words are carried in roman exactly as he sets them: "The ancients called things consecrated to the gods delicata, which we now call dedicata." Italicising them would have added two spans the plate does not have.

4.5 Unexpressed subjects that were supplied — named, per the brief

Every "he/she/it/they" not in the Latin was fixed from the lemma above or the verse it comes from, not from the sentence:

• 0563B–C · vult / praecepit / precaepit — the subject is the true David (Christ), named in the lemma's own gloss at 0563B and again at 0563C; not Saul, not the Philisthines.
• 0564D · traduxit … constituit — Abner, from 2 Rg 2:8's Abner … tulit Isboseth … et circumduxit eum, which is the lemma. Not Isboseth.
• 0566D–0567A · mittit … aedificatur … cognoscitur — Hiram, from the lemma Misit quoque (5:11); the gentilitas he figures is feminine and takes "it".
• 0568C · quem dum Michol irridet, gloriosior ancillis apparet, quae in typo Synagogae sterilis permansit — the last relative is Michol, not the handmaids: sterilis permansit is 2 Rg 6:23, Michol's barrenness, the verse this book already glossed at 0565B. The English names her by keeping "she" bound to Michol and moving the handmaids into a prepositional phrase.
• 0575A · dicit ad David — Semei, from the lemma Semei autem (16:13), against the nearer Jeroboam.
• 0578C · sedit … desideravit … reprehendit … libavit — David, named two clauses earlier (David vero alienam uxorem concupivit); Gregory's vir flagellis eruditus is the same man.
• 0579B · interfecit — Banaias, and then in the Hebrew reading the lion is Moab; the two readings are kept distinct as Migne keeps them.

§5 · TESTS RUN, AND THE SECTIONS READ CLEAN

1. Column anchors — position, not merely presence (brief 2b). All 72 anchors verbatim and in order; each checked against the word it stands beside in the Latin. Three sit inside a construction and were split to match: usque [0564A] ad → as far [0564A] as; inimicorum [0567B] divisio → the enemies' [0567B] division (the possessive was chosen precisely so the anchor keeps its two neighbours); Christo [0577B] Dei Jacob → the Christ [0577B] of the God of Jacob. Seven anchors set hard against a VERS. address keep the missing space.
2. Notes. 115 [n: …], byte-identical to the Latin twins in content and order. No [nt:]: every note in this work is a citation or a siglum, not editorial prose — including the long ones, which are attributions with book and chapter ((RAB., ex Aug. de Civit., lib. XVII, cap. 8.), (GREG., lib. XXXIII Moral., cap. 10.)). Judged by content, not length.
3. Marker-strip read. Every [var:]/[cj:] deleted and the English read back: no stranded article or preposition, no doubled punctuation, every clause keeps its verb, its negation, its conjunction and its object (the fourth question 7a⁗ does not ask). All seven markers are additive — none stands in place of a rendering.
4. Per-negative host check + 2a. Every printed negative walked (99 particle tokens in the eight chunk bodies), each attached to the single Latin word it governs and that word found in the English. The 2a population (a negative on a verb with a quantifier or correlative nearby) fired at six sites, each proved from structure rather than from what reads well: nec solum … sed etiam (0571A, on solum) · Nec pro seductione … sed pro adulterio (0574C, on the causal phrase, not on ejectus est) · Non enim … intantum humiliabantur, ut … sed adeo (0569C, on the degree) · Non enim David … studiose respexit, quia concupiverat, sed potius (0571C, on the whole causal clause) · non nisi tertio (0576D) · in summa non triginta, sed septem et triginta (0579C, on the number).
5. Punctuation, band by band (brief 3, never by total). Colons, semicolons, question marks, guillemets and exclamations compared inside every anchor-partitioned band of all eight chunks (80 segments). Zero mismatched bands, with two declared exceptions: (a) 0566A, one colon the Latin does not print, licensed by Pattern 8b — Quaeritur cur non in summa quadraginta anni et sex menses numerentur? is an indirect question closed with ?, recast as "It is asked: why are not forty years and six months reckoned in the sum?" so the mark survives in a sentence that can hold it; (b) 0575B, the two straight quotes around "not" (§4.3), which are ours and free under Pattern 6. Nine semicolons and four colons introduced on the first draft were reverted to the plate's commas in the course of this compare. All 16 question marks in the work are Migne's, in place, and the one direct question he closes with a full stop (0576B) keeps the full stop.
6. Italic spans. 275 spans, 1:1 in place. The three etymological pairs and the two Greek-derived glosses (πῦρ at 0567B, kept in Greek script) are unmoved.
7. usque ad decided per occurrence. 48 occurrences: 44 are the abridgment formula and are italic in Migne, rendered as far as; 4 are the words of the text itself and are rendered as ordinary prose — 0565B (2 Rg 6:23), 0570D (usque ad inguina), 0577C (Usque ad tres primos non pervenit), 0577C (usque ad excellentiam divinae Trinitatis). No abridgment reading was allowed to swallow a lemma's own words.
8. The fourth test — each finished paragraph read back against the Latin asking only does it assert what he asserts. This is what produced §2.1 (the obedientiam site, which every count check passes), §4.3 (the subaudis non parse, which reads smoothly either way), and the Domatum/rectorum decline. Read clean and reported so: the whole of CAPUT VII–IX (0568D–0570B), CAPUT XI's long Gregory extracts (0571C–0572D, the densest prose in the work and the one most likely to hide a smoothed connective), CAPUT XXI–XXII (0576D–0577A), and CAPUT XXIV (0580A–0580D) apart from its census number.

5.1 Candidates raised and REJECTED, with reasons

• Chapter-head numerals — CAPUT II. was not turned into "CHAPTER TWO." The launch brief offers that form, but its own tiebreaker is Ruth's shipped English, and Ruth ships ## CHAPTER II. beside ## CHAPTER ONE. Following the plate's shipped precedent, not the paraphrase of it.
• [f: …] inline citation tails — not tagged anywhere. This is a lemma-and-gloss commentary, which the runbook's false-positive 4 rules out explicitly. The one italic locator in the work, (I Paral. [0569C] XVIII.) at 0569C, is carried verbatim and untranslated inside its italic span, exactly as Migne sets it, and takes no marker.
• Column-band arithmetic — not reported. The bands run 0563A–D, 0564A–D, … with no gaps and no text broken mid-word across an anchor (runbook false positive 1).
• Quid mihi? (0574D ×2) vs Quid mihi. (0575D) — not reported as an inconsistency. Migne prints the mark twice and omits it once; both ride as printed. The lemma was not expanded to the Vulgate's Quid mihi et vobis, filii Sarviae? because "What is it to me?" is complete English on its own — the ⚑ class distinction between a clipped lemma and a complete short one.
• Nonne. (twice) and Qui abscissum. — expanded under convention 3 to "Am not I." and "Who cast forth the severed head.", because neither renders as English word for word. Both occurrences of Nonne. were expanded identically, per the ruling that a work treating one bare lemma two ways is resolved by expanding the stump, never by contracting its twin.
• Tenerrimus ligni. — expanded to "The most tender little worm of the wood." for the same reason (23:8 quasi tenerrimus ligni vermiculus), and the gloss at 0579A picks the phrase back up as ipse ligni vermiculus, which fixes the expansion rather than guessing at it.
• Migne's ( sic ) — none in this work. Nothing was stacked.
• Foot-of-page conjecture notes ([cn:]) — none present in our Latin, as everywhere in the PL block. This work's plate has NOT been read; that is a gap, not a checked zero, and it is the difference between this entry and Ruth's §1.5. The three plate candidates in §4.1 and the one suspicious tidiness below are what a reader should look at first.
• Suspiciously tidy forms (brief §4 / addendum C) — one raised: 0569D Maseth ancillae Abrahae filius, where Masek (Gen 15:2 LXX, Abraham's handmaid) is the expected form and Maseth is not a form this corpus otherwise carries. Kept as printed; named here because a reader going to the plate for §4.1's precaepit is one page away from it.

§6 · THE BLIND POLARITY READ (2026-08-24) — five sites, four repaired

A reader fenced from this file — it may not be opened before findings are fixed, because every agent given it in the 2026-07-28 sweep returned "all already logged" and found nothing — read all eight chunk pairs. Five sites, all five new to this file, four of them repaired in the English. Its rejections agreed with §4's declines site for site, which is the check that the two passes were independent and not merely duplicated.

⭐ Four of the five are the same class, and it is not negation. The batch's briefs named the unexpressed subject as the dominant defect; what actually beat this work was the referent of a word English cannot leave floating — a supplied pronoun, a genitive re-hosted, a predicate nominative fused into a compound. Parity, counts and negation-host checks all passed at every one of them. They failed only the last test: does it assert what Migne asserts?

6.1 · 0574C — a supplied "he" that convicted the wrong man (repaired)

…quo transacto expulit patrem de regno. Nec pro seductione et interfectione sacerdotum ejectus est de regno, sed pro adulterio et homicidio, Nathan cuncta praedicente.

The English read "…he drove out his father from the kingdom. Nor was he cast out of the kingdom for the seduction and the slaying of the priests, but for adultery and murder" — and that "he" inherits Absalom from the sentence before, so the page said Absalom was cast out for adultery and murder. Migne's unexpressed subject is the father: seductio et interfectio sacerdotum is Nobe, and adulterium et homicidium … Nathan cuncta praedicente is Bathsheba, Urias, and Nathan's sentence — which is the very expulsion Absalom carried out.

Now "Nor was his father cast out…". ⚑ The resolution deliberately uses Migne's own printed noun (patrem, two words earlier) rather than importing the name David: the least supplied word that fixes the referent is the right one.

6.2 · 0579A — a genitive moved off its host (repaired)

ad has tres virtutes nemo fortium David pervenit. The English read "these three virtues of David, none of the valiant men attained", taking David with virtutes four words back, across a complete prepositional phrase and a new subject. fortium David is contiguous and is this chapter's own standing phrase (0577C catalogus virorum fortium; 2 Rg 23:8 Haec nomina fortium David). Now "unto these three virtues none of David's valiant men attained." As it stood the English asserted that the virtues were David's; Migne asserts that the men are.

6.3 · 0568A — a predicate nominative fused into a compound noun (repaired)

__Aqua__ Gethaeus populus potest appellari, qui dicit… The English read "the Gethaean can be called the water-people", which dissolves the predication the sentence exists to make and reads smoothly over a rough spot — the tell the fourth test is built on. Now "the Gethaean people can be called water, who says…", which is what carries the Gal. VI tag that follows.

6.4 · 0578A — an invented predication removed, and the plate read (repaired)

…verecundiam castitatis incontinentibus quasi legentes gladio oris contra luxuriam defendit. The English supplied a clause the Latin does not license: "as though they were gleaners" — a plural subject with no available referent, manufactured to smooth a rough spot. Now "as though gleaners."

⭐ The reader also proposed lentes for legentes — a good argument: the verse glossed is 2 Rg 23:11, ager plenus lente, and the same sentence has already read the crop allegorically (infirmos velut hordeum tuetur). I read the plate. Migne prints legentes, exactly as we hold it (PL 113 p. 294, right column, 450 dpi). The conjecture is therefore about Migne's own text, not our source, and it stays in this crux and out of the apparatus: the sentence as printed asserts nothing — legentes agrees with nothing in it — so it misses Pattern 18's bar, exactly as 8958 @0511D did, where declining beat a conjecture that the plate then refuted. ⚑ The real-word class is now refuted 5 times out of 5 at the plate.

6.5 · 0569B — a case variant English cannot carry (crux only, body unchanged)

Ex Aegypto. Gente et Deo ejus vel gente et Deum ejus, gentem Israel, et Deum ejus Moysen… Migne records a construction variant — ablative Deo against accusative Deum — and the English necessarily prints the same six words twice, so on the English side alone it reads as dittography.

Nothing is changed in the body, and no marker fires. This is the exponent rule as 8992 settled it: signa was marked because English shows number, remanerent was left because English cannot carry a verb ending at all. A Latin case ending is the second kind. The parallel Latin column stands opposite and shows the difference, which is the whole reason this edition prints both. ⚑ The second limb is the one the continuation picks up (gentem Israel, et Deum ejus Moysen).

6.6 · Promoted from §3.3 at the reader's prompting

0575C ideo Absalom pax pacis dicitur, quia pacem pater habuit quam ille non habuit. The gloss's own reason (pacem pater habuit) and this work's other etymology of the name at 0573D (pater pacis vel patris pax) both require patris pax. It was filed only in the etymology-rendering list; it belongs on the record as a defect of Migne's, reproduced — the English renders pax pacis literally, which is correct and stays.