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 →
Every judgment call logged individually, per column. Patterns cited are those of translation-style.md.
0095D (chunk 0000)
Printed: duae sequentes Malachim Ramatham, sicut in Hebraeo, vel ut quidam Arimathaim, unde Arimathia in Evangelio. Rendered: "the two following, Malachim Ramatham, as in the Hebrew, or as some Arimathaim, whence Arimathea in the Gospel." Note: The sentence runs two separate notes together with no break: the Hebrew title of 3–4 Kings (Malachim) and a gloss on the place-name Ramatha / Arimathaim. Probably a lost stop or a lost lemma between Malachim and Ramatham. Rendered as printed, unpunctuated between the two words (Pattern 8: never supply a mark the plate lacks).
0096D–0097A (chunk 0001)
Printed: quinque anos aureos … quinque anos singulae singulos Rendered: "five golden anuses" … "the five anuses, each province one apiece." Note: Translation choice, not a defect. Migne prints the Vulgate's anos (1 Kings 6:4–5); the conventional English versions read "emerods." Per the scripture rule, the printed Latin word is translated, not the traditional English equivalent.
0097A
Printed: quos, licet pro dignitate specialiter civitates quasi capita provinciarum dedisse dicuntur, tamen pecunia … collecta est Rendered: "which, although for dignity's sake the cities in particular, as the heads of the provinces, are said to have given them, yet the money was gathered…" Note: Anacoluthon in the plate: accusative quos is governed by nothing (the main clause switches to pecunia … collecta est), and dicuntur is passive while civitates … dedisse wants an accusative-infinitive under an active verb. Rendered as printed; not emended.
0097C
Printed: Saed in Hebraeo plebis non habetur. Rendered: "Saed in the Hebrew 'of the common folk' is not found." Note: Saed is not a word. Almost certainly Sed ("But"), with a stray a. Carried through untranslated in italics under Pattern 7's non-word rule.
0097C
Printed: ac si dixisset. Quinquaginta millia et septuaginta percussit… Rendered: "as if he had said. Fifty thousand and seventy he struck…" Note: The full stop breaks the sentence mid-construction (ac si dixisset is left without its clause). Preserved exactly, per Pattern 8 — never delete a mark the plate prints.
0097B
Printed: Et usque ad Abel magnum? Rendered: "And as far as great Abel?" Note: A question mark closing a scripture lemma that is not a question (1 Kings 6:18). Preserved, Pattern 8.
0099A
Printed: Hebraeus habet sic: Vagham, Neza Izrahelloe Zachir, ; id est, Rendered: identical, verbatim: "The Hebrew has thus: Vagham, Neza Izrahelloe Zachir, ; that is," Note: The transliterated Hebrew is carried through untranslated. The stray , ; marks the place where Migne's plate sets (or fails to set) the Hebrew type itself; the empty slot and both marks are reproduced exactly, Pattern 8.
0099A
Printed: sine virtutis Israel, id est dans virtutem et fortitudinem Israel Rendered: "without the strength of Israel, that is, giving strength and fortitude to Israel" Note: Unconstruable as printed — sine governs the ablative, not the genitive virtutis, and the sense of the passage requires the alternative gloss sive virtus Israel ("or, the strength of Israel"), matching the Hebrew rendering quoted two lines above (porro virtus Israel non mentietur). Rendered as literally as the printed words allow; not emended (Pattern 7).
0099C
Printed: Nam coram Domino est Christus ejus? Rendered: "For before the Lord is his Christ?" Note: Christus kept as "Christ" throughout rather than resolved to "anointed," since Hugh's gloss turns on the word itself (cf. 0103B Christo Dei Jacob). Question mark preserved.
0100C (chunk 0002)
Printed: note order [n: (I REG. XXVI.)] followed by [n: (I REG. XXIV.)] Rendered: verbatim, in the printed order. Note: Migne's marginal chapter notes run out of sequence here — XXVI before XXIV — and the lemma tagged XXIV (Est apud me multis diebus vel annis) is in fact 1 Kings 29:3. The notes are reproduced exactly as printed and in the printed order (verbatim rule); no reordering.
0100C
Printed: nunc autem ac his dici, multis diebus vel annis moratum apud se Rendered: "but now ac his to be said, that he had stayed with him many days or years" Note: ac his is not a construable run; almost certainly ac si ("as though"). Carried through untranslated in italics, Pattern 7 non-word rule.
0100C
Printed: quod fortassis idcirco dicitur rectius argumentum fidelitatis ipsius caperent ex temporis prolixitate Rendered: "which perhaps is said for this reason, that they might the more rightly take an argument of his fidelity from the length of the time" Note: The plate has lost the conjunction governing caperent (expect ut after idcirco). Rendered as the printed words allow, with the purpose sense the subjunctive carries; nothing supplied in the text.
0102A (chunk 0003)
Printed: Quare cogitasti hujuscemodi rem; Rendered: "Why hast thou thought such a thing;" Note: A semicolon closing a direct question. Preserved, Pattern 8; not converted to a question mark.
0102B
Printed: ut si sensus; veni ut loquar verbum Rendered: "so that if the sense; I have come that I may speak the word" Note: Defective — expect ut sit sensus ("so that the sense may be"). The English is deliberately unidiomatic here so that the defect stays visible; the semicolon is preserved, Pattern 8.
0102B
Printed: Loquar regem si quo modo faciat rex verbum ancillae suae; Rendered: "I will speak to the king, if in any way the king may do the word of his handmaid;" Note: loquar takes a bare accusative regem (expect ad regem, as at 2 Kings 14:15). English cannot carry the case defect; rendered with "to." The semicolon closing a conditional is preserved.
0102C
Printed: Captatio est, ut attendat qui agendum sit ut sapiens. Rendered: "It is a winning of favor, that he may attend, as a wise man, to what is to be done." Note: qui stands where quid is required by the indirect question. English has no exponent for the pronoun's gender/form here, so the sense of quid is unavoidable in the rendering; logged rather than silently corrected.
0102B
Printed: in Haebraeo expressius invenitur: chi, chire, , id est, Rendered: "is found more expressly in the Hebrew: chi, chire, , that is," Note: Transliterated Hebrew carried verbatim; the doubled comma marks the slot where the Hebrew type stands (or is missing) on the plate, and is reproduced. Haebraeo is a period spelling variant, translated normally per Pattern 9 — no carry-through.
0103A
Printed: Arepham [Arafam] and Sephi [Zap] and Arepha [Arafa] Rendered: the bracketed matter carried untranslated, exactly as printed. Note: These square brackets are Migne's own variant readings on the Hebrew proper names (same footing as [ al., …]), so they pass through the English untranslated.
0102D–0103A
Printed: Ab initio messis donec, stillaret aqua super eos. Rendered: "From the beginning of the harvest until, water dropped upon them." Note: The comma falls between donec and its verb, breaking the clause. Preserved, Pattern 8.
0104A (chunk 0004)
Printed: Hasta lignum est, lancea ferrum quasi diceret. Quin in eos saevire voluerit, utetur… Rendered: "The shaft is the wood, the lance the iron; as if he said. Indeed, whoever shall have wished to rage against them will use…" Note: Two defects in one place. (1) The full stop after diceret cuts the quasi diceret clause from what it introduces — preserved (Pattern 8). (2) Quin stands where Qui ("whoever") is required; Quin is a real word, so it is rendered literally ("Indeed") and not carried as non-word type, but the reading is almost certainly a compositor's slip.
0104B — dittography
Printed: in quibus David primum ponit, ac primum ponit ac principem, secundum Eleazarum Rendered: "among whom he puts David first, [d: and puts him first] and as chief; Eleazar second" Note: The run ac primum ponit is set twice; the intended text is in quibus David primum ponit ac principem. This reads as a compositor's doubling (eye-skip on the repeated primum ponit), not a rhetorical doubling by Hugh. Both occurrences are printed; the repeat is marked [d: …] per Pattern 11.
0103D
Printed: cuncta salus mea et omnis voluntas mea., Rendered: "all my salvation and all my will.," Note: The plate sets a full stop and a comma together inside the italic lemma. Both preserved, Pattern 8 — neither deleted.
0104B–0105A — the list of the mighty men
Printed: the section under the head Incipiunt nomina fortium in Israel. consists of bare numerals — - 2 (three times, then 3, 4, … 26) — followed by five pipe-separated rows carrying names and the numbers 27–31. Rendered: numerals reproduced exactly as printed, in the printed order, including the repeated 2 and the absent 1; the five named rows translated, their numbers and pipe layout kept. Note: This is the largest defect in the work. The names of the thirty-seven fortes have dropped out of the transmitted text and only their tally column survives; the tally itself is broken at the head (2, 2, 2, 3 …, no 1). Nothing was reconstructed from 2 Kings 23. Worth Wilson's eye — this may be a defect in the TEI transform rather than in Migne's plate, and if so it is recoverable from the scan.
DIAGNOSED 2026-07-28 — upstream data loss at Corpus Corporum, NOT our chunker, and NOT recoverable from our source. The TEI at this point reads literally <list><item/><item>2</item><item/></list> and so on: the name cells are empty elements. Corpus Corporum transcribed the tally column and dropped the names, so the chunker had nothing to lose — it reproduced empty items faithfully. Note that entries 27–31 survive intact (<row><cell>Eleazar</cell><cell>28</cell></row>) because those are marked up as a <table> rather than a <list>, which is itself evidence that the loss is a markup artifact rather than a defect in Migne.
Recovery therefore requires the PL 175 scan at cols 0104B–0105A, not a re-harvest, and would land as a data/tei-patches/11055.json entry with plate provenance — the same shape as the 11085 patch. Deliberately not attempted here (it is a scan-reading task, not a translation one) and left as a recorded gap per the editorial method: where the source is missing, the gap is recorded as a gap and nothing is supplied to paper over it. Until then the page shows a bare tally, which reads as our defect and is not — worth deciding whether this section should carry a visible note to the reader before it deploys.
0104B
Printed: et cum Uria, triginta septem and, at 0105D, [non adnumerato Uria] Rendered: the first translated ("and with Urias thirty-seven"); the bracketed phrase carried untranslated as printed. Note: [non adnumerato Uria] is Migne's own bracketed editorial matter (an ablative absolute inserted into the lemma), so it passes through untranslated on the same footing as [ al., …]. It is longer than the usual one-word bracket, which is why it is logged.
0106A
Printed: [n: (II REG. XXIV)]. Et addidit furor Domini. Rendered: verbatim, with the full stop outside the note bracket. Note: Migne sets this note's terminal punctuation outside the parenthesis, unlike every other note in the work. Reproduced exactly.
0107A (chunk 0005)
Printed: In hoc patet quod oraculum viginti cubitis inferius erat reliqua domo. Rendered: "From this it is plain that the oracle was lower than the rest of the house by twenty cubits." Note: The arithmetic is wrong as printed: the temple is thirty cubits high (3 Kings 6:2) and the oracle twenty, so the difference is ten, and Hugh's own premises give that figure. Expect decem cubitis. Rendered as printed; not emended (Pattern 7).
0107B
Printed: reliquas duas adinvicem jungerent in medio oraculi tanta fieret extensio alarum, quanta fuit latitudo oraculi. Rendered: "and they joined the remaining two to each other in the middle of the oracle so great would the extension of the wings be as was the breadth of the oracle." Note: The subordinating ut before tanta has dropped out, and the plate prints no mark at the join. Rendered without a supplied conjunction or a supplied comma (Patterns 7 and 8); the English is correspondingly rough.
0107D (chunk 0006)
Printed: Hoc est quod dicit, cooperatura tres ordines, subauditur habebat Rendered: "This is what he says, the cooperatura three rows — "had" is understood" Note: cooperatura is not a lexicon form in this sense; the context (a boarded ceiling) wants coopertura, "covering." Carried through untranslated in italics, Pattern 7 non-word rule. The straight quotes around "had" are ours (Pattern 6), supplied because English needs them and the plate marks the understood word by subauditur alone.
0109C
Printed: duo ordines sculpturarum histriatarum hobus supposita duodecim Rendered: "two rows of histriatarum carvings, twelve hobus set beneath" Note: Two non-words in one clause. histriatarum is presumably striatarum ("fluted"); hobus is presumably bobus ("oxen," dative/ablative of bos, as at 0109B duodecim boves ex aere fusos). Both carried through untranslated in italics, Pattern 7.
0109D
Printed: ubi illi ordines sive versus jungebantur sulpturae variae in lineam ductae, inter versum, et versum discretationem facientes Rendered: "where those rows or lines were joined there were various sulpturae drawn out in a line, making a discretationem between row and row" Note: Two more non-words. sulpturae is sculpturae with the c dropped (the word is spelled correctly six times in the same column). discretationem is discretionem with an intrusive -ta-. Both carried through in italics, Pattern 7.
0110B
Printed: quia super quatuor numerulos in quatuor angulis erectos apparebant Rendered: "because they appeared above four little numbers raised at the four corners" Note: numerulos is a formally well-made Latin diminutive of numerus, so it is not non-word type and is rendered literally — but the sense is plainly humerulos, "little shoulder-pieces," which Hugh uses twice in the same description (0110A and 0110B, fortassis humerulos significat, qui in angulis erant columnarum). The literal English is therefore nonsense on the page; that is the intended cost of Pattern 7, and the conjecture is here.
0110D
Printed: id est, si astrictuo est adversus proximum aliquo juramento Rendered: "that is, if he is astrictuo against his neighbor by some oath" Note: astrictuo is not a word; expect astrictus (the lemma four words earlier reads quo teneatur astrictus). Carried through in italics, Pattern 7.
0109B
Printed: Annon totum circuibat, sed decem tantum cubitis ipsius ambitus sculpti fuerunt. Rendered: "Did it not go round the whole, but only ten cubits of that circuit were carved." Note: A direct question closed with a full stop (the parallel question two lines above, Quomodo decem cubitis circuire poterat…?, takes a question mark). Preserved as printed, Pattern 8.
0111A–B (chunk 0007)
Printed: Aedificavit mello, et caoequavit voraginem civitatis David patris sui Rendered: "He built Mello, and caoequavit the chasm of the city of David his father" Note: caoequavit is not a word — coaequavit, with the o and a transposed; Hugh's own gloss five words later spells it correctly (coaequavit ipsum). Carried through untranslated under Pattern 7. It sits inside an italic scripture lemma and therefore takes no second italic marker (nested italics are not expressible), per Pattern 10.
0111C
Printed: In Hebraeo sic habetur: Omnes os quod non osculatum est eum. Rendered: "In the Hebrew it stands thus: All the mouth which has not kissed him." Note: Omnes (plural) with os (singular) — the lemma five words earlier reads correctly Omne os. English can carry this one, so "All the mouth" is printed rather than smoothed to "Every mouth."
0112B
Printed: id est torrentis hujus, qui modo exsiccatus est, alveum arentem fodit altius sive profundius Rendered: "that is, the dry bed of this torrent, which is now dried up, he digs higher, or rather deeper" Note: fodit is a third-person indicative where the gloss on an imperative lemma (Facite … fossas) wants the imperative fodite. Rendered as printed, in the third person, though it breaks the syntax of the gloss.
0112C
Printed: Abiit ab eo ergo quasi aliquando terrae spatio Rendered: "He went therefore from him as it were at some time by a space of land" Note: aliquando ("at some time") stands where the whole point of the gloss — the ambiguity between a space of land and a time — requires aliquanto ("by some little"). Rendered as printed; the argument accordingly reads slightly against itself, which is the evidence worth preserving.
0113A
Printed: Joam scilicet filium ejus Rendered: "Namely Joam his son" Note: The king is Joas (4 Kings 13:25). Joam kept as printed; proper names in the lemmata and glosses are not normalized to the biblical spelling.
0113B–0114A
Printed: column anchors run [0113B] … [0114A] Rendered: verbatim, in that order. Note: Columns 0113C and 0113D are absent from the transmitted text — the sequence jumps a full page-column pair mid-work. Reproduced as given; not interpolated. Worth Wilson's eye if a continuity check flags it.
<!-- corpus-polarity-sweep 2026-07-28 --> ## Corpus polarity sweep — 2026-07-28
Blind Opus read of the Latin/English pair for sense-reversing defects (one word or one letter asserting the opposite of what the context requires). 3 site(s). Migne's defects are recorded here and rendered literally in the English; our own errors were fixed in place. Full context: data/sweeps/corpus-polarity/REPORT.md.
0096D — dropped-word (migne, medium confidence)
Printed: Postea sequitur, et lucerna Dei exstingueretur. Samuel [0096D] autem dormiebat in templo Domini**
Conjecture: antequam lucerna Dei exstingueretur (Vulg. I Reg. iii, 3 'lucerna Dei antequam extingueretur'); Migne's plate has lost 'antequam' in the re-quoted lemma
The very next clause of the same note reverses it: 'et hoc antequam lucerna Dei exstingueretur' ('and this before the lamp of God went out'), and the note then explicitly re-cites the phrase as 'Quod autem dictum est antequam lucerna exstingueretur'. Hugh's second explanation depends on the lamp still burning — 'lucerna lucente, nullum alium in templo praeter Heli esse videret' ('while the lamp was shining, he could see no one else in the temple except Heli'), which is impossible if the lamp was going out.
0099A — sense-reversing-letter (migne, high confidence)
Printed: quod Deus qui est triumphator in Israel, quia per eum Israel triumphat et vincit; sine virtutis Israel, id est dans virtutem et fortitudinem Israel; ipse non mentietur
Conjecture: sive virtus Israel — 'or the strength of Israel'; the printed 'sine' is 'sive' with n for v, and 'virtutis' follows the corrupted preposition into the genitive
The gloss attached to the very phrase says the opposite of 'without': 'id est dans virtutem et fortitudinem Israel' ('that is, giving strength and fortitude to Israel'). The clause exists to pair the two renderings just quoted — the Latin lemma 'triumphator in Israel' and the Hebrew 'virtus Israel' — so it must be the disjunctive 'sive'. Hugh uses exactly that construction a few lines later on the same lemma, where Migne prints it correctly: 'sed ad effectum perducet verbum suum, sive non parcet quasi miseratione aliqua mutatus'.
0099B — sense-reversing-letter (migne, high confidence)
Printed: *Nam coram Domino est Christus ejus? Id est nunquid iste qui tibi nunc praesentatus est, et coram adductus, provisus est a te, ut sit Christus tuus?*
Conjecture: Num coram Domino est christus ejus? (Vulg. I Reg. xvi, 6); Migne's 'Nam' is 'Num' with a for u
Hugh's own gloss on the lemma renders it as a question expecting the answer 'no': 'Id est nunquid iste ... provisus est a te, ut sit Christus tuus?' — 'nunquid' is precisely the equivalent of 'num', and Migne himself prints the lemma with a question mark. 'Nam' ('for') turns Samuel's doubting question into a flat assertion that this man IS the Lord's anointed, which the whole note then goes on to deny (the note concludes 'et verum est, quod de reliquis nullum elegit Dominus'). Our English preserved the defect, producing 'For before the Lord is his Christ?' — an assertion punctuated as a question.