Epistola ad Titum

Translated 2026-08-18, one translator, 1 chunk (1,125 Latin words → 1,719 English, 1.53×). verify-english.mjs clean. Apparatus: 2 [var:] · 0 [sic:] · 0 [cj:] · 0 [d:] · 0 [ed:] · 0 [nt:] · 3 [n:] (exact parity with the Latin twin). 14 column anchors, 47 italic spans, 4 heads, 0 guillemets, 0 question marks — all 1:1 with the Latin.

Conventions are the Ruth pilot's (src/english/8968/cruces.md §1) and are not re-opened: VERS. n.-- passes verbatim, the lemma is Englished, a fragmentary lemma is rendered as the phrase it stands for, the attribution sigla ride as ordinary [n: …].

§1 · THE LEMMA DIVERGENCE LIST — EVERY LEMMA CHECKED (7a″)

All 47 italic spans were checked word by word against the Clementine Vulgate of Titus. Agreements are listed too, so this can be told from a report by an agent that never looked. Two divergences were material enough for a [var: …]; five more are recorded and took no marker, with the reason.

Chapter I

Chapter II

Chapter III

Two spans are not lemmata at all and were checked as such: usque ad ×2 (0639B, 0639C), Migne's abridgement formula, rendered as far as after the shipped precedent at 11436/0002 col. 0350A. They keep their italics.

The two [var: …] markers

0637D · Secundum fidem electi. — the Vulgate reads secundum fidem electorum Dei. Migne's electi is a real Latin form and the gloss immediately construes it as the nominative plural — quam tenent vel qua salvantur electi Dei — so the divergence is not a slip the gloss ignores; the gloss reads the lemma as printed. English the elect is number-neutral and carries either. Marker: [var: Vulg. electorum Dei, "of the elect of God"].

0640D · gratia Dei et Salvatoris, etc. — the Vulgate has gratia Dei Salvatoris nostri (Tit. II, 11), one person, no et. Migne's et would ordinarily split it into two, and the gloss then binds it back into one — id est Deus et Salvator noster, scilicet Dei Filius. The added conjunction is therefore load-bearing for what the gloss does with it, which is the Pattern 14 test. Marker: [var: Vulg. gratia Dei Salvatoris nostri, without et].

The five divergences DECLINED, with reasons

Pattern 14 fires on divergence "in a way that changes the sense". These five do not, and marking them would be noise on a page that carries two real ones:

1. 0639A Unius uxoris virum vs Vulg. unius uxoris vir — accusative for nominative; the form is that of I Tim. III, 2, from which the phrase is habitually quoted. English has no exponent for the case (Pattern 9's own argument): "the husband of one wife" either way. Recorded, not marked.
2. 0639B esse sine crimine vs Vulg. sine crimine esse — word order inside a fragment the gloss has already recast as its own syntax (necesse est … eum esse).
3. 0640B Non vino multo vs Vulg. non multo vino servientes — adjective/noun order. Sense identical; the gloss turns on the wine, not the order.
4. 0641A Eramus enim et nos aliquando vs Vulg. Eramus enim aliquando et nos — et nos fronted. "For we also were once" renders either.
5. 0638D Haec est vita aeterna (Joh. XVII, 3, Haec est autem…) and 0639D Vidit Deus… (Gen. I, 31, Viditque Deus…) — a dropped autem and a dropped enclitic -que in two proof-texts quoted inside the gloss. Both are connectives at the head of a quotation lifted out of its sentence; neither changes what the verse asserts.

§2 · PRESBYTERI AND EPISCOPUS — THE GLOSS TREATS THEM AS ONE OFFICE, AND SAYS SO

This was the standing instruction not to resolve. The gloss resolves it itself, in as many words, at 0639C:

Hic aperte ostendit presbyterorum nomine episcopos supra fuisse designatos.

Rendered "Here he plainly shows that under the name of presbyters bishops were designated above." The chain is unbroken across three verses and one column: VERS. 5 glosses Presbyteros on why the plural is used (ut … pastoralis curae onus facilius sustineatur, per multos divisum); VERS. 7's Oportet enim — whose Vulgate continuation is episcopum — glosses it as irreprehensibilem episcopum; and then the sentence above states the identification outright. The gloss treats them as one office.

What the English does about it: nothing, deliberately. Presbyteros → presbyters, episcopum → bishop, presbyterii (Argumentum) → presbytery, each rendered as it stands and kept lexically distinct throughout, so that the identification the gloss makes is visible as the gloss's own act and not as our vocabulary. "Priest" was considered and rejected on the brief's own ground: it imports the judgement into the word. A reader who wants the identification finds it at 0639C, in Migne's sentence, where it belongs.

⚑ Note also that the Argumentum already frames the epistle as being de constitutione presbyterii — the presbytery, not the episcopate — which is why VERS. 7's identification is the interesting sentence in the chapter rather than a throwaway.

§3 · THE UNEXPRESSED SUBJECTS — EVERY SUPPLIED PRONOUN, NAMED

The brief's dominant defect class. Every "he/she/it/they" the English supplies is listed here with what fixed it. The household code (ch. II) is where the subject shifts fastest.

The Argumentum · commonefacit et instruit — "He admonishes Titus." Paul. Fixed by the closing participle scribens ei a Nicopoli, which only the letter's author can govern, and by the form itself (this is the standard prologue to a Pauline epistle).

0637D · Quidam nobilitate — "Certain men by nobility": quidam is expressed, and the exempla that follow (Moses, David, Mary) fix it as persons, not the abstract conditio.

0639A · Primo dicit quid agere debeat et quales ordinare presbyteros — two different subjects in one clause. dicit = Paul (he is the one saying); agere debeat and ordinare = Titus, from Tit. I, 5 reliqui te Cretae … ut constituas presbyteros. Likewise Dum haec recolit … monet exsequi: Paul recalls, Paul admonishes, Titus carries out, and sicut ei disposuerat = as Paul had appointed for Titus (sicut et ego disposui tibi).

0639B · quotidie fieri non quiescunt … quae … remittantur — "they do not cease to be committed daily." The subject is sins, neuter plural, and it is not printed: it is lifted out of sine peccato two clauses earlier, and confirmed by the neuter plural relative quae … remittantur. Nothing else in the sentence is plural.

0639B · nec frontem habet ad alios talium cohabitator — the subject IS printed (cohabitator); the English keeps it as a full noun phrase ("one who dwells with such persons") rather than a pronoun, because a bare "he" here would attach to the filii of the lemma instead of to the father who would be ordained.

0639D · Hoc ipsi negant — "This they deny." ipsi is expressed but empty: its referent is the Judaizers of Tit. I, 10–14 (maxime qui de circumcisione sunt, mandatis hominum), which VERS. 14 has just glossed one paragraph above. Not the mundis of the lemma.

0639D · intelligi voluit — "he willed this to be understood." Moses, from quaedam Moyses prohibuit in the immediately preceding clause; the whole sentence is about what Moses' prohibitions were for. Not God, though quas Deus creavit stands inside the same sentence — and that is exactly the trap: God is the nearer noun and the wrong subject.

0640B · Sunt enim plurima quae … vix fieri patiantur — "which at an age of this kind they scarcely suffer to be done." ⚠ Two readings construe and the choice is worth a second eye. (a) senes (the aged men of the lemma) are the subject of the deponent patiantur: "there are very many things which men of that age scarcely allow to be done." (b) plurima is the subject: "very many things which scarcely allow themselves to be done." (a) was taken, because patior deponent is active in meaning and wants an animate sufferer, and because the whole [n: (CHRYS.)] note is an inventory of what old age does — but (b) cannot be excluded from the printed words alone.

0640B · earum incessus, motus, vultus, sermo — "their gait": the aged women (anus), from Tit. II, 3 anus similiter in habitu sancto, which is the lemma's own verse. Feminine plural genitive fixes it; the senes of VERS. 2 are masculine.

0640B · Solent enim esse quaedam — "certain women": quaedam feminine plural, still the anus, continuing from Non vino multo (Tit. II, 3).

0640B · Quos … diligunt, ament, VERS. 4 — "whom they love … let them love": the young women (adolescentulae), from Tit. II, 4 ut prudentiam doceant adolescentulas, ut viros suos ament. The subject changes from the previous paragraph's anus at the verse boundary and nothing in the sentence marks it: viros suos does.

0640C · Hortare senes, et anus, et adolescentulas — the three groups are expressed; no supply needed. Same at 0640C VERS. 10 (ad senes, et anus, et servos), where the addition of servos is what makes the paragraph a summary of the whole household code.

0640C · ut ita ornent … doctrinam Salvatoris — "that they may adorn": the subject is omnes praedictos in the same sentence, i.e. all three groups. Expressed.

0640C · quantum in ipso est occidit eos — "he kills them": the man who lives ill (Omnis … qui male vivit, expressed) kills those over whom he is set (eorum quibus praepositus est, expressed). Both printed; no supply.

0640D · veniet praemia daturus — "Who will come": Christ, from magni Dei et Salvatoris in the lemma; the relative Qui is printed and its antecedent is the lemma above, per the brief's own rule.

0641A · Per quam dona nobis spiritualia abunde tribuit — "he bestowed": God our Saviour, from Tit. III, 4 benignitas et humanitas apparuit Salvatoris nostri Dei, of which the lemma Benignitas is the opening word. The relative quam is benignitas itself.

0642A · mandat Tito, ut priusquam veniat Nicopolim, praemittat eos — three subjects, two of them unexpressed. mandat = Paul (the sentence's per epistolas Pauli names him, and only he charges Titus); veniat and praemittat = Titus, named as the addressee of mandat Tito and fixed by Tit. III, 12, where it is Titus who is to hurry to Paul at Nicopolis. The trap here is veniat: read of Paul it would say Paul comes to Nicopolis, which the Vulgate does have him doing (ibi enim statui hiemare) — but then praemittat would have no subject at all, and the point of the sentence is that Zenas and Apollos are to be sped on their way before Titus leaves.

0642A · quia hujus professionis erat in Synagoga — "he was of this profession": Zenas, the lemma's own subject (Zenam legisperitum), not Apollo, who enters the sentence only as the term of comparison and gets his own lemma next.

§4 · THE FIXED WORD-PAIRS THE BRIEF NAMED

• munda / immunda (0639D) — clean / unclean, rigid across all four occurrences, including ab immundis munda discrevit → "separated clean things from unclean", where the two members meet. ⚑ The brief also named pollutis / incredulis for this verse: neither word occurs. Migne's gloss on Tit. I, 15 stops at Omnia munda mundis and never quotes the verse's second half (coinquinatis autem et infidelibus). Recorded as a checked absence, not an omission.
• crimen / peccatum (0639A–B) — crime / sin, and this pair carries the whole paragraph: Si dixisset, sine peccato, nullus … posset ordinari is unintelligible unless the two words stay apart in English. Kept apart at all seven occurrences, including the return of esse sine crimine at 0639C.
• sobrii (0640A) — sober. The brief anticipated sobrii / sobrias / sobrie across ch. II; only sobrii is printed (Tit. II, 5's sobrias and II, 12's sobrie are in verses this gloss does not touch). One word, one rendering, nothing to link.
• sanam doctrinam / doctrinam Salvatoris (0640A, 0640C) — one doctrina, one English word: sound doctrine / the doctrine of the Saviour, and doctrine again at both non-lemma occurrences. Not varied.
• benignitas / humanitas (0641A) — humanitas is not printed. Migne's lemma stops at Benignitas, etc. Rendered Benignity rather than "kindness", deliberately, so that the English word for humanitas stays unspent should the pair ever be quoted whole elsewhere in the Glossa block.
• sectator / sectantes (0640D) — a pursuer of good works / pursuing a pure life, kept on one root, because the gloss's argument is that the second is what the first means.
• contradicunt / contradicentes / contradicere (0639C) — all contradict; arguere → reprove, redarguere → confute, kept distinct so that the lemma's own verb and the gloss's intensified one do not collapse into one English word.
• facta / verba / mores (0640A) — deeds / words / manners, rigid through VERS. 16, where the antithesis is repeated five times.

§5 · THE CRUCES PROPER

0640A · fides tantum meorum proprie fides est daemoniorum — RENDERED LITERALLY, NO MARKER

Migne prints, closing the chapter:

Nam si confitemini verbis et non factis, fides tantum meorum proprie fides est daemoniorum.

meorum ("of my own", gen. pl. of meus) has no antecedent anywhere in the work and no grammatical work to do beside fides tantum. Rendered as printed: "the faith only of my own is properly the faith of devils."

Two conjectures, both plausible, neither adopted. (a) verborum: "the faith of words only is properly the faith of devils" — this is what the sentence's own antithesis (verbis … non factis) demands, and it is the reading the Augustinian source behind the paragraph supports. (b) talium: "the faith of such men", picking up Tales sunt omnes mali catholici three sentences earlier.

[cj: …] declined, and the ground is Pattern 18a. The bar is that the printed word make the sentence assert something the author is arguing against. It does not: it makes the clause fail to construe, which 18a explicitly leaves declined ("mood, voice, number and person anomalies that merely fail to construe stay declined"). meorum is neither a negation nor a correlative, so the 2026-08-15 extension does not reach it either. And with two live conjectures competing, a [cj:] would have to choose one and print it on the reading page as ours. It stays here.

⚑ Note what is NOT wrong with the sentence: every negation is carried (non factis), and the surrounding argument survives intact, because the corruption is in the subject of the last clause rather than in its force. This is the shape 7a⁗ warns about in reverse — nothing is quarantined, so nothing is swallowed.

0637D · Conditione non tamen peccati misera servitute. sed Dei. — MIGNE'S FULL STOP KEPT

The plate sets a full stop between servitute and sed, mid-construction. Pattern 8: never delete a mark he prints. The English keeps the stop and the lower-case but: "By condition, yet not by the wretched servitude of sin. but of God." The same shape recurs at 0638D (Haec res. hic fructus apostolici officii.) and is kept the same way: "This is the matter. this the fruit of the apostolic office."

Capitalization was the live question, since English capitalizes after a stop. It is left lower-case because capitalizing would be our repair of his punctuation, and the whole point of carrying the stop is that a reader sees the plate's own sentence division.

0639C · Quis enim audet apertissime loquendo contradicere veritati. — A QUESTION MIGNE PUNCTUATES AS A STATEMENT

Quis … audet is interrogative in form and the sentence is plainly rhetorical, but the plate prints a full stop. Pattern 8: never supply a ? he does not print. Rendered "For who dares, by speaking most openly, to contradict the truth." with the stop. This is the work's only would-be question, and it is why the English carries 0 question marks against the Latin's 0 — the parity is real, not an accident of a dropped clause.

0637D · et ideo per ea magis laborandum — ea NEUTER, NO CLEAN ANTECEDENT

Rendered "and therefore one must labour the more through these things." ea is neuter plural and the only candidates in the sentence are feminine (fidem, quam, qua) or masculine (electi). Conjecturally eam (= fidem), "one must labour the more for it". Rendered as printed; no marker — the sentence asserts nothing false, it merely leaves the referent open, and English "these things" is as open as ea.

0642A · praemittat eos ita solite — solite FOR VULGATE sollicite?

Tit. III, 13 reads Zenam legisperitum et Apollo sollicite praemitte, and the gloss is paraphrasing that very command. Migne prints solite, "in the accustomed manner" — a real adverb, which occupies its slot and yields sense. Rendered as printed ("he send them on ahead in the accustomed manner"). No [cj:]: under 18a the printed word must make the sentence say something the author is arguing against, and "in the accustomed manner" is merely a weaker claim than "carefully", not a contrary one. Recorded because a later reader with the plate in hand may want to check the type.

0641A–B · Haereticus est qui … qui vitandi sunt — SINGULAR TO PLURAL, KEPT

The gloss defines the heretic in the singular (Haereticus est qui … impugnat … astruit … confirmet) and then closes on a plural relative (qui vitandi sunt … correcti exercitatiores essent). English has a full exponent for this shift, so it is rendered visibly: "A heretic is he who … ; who are to be avoided: because, being the more frequently corrected, they would be the more practised for evil." No marker. Pattern 9 normalizes only where English cannot show the defect; here it can, so showing it is the honest treatment (Pattern 12's test (2), run in the other direction).

0641B → 0642A · THE COLUMN JUMP — CHECKED, NOT REPORTED

The anchors run … 0641A, 0641B, 0642A, 0642B, skipping 0641C–D. Per the runbook's known false positive 1, band arithmetic manufactures phantom gaps and only text broken mid-word or mid-construction across an anchor is a signal. The break here falls cleanly between qui vitandi sunt: and quia frequentius correcti — a colon boundary, both halves complete. Checked and deliberately not reported; no [ed:], no patch, no plate call.

§6 · SUPPLIED WORDS IN FRAGMENTARY LEMMATA (Ruth convention 3)

Three lemmata are fragments that are not English on their own. Each is rendered as the phrase it stands for, per the pilot's settled convention, and each supply is named here because the convention licenses supply and therefore has to be auditable:

1. 0640B Ut viros suos. → "That they may love their own husbands." Supplies ament from Tit. II, 4. The gloss's very next words (ament etiam pudica dilectione) confirm the verb the fragment opens.
2. 0640B Ut non blasphemetur. → "That the word of God be not blasphemed." Supplies verbum Dei from Tit. II, 5. ⚠ Where Migne repeats the phrase in his own running prose at the end of the same paragraph (ut non blasphemetur, etc. — not italic, so not a lemma), the English supplies nothing and reads "that it be not blasphemed, etc." The convention covers lemmata; it does not license supply in the gloss's own sentences.
3. 0640C Ut doctrinam Salvatoris. → "That they may adorn the doctrine of the Saviour." Supplies ornent from Tit. II, 10 — which Migne then picks up as a lemma of its own two sentences later (Ornent. → "Let them adorn."). The small overlap is his, not ours.

A fourth was considered and declined: 0640A Factis autem. → "But by deeds." The Vulgate continues negant, but "But by deeds" is already English and already the phrase the gloss expounds; supplying "they deny" would add a verb the gloss itself supplies four words later (Si negatio … fit).

§7 · READ CLEAN

Named so a blind reader can tell checked ground from unexamined ground:

• The ARGUMENTUM (0637D) — one sentence, one supplied subject (named in §3), no lemma, no divergence, no defect.
• CHAPTER I, VERS. 1–5 (0637D–0639A) — read clean apart from the electi divergence (§1), the mid-construction full stops (§5) and per ea (§5).
• CHAPTER I, VERS. 6–9 (0639A–0639C) — read clean. The crimen/peccatum paragraph is the densest negation in the work (non quia … non dimittatur … non quiescunt … non est tamen sine peccato) and every one of the six negatives is carried; the count was made twice, against the Latin, before and after drafting.
• CHAPTER I, VERS. 12–16 (0639C–0640A) — read clean apart from meorum (§5). Note that the Epimenides line (Cretenses semper mendaces) is not printed: Migne abridges to in cujus libris hoc invenitur, etc., usque ad …, so there was nothing to render and the warning in the brief had no site to fire on. Recorded as a checked absence.
• CHAPTER II entire (0640A–0640D) — read clean. The subject shifts are the whole risk and every one is named in §3.
• CHAPTER III entire (0641A–0642B) — read clean apart from solite (§5) and the singular/plural heretic (§5).

Candidates rejected, with reasons (beyond the five lemma divergences in §1 and the conjectures in §5):

• 0639B nec saltem de luxuria accusatos — nec saltem where the Vulgate has non in accusatione luxuriae. Considered as a divergence; rejected, because the gloss is paraphrasing here, not quoting: the words are outside the italic lemma, and Pattern 14 applies to quoted scripture.
• 0642A non quia peritior sit Apollo — ambiguous, and both readings construe. (a) Apollo as ablative of comparison: "not because he [Zenas] is more skilled than Apollo"; (b) Apollo as the indeclinable nominative: "not because Apollo is more skilled." (a) was taken, because the sentence exists to explain why Zenas alone carries a title in the verse, and (b) would make the clause say nothing about Zenas at all. Flagged rather than settled: the printed words permit both.
• 0640B In habitu sancto rendered "In holy habit" rather than "in holy attire". The gloss expounds it as bearing (incessus, motus, vultus, sermo), not dress, and English habit is the one word that carries both senses as habitus does. Choosing "attire" would have narrowed the lemma against its own gloss; choosing "behaviour" would have thrown away the sense the familiar English carries. Recorded because it is a judgement, not a gloss.
• 0639C arguere/redarguere — rendered reprove/confute. Considered collapsing both to "reprove"; rejected, because the gloss escalates deliberately (quos redarguere bonum opus est, grandis sarcina, clivus arduus).
• Em-dashes — none used. Guillemets — none in either language; the two quotations the gloss carries (Gen. I, 31 and Joh. XVII, 3) are marked by italics on the plate, so Pattern 6 leaves them italic and adds no marks of ours.

§8 · MIGNE'S FOOT-OF-PAGE CONJECTURE NOTES — NOT RECOVERED FOR THIS WORK

No plate read was performed and no [cn: …] notes exist for 8997. data/plate-notes/ has no entry for this idno. This is a skip, not a checked zero, and is recorded as such so a later reader does not mistake it for Ruth's tier-3 full coverage. Under the 2026-08-18 ruling the recovery belongs per volume at chunking; PL 114 has not had that pass.

§9 · BLIND POLARITY READ — 2026-08-18, and the adjudication

A fenced reader (Latin twin + English only, never this file), all four sections read end to end. 6 sites raised · 3 confirmed and repaired · 3 declined.

Zero polarity defects across the whole work, and the census is named. Every negation Migne prints survives in both directions — Conditione non tamen peccati · non tepidius agendum · Non recens promissio · catholicam non privatam · non autem singulariter · nullus… posset ordinari · neminem dixerim · non quia peccati aliquid · non dimittatur · quotidie fieri non quiescunt · non est tamen sine peccato · Contradicentes non uno modo · Non contradicunt multi lingua · nullum exclusit · non solum erudivit · Ut non aspere accipiatur · non quia peritior sit Apollo · ut nihil… desit · non illis deessent carnalia — plus the whole Titus 1:15–16 stretch (non secundum naturas · nec nobis omnia apta sunt · si negatio non tantum lingua · qui non verbis, sed factis negant · Nolite ergo esse · et non factis). Zero conformations: Sine crimine held as "Without crime" and never became AV's "blameless" across four re-uses; Eos qui contradicunt, arguere did not become Douay's "to convince the gainsayers"; no dilecto supplied at Filio secundum communem fidem; no "of God our Saviour" supplied at Ut doctrinam Salvatoris; no autem restored to Haec est vita aeterna.

⭐ The presbyteri / episcopus handling was independently confirmed as clean — the two words are held apart everywhere ("presbytery", "presbyters", "under the name of presbyters" / "bishop", "bishops"), never "priest", never "elder", never collapsed, so the identification at 0639C reads as Migne's own act: "Here he plainly shows that under the name of presbyters bishops were designated above." Nothing was decided by our vocabulary.

✅ 0642A · VERS. 13 — CONFIRMED, REPAIRED. The dropped correlative was concealing a broken plate.

Printed: …praemittat eos ita solite, ut nihil in viatico illis desit. Was: "…he send them on ahead in the accustomed manner, so that nothing be lacking…" Now: "…he send them on ahead so customarily that nothing be lacking…"

ita had gone unrendered, and its bare "so that" had nothing to answer to. ⭐⭐ This is the finding worth carrying forward, because the dropped particle was doing damage beyond itself. Rendered with the correlative intact, the sentence visibly does not construe — and that is the point: solite is almost certainly the plate's corruption of sollicite, which is the Vulgate's own word at the very verse being glossed (Titus 3:13, Zenam legis peritum et Apollo sollicite praemitte) and which makes ita … ut work perfectly ("send them on so carefully that nothing be lacking"). Our English rendered the corrupt word faithfully, per Pattern 7 — and then quietly dropped the particle that would have exposed the break, so the page read as a bland, complete sentence.

⚑ A defect can hide a second defect by making the page read smoothly. Fidelity to the words is not fidelity if the connective that reveals the trouble is left out. No marker proposed: Pattern 18's census is closed (18a), and this is recorded as a reading problem.

✅ 0640B · VERS. 2 — CONFIRMED, REPAIRED. An invented subject with no antecedent.

Printed: Sunt enim plurima quae in hujusmodi aetate vix fieri patiantur. Was: "…which at an age of this kind they scarcely suffer to be done." Now: "…which at an age of this kind scarcely suffer themselves to be done."

Both readings construe and the choice is not free of doubt — §2 had taken senes as the subject on the ground that deponent patior wants an animate sufferer. But there is no antecedent for "they" in reach: the subject of the two preceding sentences is senectus, singular, and the only plurals nearby are the vices themselves; a reader must jump back past a singular subject to Senes in the lemma three sentences up. Taking quae as the subject of the deponent in a relative clause of characteristic supplies nothing, and it is what the whole Chrysostom excerpt is about — what old age can no longer manage.

⚑ The rule applied: where two readings construe, prefer the one that supplies no unprinted word. The near-unparseable English was itself the tell.

✅ 0639B · VERS. 6 — CONFIRMED, REPAIRED. An idiom rendered word by word.

…nec frontem habet ad alios talium cohabitator. "A face for others" → "the face to confront others." frontem habere ad is the idiom of effrontery — the standing to face someone down. Pattern 12's boundary governs: normalize the language, mark the page. The harder half of the sentence was already right, the nominative cohabitator correctly identified.

⛔ 0640A — fides tantum meorum: DECLINED as an edit, and the reader agrees it is Migne's

Nam si confitemini verbis et non factis, fides tantum meorum proprie fides est daemoniorum. The reader independently reached §2.1's conclusion — the argument has run its whole length on verbis / factis, and the clinching sentence wants verborum: "a faith of words only is properly the faith of devils." It also confirmed that this is the plate's loss, not ours: meorum is a real form, so [sic:] has nothing to wrap, and 18a forecloses the other marker. Rendered as it stands. The English is unreadable at the argument's climax and that is Migne's page, faithfully shown.

⛔ 0639B — negation scope, DECLINED: the English mirrors the Latin's own looseness

Id est, nec saltem de luxuria accusatos, vel sibi non subditos. The reader warned that "not so much as accused of wantonness, or not subject to him" lets a reader take the second member as a property of faithful children. But the Latin is loose in exactly the same way: nec negates accusatos, while the second member carries its own non, so a reader of Migne must distribute the sense precisely as a reader of our English must. Tidying it would be repairing his syntax, not translating it. Reader's own confidence: medium, with the admission that the distribution is available. Named here rather than smoothed.

⛔ 0640C — the bare "it" in the re-quoted ut non blasphemetur: DECLINED, and it is an artifact of the convention

The lemma Ut non blasphemetur. is expanded to "That the word of God be not blasphemed" (convention 1.2, a fragment rendered as the phrase it stands for); the identical Latin re-quoted four lines down is rendered bare, "that it be not blasphemed." The inconsistency is real but it is produced by the convention working correctly — the lemma is expanded because it is a lemma, and the re-quotation is not. Carrying "the word of God" into the second would supply words Migne does not print there.

⚑ Worth knowing for the remaining Glossa books: fragment expansion in a lemma will routinely look inconsistent with the same words re-quoted bare in the gloss. That is by design, not drift.

Read clean, and named

All four sections — ARGUMENTUM · CAPUT PRIMUM · CAPUT II · CAPUT III — read end to end. 14 column anchors present and in order with no text broken mid-word or mid-construction at any of them (each checked); 3 [n:] verbatim and untranslated, with the two that sit inside lemma spans still inside them; ~45 italic spans counted section by section; ? 0 : 0, with Quis enim audet apertissime loquendo contradicere veritati. — a rhetorical question Migne closes with a full stop — correctly riding the stop; five Pattern 17 sites all correct.

Rejected with reasons: both band jumps (false positive 1, doubly so here since the gloss on Philemon interleaves these same columns); the [0639D] anchor placement (word order forces it, nearest achievable); decent vs Vulg. decet ([var:] arguable, but the sense does not turn on it — Pattern 14's test); irreprehensibilem episcopum → "blameless bishop" (glosses irreprehensibilis, which means blameless — and the lemma sine crimine was pointedly not conformed to it four words earlier); Migne's lowercase after a full stop, mirrored deliberately; quidam nostrum… carebant (the person shift is his); priusquam veniat Nicopolim (unexpressed in the Latin too, equally ambiguous — unlike the two repaired sites, nothing was rearranged); non quia peritior sit Apollo (ablative of comparison confirmed, and "Apollo" kept rather than conformed to the English Bible's "Apollos"); scientes quod singulis persolvi oporteat → bare "it" (as vague on both sides).

⚑ One class recorded rather than reported, for the house to rule on if it wishes: five Latin commas rendered as English semicolons or a colon (0640A, 0640B, 0640C, 0640D, 0641B). No mark is deleted or supplied and the sentence count is unchanged, and Pattern 8's worked cases are all ? and full stops. The 3 John read (8989 §4) found that comma-level substitutions can net to zero and pass every count-based check, so the instances are named here in case Pattern 8 is later read as covering them.

⚑ Correction to §8 — the plate WAS checked

§8 logged Migne's foot-of-page apparatus as a skip for PL 114. It is now a checked zero: the full-width foot bands of leaves n323–n325 of archive.org patrologiaecurs06saingoog (leaf n = (column + 9) / 2, calibrated against the running head at n330 = cols 651–652) cover this work's columns, including n325, where the Titus / Philemon division falls mid-page — the case that lost three notes on PL 202. Migne prints no foot-of-page apparatus there: no foot rule, no numbered notes, the columns simply end. So solite at 0642A and meorum at 0640A are not cruces he already answered. Recorded in data/plate-notes/coverage.json as tier 3, coverage full, notes_recovered: 0.