These are the edition's working notes, published as they were written. They record, work by work, where Migne's plate is defective or the reading uncertain: the printed form, what the English does with it, and why. They are not a groomed apparatus. Where a note argues with itself, withdraws an earlier decision, or leaves a question open, that is the record of how the reading was actually reached, and it is left standing — a conjecture that is visible can be checked, and one that has been tidied away cannot.
The English never silently repairs the page. A construable but wrong reading is rendered literally and the conjecture is recorded here, never applied to the text. How the English is produced and checked →
Translated 2026-08-18, one translator, 2 chunks (1,445 Latin words → 2,233 English), on Opus. Structure: ARGUMENTUM, CAPUT PRIMUM, CAPUT II, CAPUT III.
Apparatus: 3 [var:] · 2 [sic:] · 0 [cj:] · 0 [d:] · 0 [ed:] · 0 [nt:] · 2 [n:] (exact parity with the Latin twin). 14 column anchors (10 + 4), 58 italic spans in the Latin (45 + 13) against 63 in the English (47 + 16), the difference being exactly the 3 [var:] and 2 [sic:] contents, which the verifier strips. Both [sic:] markers are at 0624B and both were added after a blind polarity read; see §9, which is the section to read first. 1 question mark and 1 exclamation mark, both in chunk 0, both at parity. Zero guillemets in either language. node scripts/verify-english.mjs 8987 returns clean.
Conventions follow src/english/8968/cruces.md §1 (Ruth, the Glossa pilot) and shipped 8978 (Jude, same volume, same form): VERS. n.-- passes through verbatim; the lemma is Englished; the attribution sigla ride as ordinary [n:] notes with contents untranslated; italic usque ad → as far as (8978 cruces §205).
§0 · PLATE STATUS — a CHECKED ZERO
Foot-of-page apparatus for cols. 0619–0624: notes_recovered: 0, coverage full, tier 3. Read this session by the orchestrator, not by the translator.
Method. archive.org patrologiaecurs06saingoog, leaf n = (column + 9) / 2, calibrated against the running head (n312 = cols. 615/616). Leaves n314, n315, n316 read at full page width across the whole foot. No foot rule, no numbered notes: the columns simply end.
⭐ Two of the three leaves carry a mid-page work division and were read there as well as at the foot: n314 (1 Thessalonians ends; EPISTOLA II AD THESSALONICENSES opens below the rule, which is where this work begins) and n316 (this work ends; 1 Timothy begins). That is the configuration that lost three notes on PL 202. It is clean here.
⛔ This is a checked zero, not a skip and not "unchecked." Consequently any [cj:] in this work would be ours, not a conjecture Migne had already answered at the foot — and this work takes none.
Two corroborations from the same plate read (used, but not treated as licence to skip the translator's own reading):
- The ARGUMENTUM is confirmed word for word against n314, and
CAPUT PRIMUM.opens atVERS. 3. — Gratias agere, etc., exactly as our Latin twin has it. - ⭐ The question mark at 0621A is Migne's own: the plate prints
quid faciet induratis ?(n315, left column). It is therefore preserved, not supplied, and the Pattern 8 parity for chunk 0 stands at one question mark against one, as recorded in the header.
§1 · ⛔ MIGNE'S OWN VERSE NUMBERING — TWO INDEPENDENT DEFECTS. NOTHING RENUMBERED.
Verse addresses pass through verbatim by the settled convention. Both of the following are recorded, not repaired; the reader's Latin column would otherwise disagree with our English.
1a · VERS. 5. at 0622A is a misprint for VERS. 3, and it puts the sequence out of order
Chapter II runs on the plate: 1 · 2 · 5 · 4 · 6 · 7 · 8 · 9,10 · 12 · 13 · 15.
The lemma printed under VERS. 5. is Quoniam nisi venerit / Nisi venerit refuga primum / Filius perditionis. All three words stand in 2 Thess 2:3 (quoniam nisi venerit discessio primum, et revelatus fuerit homo peccati, filius perditionis). 2 Thess 2:5 contains none of them (non retinetis quod cum adhuc essem apud vos, haec dicebam vobis?). The following address is VERS. 4., and the gloss under it (In templo) is genuinely 2:4 — so read as 3, Migne's run is 1 · 2 · 3 · 4 · 6 …, continuous. The defect is one digit.
⚠ Note also that VERS. 5. and the true VERS. 5. are not both present: 2:5 is simply not glossed. So no gloss has been displaced and no text is lost; no [ed:].
1b · From VERS. 12. at 0623A the addresses run one behind the text they gloss
| Migne's address | lemma printed | where the words actually stand |
|---|---|---|
| VERS. 12. | Primitias in salutem. | 2 Thess 2:13 (elegerit vos Deus primitias in salutem) |
| VERS. 13. | In acquisitionem gloriae Domini nostri Jesu Christi. | 2 Thess 2:14 |
| VERS. 15. | Et dedit consolationem. / In gratia. | 2 Thess 2:16 (dedit consolationem aeternam … in gratia) |
A consistent −1 offset over the last three addresses of Chapter II, beginning immediately after the AUG. note. Chapter III's addresses (1 · 2 · 6 · 7 · 9 · 10 · 11 · 13) are all correct against the Vulgate, so the offset does not propagate.
⛔ Not corrected. The two defects have different shapes (1a is one wrong digit inside an otherwise sound run; 1b is a sustained offset), which is itself evidence that neither is our business to normalise.
§2 · THE LEMMA DIVERGENCE LIST — ALL 57 SPANS CHECKED AGAINST THE VULGATE
Every italic lemma span in both chunks, checked word by word against the Clementine Vulgate before the English was written, and each rendering then held fixed against every re-quotation of it in the gloss below. Agreements are listed too, so this can be told apart from a report by someone who only listed findings. (The 58th italic span, usque ad at 0623A, is Migne's abridgment formula, not a lemma; see §5.)
2a · The three that diverge — all marked [var: …]
| col | Migne prints | Vulgate | marker + reasoning |
|---|---|---|---|
| 0621A | VERS. 7 In revelationem. | in revelatione Domini Jesu (1:7) | [var: Vulg. in revelatione, ablative, "at the revelation"]. Migne's accusative turns a point of time into a terminus, so the English reads "Unto the revelation." The gloss below settles nothing either way (Incipit ostendere quam horribilis tribulatio…), which is exactly why the printed case had to decide it and not the sense. ⭐ The rendering an eye that knows 1 Thess/2 Thess reaches for first is Douay's "at the revelation" — that would have been a silent repair of the case ending. |
| 0622A | VERS. 5 Nisi venerit refuga primum. | nisi venerit discessio primum (2:3) | [var: Vulg. discessio, "the departure"]. Material and load-bearing: discessio is an event (a falling away), refuga a person (a deserter), and the whole gloss turns on the person: quem refugam vocat: utique a Domino Deo — "whom he calls a deserter: from the Lord God, assuredly." Rendered "Unless the deserter come first," with refugam in the gloss carried as the same word, "a deserter." ⚠ Migne himself flags it (Sic aliqui codices habent), and that flag is not a substitute for the marker: his sentence tells the reader some codices read so, ours tells him what the received text reads. Note also that the immediately preceding gloss uses the Vulgate word (vel discessione a fide), so both readings stand on the same plate within two lines. |
| 0623C | VERS. 7 Scitis quomodo oporteat nos imitari. | scitis quemadmodum oporteat imitari nos (3:7) | [var: Vulg. quemadmodum oporteat imitari nos, "how you ought to imitate us"]. Sense-neutral (quomodo and quemadmodum are synonyms; the word order does not change which of nos/imitari governs). Marked anyway because it is a real lexical substitution in a quoted verse and the launch instruction is to name divergences, not only sense-changing ones. The [var:] gloss deliberately repeats the same English to say so: the divergence is in the Latin, not in what it means. ⛔ Do not read the marker as a claim that the sense shifts. |
2b · The one that is orthographic only — checked, deliberately NOT marked
| col | Migne prints | Vulgate | why no marker |
|---|---|---|---|
| 0621A | 1:8 His qui non noverunt Deum. | iis qui non noverunt Deum | his for iis is the ordinary medieval/Migne spelling of the same dative-ablative plural, not a different word. Nothing in the gloss turns on it. A [var:] here would train a reader to discount the three that matter. Recorded so the absence of a marker is a decision, not an oversight. |
2c · The alternative-version reading, quoted as such — checked, deliberately NOT marked
| col | Migne prints | why no marker |
|---|---|---|
| 0621A | Siquidem justus est, etc. | Migne labels it: unde alius interpres ait. It is not offered as the text of 2 Thess 1:6 but as another translator's rendering of it, so it is not a divergence in a scripture citation but a cited variant version. Pattern 14 does not reach it. ⚠ But see §3 — the masculine justus against justum is the whole point of quoting it, and the English has to carry that. |
| 0622B | in templum Dei sedeat, | Migne prints both readings and argues between them in his own prose (non in templo, sed in templum Dei sedeat, sicut in Graeco est). The Vulgate reading he rejects is already on the page in the same sentence. A [var:] would only re-state what the gloss says out loud. |
2d · The fifty-two that agree with the Vulgate exactly (checked, no marker)
Chunk 0, Chapter I (1:3–1:11): Gratias agere, (1:3) · In invicem. (1:3) · In exemplum justi, (1:5) · Justi: si tamen justum est, (1:5 + 1:6) · si tamen, (re-quote) · Domini Jesu. (1:7) · in flamma ignis (1:8) · Qui poenas dabunt. (1:9) · In interitu aeternas. (1:9) · Cum venerit glorificari, (1:10) · Quia testimonium. (1:10) · In quo. (1:11) · in quo, (re-quote) · Opus fidei. (1:11) · In virtute (1:11).
Chunk 0, Chapter II (2:1–2:16): Rogamus autem vos, fratres, per adventum Domini nostri Jesu Christi, (2:1) · Non cito moveamini. (2:2) · Neque per spiritum. (2:2) · Neque per Epistolam. (2:2) · Quasi instet dies Domini. (2:2) · Quoniam nisi venerit. (2:3) · Filius perditionis. (2:3) · In templo. (2:4) · templum. (2:4, accusative variant named by the gloss) · in templo, (re-quote) · Scitis. (2:6) · Suo tempore. (2:6) · Nam mysterium, (2:7) · Donec de medio fiat. (2:7) · De medio fiat. (re-quote) · Spiritu oris sui. (2:8) · Mendacibus et in omni seductione. (2:9–10) · Ideo mittet illis Deus. (2:11) · Primitias in salutem. (2:13) · In acquisitionem gloriae Domini nostri Jesu Christi. (2:14) · Et dedit consolationem. (2:16) · In gratia. (2:16).
Chunk 0, the two spans that are the gloss's own re-ordering, not quotations: In quo die oramus, ut dignetur, and Semper oramus. — Migne prefaces them Ordo verborum talis est, i.e. he is re-arranging 1:11 (In quo etiam oramus semper pro vobis: ut dignetur…) to show the construction. Every word is 1:11's; only the order is his, and he says so. Rendered as the re-arrangement he prints.
Chunk 1, Chapter III (3:1–3:13): De caetero, fratres, orate. (3:1) · Ut liberemur. (3:2) · Et malis, (3:2) · liberemur, (re-quote) · Non enim omnium est fides. (3:2) · Denuntiamus. (3:6) · Sed ut nosmetipsos. (3:9) · Quoniam si quis non vult operari. (3:10) · Curiose agentes. (3:11) · Nolite deficere. (3:13).
Chunk 1, the Matthew citation, quoted twice: Nolite solliciti esse, quid manducetis [0624A] and Nolite solliciti esse quid manducetis, [0624B]. Matt 6:25 reads Nolite solliciti esse animae vestrae quid manducetis. Migne's two quotations both drop animae vestrae and both close with etc. — i.e. this is abridgment of a citation, not divergence: the omitted words sit between the quoted ones and the etc. announces the truncation. No [var:], per the fragment convention (brief §3). ⚠ The two quotations differ from each other by one comma (esse, quid vs esse quid); both are reproduced as printed.
§3 · 0621A · si tamen — the argument the English had to leave standing
Migne: Justi: si tamen justum est, quod constat esse. Non enim ponitur hic, si tamen, pro dubitatione, sed pro affirmatione, unde alius interpres ait: Siquidem justus est, etc.
The gloss's whole claim is that si tamen is not conditional-dubitative here. So the English of the lemma may not be a rendering that only reads as doubt.
- Rejected: "if however it is just" and "if nevertheless it is just." Both force the concessive, which is precisely the reading the gloss denies; the paragraph would then contradict its own lemma.
- Rejected: "since it is just." This resolves the argument inside the lemma and leaves the gloss with nothing to argue. It would also be a silent repair of a conditional into a causal.
- Adopted: "if indeed it is just." English if indeed genuinely carries both values, so the gloss can say which one is meant and the reader can see that it needed saying. The re-quotation two clauses later is held to the identical words, if indeed.
⭐ The masculine in Siquidem justus est. justum est (neuter, "it is just") against justus est (masculine, "he is just") — the alternative version supplies a personal subject, and that is a second, unremarked way of turning the concession into an assertion. The unexpressed subject is God, from apud Deum in 1:6 (si tamen justum est apud Deum retribuere). Rendered "Since indeed he is just," and the "he" is named here rather than supplied silently. ⚠ A blind reader may want to weigh "Since indeed it is just" — but the printed justus is unambiguously masculine nominative and justum stands three words earlier in the same sentence for contrast. The call is not close; it is only easy to miss.
§4 · THE UNEXPRESSED SUBJECT — every supplied pronoun, named
Sites where English forced a pronoun the Latin does not print. Antecedent named, with what fixes it.
| col | Latin | supplied | what fixes it |
|---|---|---|---|
| 0620D | Hic gratias agit Deo | "he" = the Apostle | The ARGUMENTUM's subject (scribit Epistolam Apostolus), and gratias agit answers the lemma Gratias agere of 1:3, whose subject is Paul. |
| 0620D | ne … extollantur … habentes | "they" = the Thessalonians | de bonis eorum in the same clause. |
| 0621A | Si enim tam severe punit … quid faciet induratis? | "he" = God | The lemma is in exemplum justi (judicii Dei); and the next clause names him: non parcit impiis. ⚠ Not the Apostle, who is the subject of the following sentence (Hoc dicit). The two sentences change subject without saying so. |
| 0621A | Hoc dicit, ut intelligatur | "he" = the Apostle | dicit of the writer, contrasted with punit of God above. |
| 0621A | quod constat esse | "it" = that it is just | Refers to justum est immediately preceding. |
| 0621A | Incipit ostendere | "he" = the Apostle | Same speaker as Hoc dicit. |
| 0621A | quam horribilis tribulatio illis, et quam jucunda requies istis | "those" / "these" | 1:6–7 supplies both: illis = the persecutors (iis qui vos tribulant), istis = the Thessalonians (et vobis qui tribulamini, requiem). Rendered "for those" / "for these" so the deixis survives; ⛔ collapsing them to "them" and "you" would have destroyed the contrast. |
| 0621A | Venientis … quia inimicos exuret | "he"/"his" = the Lord Jesus | The lemma directly above is Domini Jesu, and Venientis is its genitive participle. |
| 0621A | Nam alii sunt qui nesciunt legem | none supplied | Impersonal; rendered "For others are they who know not the law." |
| 0621B | Id est sustinebunt | "they" = the ungodly of 1:9 | From the lemma Qui poenas dabunt. |
| 0621B | Semper sentient poenas, non tamen semper deficient | "they" = same | See §6 on the negation. |
| 0621B | Ipse enim clarus … videbitur … apparebit … coarctabit | "he" = the Lord | Ipse is printed, and the lemma is Cum venerit glorificari (1:10, of the Lord). |
| 0621B | Ideo vobis tribulatis dat requiem nobiscum | "he" = God/the Lord | Subject carried from Veniet punire malos in the sentence before; vobis and nobiscum are printed. |
| 0621B | ut dignetur, id est dignos judicet | object "you" | ⚠ The Latin prints no object. Vulgate 1:11 reads ut dignetur vos vocatione sua Deus noster, so the object is the Thessalonians and the subject is God. Rendered "may judge you worthy." ⭐ This is the one place a supplied word came from the verse behind the lemma rather than from the sentence, and it is flagged for that reason. "may judge them worthy" also construes; it was rejected because the whole paragraph is second-person (det vobis perseverantiam four lines later). |
| 0621C | ut a confessione nullo modo flecti possint | "they" | Latin prints 3rd plural possint. ⛔ Not conformed to the surrounding "you": the Latin changes person here and the English follows it. |
| 0621C | ut scilicet det vobis perseverantiam | "he" = God | dignetur … Deus noster of 1:11. |
| 0621C | Hucusque dixit se eis gratias agere | "he" = the Apostle | se + eis, writer and addressees. |
| 0622A | Occulte loquitur … ne irritaret eos | "he" = the Apostle | Continues dixit … determinat from the preceding paragraph. |
| 0622A | eum dixisse, quem refugam vocat | "he" = the Apostle; "whom" = Antichrist | de Antichristo eum dixisse is printed. |
| 0622B | Eos scire dicit quid detineat | "he" says, "they" know | dicit = the Apostle; eos = the Thessalonians, from the lemma Scitis (2:6, et nunc quid detineat scitis). |
| 0622C | sicut per illum tunc aperte facturus est | "he" = the devil, "him" = Antichrist | Diabolus … occidebat is the printed subject; illum is the Antichrist of the same paragraph. |
| 0622C | qui fidem tenet, teneat / qui detinet illum … detineat | "he" | Both relatives are printed masculine; the English keeps "he who," ⚠ even in the second case where the apposition id est accessio ad fidem is feminine. Not tidied. |
| 0622C | jam … revelabitur | "he" = that wicked one | 2:8, tunc revelabitur ille iniquus; and aderit ille iniquus stands at 0622B. ⚠ revelabitur could formally take discessio (fem.) as subject, but discessio has just been given its own verb in the same sentence (revelata discessione), so a second subject is required. Flagged; the call is made. |
| 0622C | quia ab eo procedit | "he" = the Spirit, "him" = the mouth's owner | qui dicitur spiritus oris names the Spirit; ab eo is the one whose mouth it is. |
| 0623A | ad mendacium trahent | "they" = the signs | Signa illa vocantur mendacia opens the paragraph. Object "men" supplied for English; noted. |
| 0623B | dedit … speramus | "he" gave, "we" hope | quam dedit and speramus are both printed; only the subject of dedit (God) is supplied. |
| 0623C | ne nostros seducant … tendant | "they" = the false ones | A pseudo opens the clause. |
| 0623C | acceperunt omnes fidem per me | "me" printed | Paul speaking in his own person, as Ideo dico liberemur shows. |
| 0624A | cum illi haec non faciant, debetis vitare | "them" as object of vitare | illi is printed as subject; English requires the object. From 3:6, ut subtrahatis vos ab omni fratre ambulante inordinate. |
| 0624B | cum multis aliis locis epistolarum quid hic sentiat Apostolus apertissime doceat | the Apostle teaches, and "he" means | ⛔ REPAIRED after the blind read, see §9.3. The first rendering was "since he most plainly teaches what the Apostle here means," which reads as two agents where the paragraph has one. The cum-causal is the answer to the mist-drawers: Paul himself settled it elsewhere. Now "since the Apostle in many other places of the epistles most plainly teaches what he here means." |
| 0624B | Vult enim servos Dei corporaliter operari | "he" = the Apostle | Same subject as doceat. |
| 0624C | possunt tamen nonnullis indigere | "they" = those who work | etsi operentur immediately before; the workers of 3:10–12. nonnullis read as neuter ("some things"), not masculine ("some men") — indigere takes the thing lacked. |
| 0624C | Et ideo monet | "he" = the Apostle | Continues the paragraph. |
§5 · SMALLER CALLS, MADE AND RECORDED
- 0623A · usque ad. Deus diabolum ista facere permittet, etc., usque ad alias injustissime judicatum. Migne's abridgment formula, rendered as far as per 8978 §205, italics kept. The tail it points to, alias injustissime judicatum, is the closing fragment of the abridged Augustine passage, not a complete clause, and is rendered as the fragment it is: "otherwise most unjustly judged." ⛔ Not completed into a sentence.
- 0621B · Qui poenas dabunt must be rendered "give," not "suffer." The entire gloss is a lexical note on that verb (Dare enim aliquando ponitur pro sustinere). Douay's "shall suffer eternal punishment" would delete the thing being explained. The gloss's own dare and sustinere are then carried as "to give" and "to undergo," in double quotes as mention rather than use.
- 0622B · Sedet in amicum, id est ut amicus. Rendered "He sits into a friend, that is, as a friend." ⚠ Deliberately un-English. The sentence exists to illustrate that in + accusative can mean "as," which the reader can only see if the accusative survives the crossing. "He sits as a friend, that is, as a friend" would be a tautology on the page. Migne prints it without italics; none were added.
- 0621C · Non cito moveamini glossed Si moveamini; cito est. Rendered "If you be moved; it is soon, for nothing is brought upon you which lasts long." ⚠ The semicolon is Migne's, in that position, and it makes the sentence read oddly in both languages: the apodosis cito est has no expressed subject and no obvious complement. Read as "it is [only] for a short while," which nihil … quod diu duret confirms. Nothing supplied into the English; the elliptical shape is preserved and logged here instead.
- 0621C · In virtute perseverantiae. Migne's italic closes at virtute, so the English italic closes at "power" and "of perseverance" stands outside it, exactly as printed.
- 0623C · importunis / portu. liberemur ab importunis … nec stant in portu quietis. There is a play on importunus / portus which no English carries. Rendered "importunate men" and "haven of quiet," which keeps the Latinate first word so a reader with the parallel column can see the figure. ⛔ Not marked; a lost pun is not a defect (Pattern 9).
- Register. Addressees are plural throughout (Thessalonians, vobis, fratres), so Pattern 17 gives you, never thou, in every second-person clause of both chunks. No singular addressee occurs in this work.
- Both notes are citations, not prose.
(AUG., lib. XX, de Civit. Dei, c. 19.)and(Matth. VI)stay[n:]with contents untranslated. Neither is a candidate for Pattern 15; judged by content, both are pure locators.
- The AUG. note sits inside the italic span at 0623A (
Ideo mittet illis Deus. [n: …]). Reproduced in that position, inside the italics.
- Em-dashes: none introduced. Where Latin colons and semicolons invited one, a comma, colon, or full stop was used instead.
§6 · NEGATION — the passages read slowly, and what each one says
Every printed negative is carried. None is added. Listed because the launch brief names them.
- 0621A · non parcit impiis qui non parcit justis. "he spares not the ungodly who spares not the just." Both negations stand; the sentence is an a fortiori and construes as it is.
- 0621A · His qui non noverunt Deum. Nam alii sunt qui nesciunt legem. Two different negated verbs (non noverunt / nesciunt) distinguishing two groups; rendered "have not known" and "know not" so the distinction the gloss is drawing survives.
- 0621B · Semper sentient poenas, non tamen semper deficient. ⭐ Rendered "They will always feel the penalties, yet they will not always fail." deficere = to give out, to be used up. The point is that the damned are not consumed by what they suffer, so the suffering can be endless. ⛔ Rejected: rendering deficient as "cease" or "perish" to make the antithesis sharper in English — both are readings, not translations, and the gloss's own antithesis is sentient against deficient, feeling against being used up. The apparent paradox is the plate's and is left standing.
- 0621C · ut a confessione nullo modo flecti possint. "in no wise." Kept.
- 0621C · Non cito moveamini … Neque per spiritum … Neque per Epistolam. Three negations in three consecutive lemmata, all carried: "Be not soon moved," "Nor by spirit," "Nor by epistle."
- 0621C · nolite credere. "do not believe him." Object supplied (si quis dicat); named in §4's spirit, recorded here.
- 0621C · ⭐ Non hoc ita dicit quin velit, ut habeant succinctos lumbos… The hardest negation in the work: non … quin is a double negative meaning "he does not say it in such a way that he would not wish." Rendered "He does not say this in such a way as not to wish that they should have their loins girt…" ⛔ Rejected: "He does not say this so as to forbid…" and "He says this without meaning that they should not…" — both collapse the two negatives into one and would be a repair. Both printed negatives appear.
- 0622A · nulli dubium est. "it is doubtful to none," not "it is certain." The negation is Migne's construction and stays.
- 0622A · non per naturam, sed per imitationem. Kept.
- 0622B · nec aperte exponit, et ideo nos nescimus, nisi quod… Three negatives (nec, nescimus, nisi) in one sentence, all carried: "and he does not openly expound it, and therefore we know not, save that…"
- 0622B · Apostolum non id aperte scribere voluisse. "the Apostle was unwilling to write this openly." The negative attaches to voluisse, not to scribere; a rendering "wished not to write" would say the same, but "was unwilling to write" keeps voluisse as the negated verb.
- 0622C · ut non erubescat homo adulterari vel furari, sicut nec ambulare vel loqui. "so that a man blush not to commit adultery or to steal, as neither to walk or to speak." ⚠ The nec clause is elliptical (it depends on erubescat), and the English keeps the ellipsis rather than expanding it to "as he does not blush to walk or to speak."
- 0623A · vel quia non vera ut videntur, sed phantastica erunt. "either because they will not be true as they appear, but phantastic." ⚠ Migne opens with vel and the second arm follows (Vel si vera), so "either … or"; the negation sits inside the first arm.
- 0623B · haec omnia dedit non merito nostro, sed in gratia sua. Kept.
- 0623C · ⭐ quia nondum acceperunt omnes fidem per me qui accepturi sunt. "because not yet have all received the faith through me who are to receive it." ⚠ Word order in English follows the Latin so nondum keeps initial force in its clause.
- 0623C · nec stant in portu quietis. "and stand not in the haven of quiet."
- 0623C · nec vobis orantibus omnes sunt credituri. "not even with you praying will all believe." nec read as "not even," which the sed tantum praeordinati that follows requires.
- 0624A · cum illi haec non faciant / non ambulant secundum traditionem / quod in illis non videtis / Non est opus ut exponam / quod non inquieti fuerimus. Five negations in one paragraph, all carried.
- 0624A · Apostolus non solum verbis docebat, sed hortabatur et factis. Kept.
- 0624B · ⭐⭐⭐ ut quod utiliter charitas monet, non solum facere velint, sed nec intelligere. ⛔ REPAIRED after the blind read, see §9.1. Now "so that they not only wish [sic: velint] to do what charity profitably admonishes, but not even to understand it." velint is wish, as everywhere else in this work. The first rendering ("are not only unwilling to do") translated it as its opposite by pressing the non of non solum … sed into double duty, which that correlative does not permit.
- 0624B · ⭐ ut non compellantur egestate necessaria petere. "that they be not compelled by want to beg for necessaries." ⚠ necessaria is the object of petere, not an adjective on egestate; "compelled by necessary want" is the misreading to watch for.
- 0624B · ⭐ the densest sentence in the work: Non enim hoc ideo dictum est, ut non procurent quantum necessitati sat est, unde honeste potuerunt, sed non ut ista intueantur, nec propter ista faciant quidquid… Four negatives. ⛔ PARTLY REPAIRED after the blind read, see §9.2. All four were present, but the third had been moved: sed non ut was rendered "but that they should not look," i.e. as ut non, which is a different scope. Now "but not in order that [sic: sed non ut] they may look upon these things." ⚠ A negation can be fully present and still be wrong, because scope is not tallied. That is why the counting check did not catch it.
- 0624C · Non enim in reprehensionem venit qui humanus est in largiendo, sed hic qui… "For he does not come under reproof who is kindly in bestowing, but he who…" Kept.
- 0624C · Nolite deficere. Rendered "Faint not." ⛔ Rejected: Douay's "do not be weary in well doing," which imports benefacientes from 3:13 that Migne does not print in the lemma.
§7 · READ CLEAN — sections examined and found without defect
- ARGUMENTUM (0619D). Three clauses, no divergence, no crux. acolythum rendered "acolyte," the technical order-word, per the register rule on institution vocabulary.
- 0620D, VERS. 3 entire. Both lemmata agree with 1:3; the gloss construes without strain.
- 0621A, VERS. 7 gloss and VERS. 8 gloss. Clean apart from the lemma case noted in §2a.
- 0621B, VERS. 10 gloss. Clean.
- 0622A, VERS. 4 gloss (Augustine's in templum discussion). Clean; the Greek-reading argument is Migne's own and is reproduced whole, see §2c and §5.3.
- 0622B, VERS. 6 gloss on Suo tempore. Long and clean; a single ablative-absolute chain (Completa accessione … qua impleta … impleto tempore) rendered as three, with the sense order preserved.
- 0622C, VERS. 7 gloss. The Nero/Enoch/Elias paragraph is clean; the four Vel limbs of Donec de medio fiat are four distinct readings and are kept as four, not merged.
- 0623A–0623B, VERS. 12/13/15 glosses. Clean apart from the address offset of §1b.
- Chunk 1 entire, VERS. 1, 6, 9, 11, 13 glosses. Clean.
§8 · WHAT A BLIND READER SHOULD HIT HARDEST
In order of how much I want a second pair of eyes.
- ⭐⭐ 0623C · Et malis, qui persequuntur infideles. The one call I am least sure of, and it is the site to check first. Two readings both construe: - (a) infideles accusative, the object: "who persecute unbelievers." The grammatically dominant reading of the bare words. But it makes no sense in the argument: the gloss is dividing Paul's importunis et malis hominibus into unbelievers who resist (already defined as infidelibus one clause earlier) and wicked men who persecute — and what wicked men persecute is the faithful, not other unbelievers. - (b) infideles nominative, predicative to qui: "who, being unbelievers, persecute." Requires no emendation, keeps every printed word, and construes with the argument. - Adopted: (b), rendered "who, being unbelievers, persecute." Reading (a) was rejected on sense; a third possibility, that infideles is a compositor's error for fideles ("who persecute the faithful"), was rejected because it needs the plate to be wrong and (b) does not. ⚠ If the foot-of-page apparatus or the underlying Glossa tradition shows fideles, this becomes a
[sic:]site and the English should change. Flagged, and a call was still made rather than left open. - ⭐ 0621B · non tamen semper deficient (§6). The rendering "will not always fail" is literal and the antithesis is genuinely odd in English. Check that it has not been quietly improved.
- ⭐ 0622C · jam … revelabilur — the subject of revelabitur (§4, last entry of the 0622C block). ille iniquus is adopted over discessio; both formally construe.
- 0621B · ut dignetur, id est dignos judicet — the object "you," supplied from Vulgate 1:11 rather than from the sentence (§4). The single most Vulgate-dependent word in the English.
- 0622A · refuga rendered "deserter" (§2a). Alternatives "renegade" and "apostate" were weighed; "deserter" was chosen because utique a Domino Deo names what he deserts from, and "deserter … from the Lord God" carries that in English where "apostate from" is pleonastic and "renegade" loses the direction.
- 0624B · the four-negative sentence (§6, §9.2). All four survive and each is now in its printed position; a tally alone will not confirm this, so read the scope.
- The verse addresses (§1). Confirm nothing was renumbered anywhere.
§9 · WHAT A BLIND POLARITY READ FOUND, AND THE MECHANISM
A blind polarity reader, which never saw this file, read 8987 and returned three real defects, all in col. 0624B. The orchestrator then confirmed all three against the plate (archive.org patrologiaecurs06saingoog, leaf n316, right-hand column, read at full resolution), so the Latin is settled and these were repairs, not proposals. Two further punctuation restorations and one negation-exponent fix came with them. All are now in the text.
⛔ Where this happened matters more than what it was. Each of the three sits one paragraph from the [var:] at 0623C that this file already reports as correctly taken. Consistent apparatus use two columns away is not evidence of use here. That is the documented shape of the failure, and §2's list of fifty-seven checked lemmata did not protect the prose between them, because the lemmata were what I was told to check and the gloss prose was not.
9.1 · ⭐⭐⭐ velint rendered as its opposite — the serious class
Migne prints non solum facere velint, sed nec intelligere. I wrote "not only unwilling to do." velle is "wish" everywhere else in this same work, including ut velint haberi ministri Christi at 0622C and quin velit at 0621C, both of which I rendered "wish." Only here did it turn negative.
The mechanism, and it is the useful thing on this page: a negation was relocated to make a sentence run. The literal rendering does not construe, because Augustine's received text reads nolint and Migne's plate has lost the n. Facing a sentence that would not construe, I took the non of the correlative non solum … sed and spent it a second time on velint. That is free of charge in English and invisible afterwards: no negative is added and none is lost, so every count-based check passes. Nothing in the apparatus recorded it, and the resulting sentence reads better than the true one.
⛔ That is precisely the case the rule is written for. The defect belongs to Migne, and this edition exists to show it. Now rendered "so that they not only wish to do what charity profitably admonishes, but not even to understand it," with [sic: velint] on the verb. The sentence is odd. It is supposed to be.
9.2 · ⭐⭐ sed non ut smoothed into ut non — the same mechanism, one line later
Migne prints sed non ut ista intueantur: "but not in order that they may look upon these things." I wrote "but that they should not look upon these things," which is ut non — the same two words, the negation moved across the conjunction, and the scope changed with it.
⚠ This is why §6 could report "all four negatives are present in the English" and still be wrong. Presence is a tally; scope is not. A relocated negation is invisible to any check that counts. It is also invisible to the translator, because the version that survives is always the one that reads better.
And, exactly as in 9.1, the sense the passage wants is the smoothed one: the following nec … faciant shows it, and Augustine reads sed ut non. ⛔ That is the argument for logging the defect, not for translating it away. Now literal, with [sic: sed non ut].
9.3 · ⭐⭐ doceat given a second agent
cum multis aliis locis epistolarum quid hic sentiat Apostolus apertissime doceat. I supplied "he" for doceat and left Apostolus on sentiat, producing two agents where the paragraph has one. The cum-causal is the reply to the mist-drawers, and its whole force is that Paul himself settled the question elsewhere; Augustine's transmitted text has apostolus as the subject of doceat. Repaired to "since the Apostle in many other places of the epistles most plainly teaches what he here means."
⚠ §4 had flagged this site, set out both readings, and then supplied the wrong one anyway. A flagged passage is not a checked passage. The flag stands in this file as evidence that the warning was read and did not work.
9.4 · Punctuation restored — matched by position, never by tally
Migne's mark is authoritative wherever it falls, and four had been normalised toward English habit rather than reproduced:
| col | Migne prints | I had | now |
|---|---|---|---|
| 0621C | colon at quod fides exigit: et sine quibus | comma | colon |
| 0622B | two colons (ab utroque imperio: … intrat ad fidem:) | semicolon, then colon | colon, colon |
| 0622C | colon at de medio mundi: quia omnes undique | comma | colon |
| 0624B | Matt. VI quoted with an internal comma at its first occurrence and without one at its second | printed without both times | each reproduced as printed |
⭐ The 0624B item is its own small lesson: I had conformed the two citations to each other, which is the same instinct as conforming a lemma to the Vulgate, applied to punctuation. The plate prints them differently and now so do we.
9.5 · non … voluisse at 0622B given an exponent
Apostolum non id aperte scribere voluisse. "Was unwilling to write" is not wrong in sense, but it leaves the printed non with no exponent in the English, so a later checker grepping the negations will not find it, and voluit/noluit is a named class in this work's defect vocabulary. Now "did not wish to write this openly." §6's entry for this line, which argued for "was unwilling," is superseded: it weighed the sense and forgot the checker.
9.6 · Confirmed correct by the plate and by the reader — recorded so they are not churned
- 0623C · qui persequuntur infideles read as nominative-appositional. My §8.1 call stands; the preceding clause settles it. ⭐ The site I was least sure of was not the site that was wrong.
- 0622C · revelabitur with ille iniquus as subject: the ablative absolute has already revealed the discessio.
- 0621C · the non … quin … ut triple: all three members survive.
- 0624C · eorum enim deus venter est, qui … provident, the mismatched relative: confirmed on the plate, rightly not smoothed.
- Both misnumbered
VERS.runs in CAPUT II (§1a, §1b): correctly reproduced. - 0622C · tollatur and manifeste appareat rendered jussively: declined by the orchestrator as not worth churning. The subjunctive is the lemma's own, carried over from the suppressed donec, and the distinction is drawn correctly three lines on.
- 0621B · the double quotation marks on "to give" / "to undergo": kept. English needs them to mark cited words.
9.7 · CLOSED · the one-word discrepancy was the CROP, not the plate and not the twin
Resolved on a re-read of leaf n316 with the left margin widened. The plate prints quidquid Evangelii praeconio facere jubentur.* Our Latin twin agrees, and the English "whatever they are bidden to do by the proclamation of the Gospel" is correct. Nothing changed; no source-pipeline gap.
⭐ Keep the mechanism, because it cuts against the plate reader as much as the translator. The first transcription read debentur. This Google scan clips line-initial characters down the inner column of several leaves, and jubentur begins its line, so the crop returned …bentur — and the missing head was completed from context. That is exactly how a false reading enters the record: not by misreading a letter, but by supplying one that was never seen, in a word that then construes.
⚠ It is also the same mechanism as §9.1 and §9.2, one level up. There, a sentence that would not construe was made to run by moving a negation; here, a word that would not construe was made to run by restoring a letter. Both repairs are invisible afterwards and both produce something that reads better than the truth. The clipping is a known property of this scan: it is what made 8970's Nephthali need a 900 dpi read.
The rule, for the next book: when a plate transcription and the Latin twin differ by one word, widen the crop before believing the plate. ⛔ And note which way this one went: the twin was right and the plate reading was wrong, so "the plate settles it" is true only of a plate that has actually been seen whole.
Also confirmed on the same re-read: sed non ut ista intueantur, nec propter ista faciant. Repair 9.2 stands exactly as made.
9.8 · The standing lesson for the next Glossa book
A negation relocated to make a sentence run is the defect that no count can see. Both 9.1 and 9.2 add nothing and drop nothing; they move something. Every parity check in the toolchain passed on both, twice, and this file argued the wrong reading confidently in prose.
The only test that catches it: for each printed negative, name the single word it attaches to in the Latin, then find that same word in the English and check the negative is on it. Not "are all the negatives present," which was the test I ran and reported passing in §6.
⭐ And the second lesson: the failure was in the gloss prose, not in the lemmata. The launch instructions named the lemmata, I checked all fifty-seven, and all fifty-seven were right. The three defects were in the sentences between them, where nothing had been named.