At a glanceKnown key applied · partial (95% of the cipher)
- The letters
- Guillaume Ancel, French resident at the Imperial court in Vienna, to a “Monseigneur” in France: 22 October, 2 November, 14 and 21 December 1577. Clear letters with long passages in cipher.
- Where
- Bibliothèque nationale de France, Cinq Cents de Colbert 398, pp. 524–526, 539, 561–563 and 565 (Gallica views 558–599). Most other cipher pages in the volume carry a decipherment in the margin; these do not. No DECODE record.
- The cipher
- A symbol alphabet with two to four signs per letter, signs for double letters, a few word signs, and code words written in plain letters (saturnus, Juno, my) mixed with plain-letter nulls.
- How it was deciphered
- Harumi-Matsuba transcribed the signs with Claude and checked them with OpenAI Codex, then applied Satoshi Tomokiyo’s reconstructed key. Here the files were re-measured with this project’s language models and three lines were compared with the images.
- What it says
- Archduke Matthias slipped away to the Low Countries; Ancel thinks the Emperor will deny any part in it, but notes he plays both sides with Spain while Don John’s envoy Varambon waits for leave to levy troops.
- Credit
- Reading, transcription and checks: Harumi-Matsuba (issue #53, 20 October 2026), done with AI only (Claude and Codex). Key: Satoshi Tomokiyo. No human has checked the transcription.
- Still open
- A human check of the transcription; most code words; 29 of 125 lines the contributor marks as conjectural.
01 The letters
Cinq Cents de Colbert 398 gathers original letters on German affairs addressed to Charles IX, Henri III, Catherine de Médicis and their ministers, among them the despatches of the two French residents at the Imperial court, Jean de Vulcob (1575–76) and Guillaume Ancel (1576–78, 1584). Most of their cipher passages were deciphered on arrival in the left margin. Four letters of late 1577 were not:
- 22 October 1577 (heading on p. 523; cipher from the end of p. 524 to p. 526): the flight of Archduke Matthias, the Emperor’s brother, to the Low Countries.
- 2 November 1577 (p. 539): six lines.
- 14 December 1577 (pp. 561–563): the Marquis of Varambon, Don John of Austria’s envoy, and the Spanish ambassador Borgia. It builds coherent sentences but reads at 94.8%, under the 95% bar.
- 21 December 1577 (p. 565).
The letters open “Monseigneur” rather than “Sire”, so the addressee is not the King himself; who it is has not been settled. The single cipher line at the top of p. 570 turned out to be deciphered in the last margin line of p. 569, and the key gives the same words.
02 The key
Satoshi Tomokiyo reconstructed the cipher of Vulcob and Ancel (1575–1584) from the margin decipherments of this volume and published the table on Cryptiana (“French Ciphers for German (and Eastern) Affairs”, 15 March 2026), noting that it would allow the letters without a contemporary decipherment to be deciphered; he published no readings. The table has two to four homophones per letter (4, 7 and α for a; ω and π for e; ∂ and μ for i), single signs for doubled letters (ss, tt, pp, rr, mm, nn), and word signs such as plus = que, ros = pour, ucl = est. He also notes that Vulcob’s letter of 2 August 1572 (BnF fr. 18989) uses it. George Lasry’s 1573 letter of Charles IX to Vulcob (vulcob1573) is in a related cipher: he compared it with a key in this volume and found 14 shared letter forms.
The contributor added a few values from context and from the margin of p. 33: an m-shaped sign = uu, a flat L-like sign = g, the crossed q = ff, a lone c = null. Names and titles hide in plain-letter code words. Those checked against margin decipherments elsewhere in the volume: saturnus = l’Empereur (p. 258), Juno = le Roy d’Espagne (p. 440), my = qu’il and my3θ = qu’ilz (pp. 258, 321).
03 The reading
The contributor’s line-by-line key output and word division are in READING.en.md; code words are in brackets. Two passages:
22 October 1577, p. 525 ll. 9–13: “…tellement que a peine pouvoit il faire un pas hors de sa chambre sans le congé d’iceluy seigneur [saturnus] son frere; tant s’en faut qu’a une telle heure il eust peu sortir du chasteau, qui ferme toute la nuit…” — Matthias could hardly step out of his room without his brother the Emperor’s leave, still less leave a castle shut all night; so, Ancel argues, the Emperor will deny any part in the escape.
14 December 1577, pp. 561–562: the Emperor must declare himself for the King of Spain [Juno] or for the Archduke; if for the first, he will show it by giving Don John leave to levy troops in Germany, as Varambon has asked; meanwhile Borgia says that if the Emperor backs his brother he must do it in earnest, “d’autant que le roy catholique fera toutes les provisions possibles pour maintenir son droit”.
On 2 November the Emperor “fait toute la demonstration [qu’il] luy est possible pour persuader les Espagnolz que tel depart est advenu a son desceu”; on 21 December Varambon is back from Archduke Charles’s court with nothing, and the Emperor waits for the King of Spain’s answer.
04 How it was checked
By the contributor (all with AI): a first transcription in sign names by Claude Sonnet 5.5 and Opus 5.5 from enlarged tiles; a cross-check transcription by OpenAI Codex from the line images, without the key; two rounds of image adjudication by Claude Sonnet; then three independent Codex transcriptions, which agree with the final one on 95.2–96.1% of values (lowest page 88.8%, p. 565). With 1,000 shuffled keys, none scores as well as the real key on any page. As a calibration, two pages that do have a margin decipherment (pp. 258 and 285) were run through the same procedure without the margin being seen: 85.9% of letters agree with the margin.
Here (21 October 2026): the files were copied into the repository and re-measured (measure.py). Of 4,385 cipher tokens on the seven pages, 98.4% get a value from the key or a checked code word. Under the project’s 16th-century French model the key output scores −1.8 to −2.3 per character, against −6.3 to −6.6 for shuffled keys. Split into words of a 16th-century vocabulary, 95.4% of tokens fall inside words, against 50% for shuffled keys; this is a lenient measure, an upper bound on what reads. Recomputing the calibration from the contributor’s files gives 472 of 556 letters, 84.9%. Three lines were compared sign by sign with the BnF images (p. 525 ll. 9–10, p. 565 l. 5): all 126 signs agree in order and count; two ∂/δ (i/n) choices could not be told apart at 2,400 pixels and rest on context, as the contributor says of that pair.
The figures support the reading as French and the key as right. They do not make it a checked reading: the transcription is AI-made throughout, the adjudications are AI judgements, and the contributor marks 29 of 125 lines as conjecture or unclear syntax. The extent therefore stays partial.
05 What remains open
- A human check of the transcription against the images, starting with the 29 conjectural lines.
- Code words without a meaning: manua/manue (a person), mare, grinter, totum (probably l’Empire), nez, bel, mim; probable nulls fec, sec, sanc, sane, pome, estre, roc, vib.
- Look-alike sign pairs settled by context: 3/ʒ (m/r), p/φ (f/m), ∂/δ (i/n), 9/q (d/l), the cross/4 (o/a).
- Unclear stretches: p. 526 ll. 9–14, p. 539 ll. 2–3, p. 561 ll. 1–4 and 18–19, p. 562 ll. 25–26 (line ends lost in the gutter), p. 565 ll. 9 and 15–16.
- The addressee (“Monseigneur”) and the signatures were not checked.
Full texts
The transcription and the decryption of every document on this page, so that the reading can be used and checked. Short files are shown here; all are in the repository.
22 Oct 1577, pp. 524 end-526
Ancel on Archduke Matthias's flight to the Low Countries: he thinks the Emperor [saturnus] will deny any part in it, since the young prince could hardly step out of his room without his brother's leave or leave the castle, shut at night (contributor's summary).
Transcription
# BnF Cinq Cents de Colbert 398, p.525 (Gallica btv1b525218133 f559). Transcribed from tiles gallica/tiles/c398_f559_r{1..6}c{1,2}.jpg (symbol transcription only, not deciphered).
# i (hook) vs dl (tall-stem delta) is often hard to tell at tile resolution; those readings carry "?" when doubtful. "at" = circled "au" sign (l.7). Page appears to end at l.25 (nothing below in tile r6).
# Letter date: p.523 (f557) heading '22. octob. 1577'; p.523-526 are one letter (p.524 continues without a new heading). Opening: 'Comme la negotiation du Sr de Malstet envoyé depuis un mois en ça de la part des Estats du pays bas vers l'Empereur ...'
# p.524 last line (f558, gallica/tiles/f558_lastline.jpg), after clear '... ce qu'il se remembroit bien vest(?).' :
# Adjudicated by image against an independent Codex transcription (second/p525_pass1.txt), 2026-10-20
# Reverted by context, marked "?" (not an image decision): phi -> p and z -> 3 everywhere on this page (the shape-based calls gave mait/maire/mrere/maut/merme and serblable/directerent/tellerent; p (f) and 3 (m) make the words). Shape alone does not separate these here.
# Adjudicated by image a second time, against three further independent Codex readings (second/p525_pass3[abc].txt), 2026-10-20
524.99 P 4 b | 8 w ?
525.01 oc x b S | {Monseigneur} | i ne | {fec?} {ar} | 4 i? S w 7 8 oc qq oc i? S D b w 9? pi | plus | ne
525.02 q w p? 7 8 oc dl ne w 9 i? D S al i? g dl w x b | {fec} | {saturnus} | 9? w S i? b w qa x oc dl i dl D w b
525.03 P b w D pi | S oc dl? i? dl D w dl? D i oo di | {fec?} | 7 dl D x b 7 2 | S w 3? o ne? 7 o q pi | {pur} | {my}
525.04 i w x oc x 9? b 7 4 x oc x w b 9 i? b pi 8 D w 3? pi dl D 8 w p? 7 i D 9 w x 7 dl D | {sanc}
525.05 {Juno} | 7 i dl S P ne x S D oc S D n w dl pi c 8 x S w b c ne x 2 P w b S x 7 9? w b
525.06 S i? ne P w x D 9 pi dl 2 7 x oc i b i? 7 3 4 i? S {puis} P w dl S pi 3? 4 i? S 8 oc dl S i?
525.07 9? pi b w ne 7 dl 7 D x b w 9 oc x 8 w 9 x i? pi x dl w P b i? dl 8 w c ne 7 dl oc x at i? D x
525.08 b w 9 i? 8 w ne x 2 qa x i? 7 D oc x S i? oc x b S pi S D w pi dl 8 b 7 i? dl D w c b w x pi
525.09 ? pi dl 8 w D pi ne q w 3? pi dl D plus | 7 P w i dl w P oc m oc i D i? ne p? 7 i? b w x dl P 7 S
525.10 h oc b S 9 pi S 7 8 h 7 3 o b w S 7 dl S ne w 8 oc dl g w 9 i? 8 w ne x 2 S w i? g dl pi
525.11 x b {saturnus} n oc dl p? b w b pi D 7 dl di S w dl p? 7 x D qa x 7 x dl w D pi ne q pi
525.12 h w x b pi i? ne w x S D P pi x S oc b D i? b 9? x 8 h 7 S D w 4? x qa x i p? w b 3 pi
525.13 D oc x di? pi ne 7 dl x i? D P 7 b w i? ne q pi 3 w dl D dl 7 2 4 i? D oc dl 8 plus n w x 7 x 8 x dl
525.14 P w b S oc qq 7 g w 7 x P b w S 9 pi ne x 2 9 x plus ne i? ne w x S D | {estre} | {mim}
525.15 9 pi 8 oc dl p? i 7 dl 8 w plus 9 w P b pi dl 9 b pi x dl pi S w 3? o ne? 7 o ne? pi b w S oc ne? x
525.16 D i? oc dl 7 x 9 pi hash w x 9 pi S 7 3 4 g w S D pi i 3 P w b i? 7 ne w 7 x w 8 ne 7
525.17 qa x pi ne q w 7 x 8 oc dl D b 7 i? b w i? ne | {fec?} | 8 w b D 7 i dl | {my3} | 7 P ne? x S 8 oc dq x
525.18 dl i? qa x w 9 pi P x i? S P w x 9 w i? oc x b S | {my} | dl 7 x oc i D 9 pi 8 oc x S D n? 3 w
525.19 7 box w dl? 9 x 4 ? hash i? plus ne w S i pi x b 9 w 3 7 ne S D w di S 8 oc dq pi
525.20 9 w P x i S i? ne S {fec?} x pi b i p i pi ne 7 x oc i? D 9 pi 3 7 dl 9 w 7 x 3 w S 3 pi
525.21 {saturnus} | dl oc di? | ros | g oc m w b dl pi x b 7 i dl S i qa x 7 x 8 x dl S 9 i? S oc i? w dl? D
525.22 3 7 i S pi dl D i w b pi 3 w dl D ros S w i g dl? pi x b 9? pi S 9 i? D th P 7 2 S o 7 S c w dl?
525.23 D b 7 i di oc i D 7 x w 8 S 7 3 7 g w S D pi i 3? P w b i 7 ne pi dl oc dl 7 x w 8 ne pi
525.24 9 i D S w {ucl} g dl? w x b 7 b 8 h i 9 x 8 | {fec} | {pur} | i? w dl pi x oc 2 i 8 2 P w b S oc qq w
525.25 qa x i 3 oc dl S D b pi 9 pi 9 oc x D w b pi dl S 7 8 oc dl S 8 i? w dl 8 pi plus ? w ne q piTranscription
# BnF Cinq Cents de Colbert 398, p.526 (Gallica btv1b525218133 f560). Transcribed from tiles gallica/tiles/c398_f560_r{1..7}c{1,2}.jpg (symbol transcription only, not deciphered). Same letter as p525; continues from it.
# Page = 16 cipher lines (526.01-526.16), then plain text (see "# clear" lines). Tiles r6/r7 are plain text and blank paper.
# Conventions as p525. Reading rules used: cross with a V/triangle at lower right = oc; cross with a diagonal arm at left = 4. The "(y" curl = S; beta with long left stem = b. Beta tails "(" are not separate tokens. i (small curled hook) vs dl (tall loop) decided from 1.85x stitched crops; doubtful ones carry "?".
# Plain "g" shape ("[" with curved top) = g. Punctuation dots (after l.01 "7", l.05, l.07, l.09-10, l.14, l.16) are omitted. 4-bar heavy "ne"-like sign (l.01, l.02) read as ne? (possibly a distinct sign). Stem + filled diamond (l.02, l.10) = t4? (unmapped shape). Big pen flourish before "Admira" at end of l.16 is not tokenised.
# clear (l.16 end, continues): "Admira[l] ..." ; next line: "Castille qui bien d'artifice a trouue aste nouuelle bien estrange." (plain text from here; clear lines below not cipher)
# clear: Castille qui bien d'artifice a trouue ... nouuelle bien estrange.
# clear (new paragraph): Mardy dernier aux Estatz de la baße Austriche en presence des Archiducs / Ernest, Mathias et Maximilian / de fidelite et obeissance. En quoy ilz s'estoient monstrez difficiles au commancement ... desirant que ... confirmast par escrit les libertez ... (about 8 further plain lines, ends "... pour le regard de la ... Archiducs primi..."; not transcribed in detail)
# Adjudicated by image against an independent Codex transcription (second/p526_pass1.txt), 2026-10-20
# Reverted by context, marked "?" (not an image decision): al -> qa back (the image call gave auon/auenuiron; qa (q) makes qu'on/qu'environ). The flat L sign is read g (see src/vk.py).
# Adjudicated by image a second time, against three further independent Codex readings (second/p526_pass3[abc].txt), 2026-10-20
526.01 8 h oc? S pi dl 7 2 w pi n D pi? | ros | P w dl S ne? 9 i 8 w ne x 2 S w i L dl pi x
526.02 9 pi ne i o pi b ne? c 8 oc? dl 8 q x w pi dl S oc? dl 8 oc? dl S w i ne dl oc t4? oo? o S? D?
526.03 D oc? x di w S ne w S 7 x D b w S 8 oc? dl i w 8 D x b pi S qa? x oc? dl x w i ne
526.04 7 3 pi dl w b 7 ne w dl 8 oc? dl D b w c b? i w dl | plus | P ne x S i w x b S P b w D?
526.05 dl D {my} dl 7 2 w pi x oc? M 7 S i oc? dl 9 7 x 4 dl 8 w b pi dl D w ne 8 7 S x dl
526.06 p? b pi b w | ros | 9 pi S P oc? x i ne w b S oc? dl oc? dl 8 ne w c S w S dl pi x w x
526.07 8 w | plus | i ne ne pi x b 7 up? 7 b D i w dl D 9 7 x 4 dl D 4 L w i pi dl D w dl 9
526.08 ne i pi x qa? x oc? dl 7 w x 4 hash w th 9 pi P w i dl w 7 i dl 8 i D w b | ros | 8 pi
526.09 ne w 9? i D S w i L dl pi x b 7 b 8 h i 9 x 8 ne w | plus | q 7 x D b pi 3 w dl? D
526.10 D b oc? P L b 7 dl 9 pi x oc ne oc? t4? D w ne oc? dl 7 9 i oc? x S D w 7 qa? hash i | plus?
526.11 9 i D S w i L dl w x b | {saturnus} | pi x S D 9 pi S i b w ne w p? 7 i b w 7 {au}
526.12 i ne w x S D P w x 9 pi P w S 8 h w b ne pi S S x S 9 i D th L w dl D i ne th? h ?
526.13 P ne x S 3 7 D i dl 8 oc? dq pi i ne w dl pi x D b? i w dl ne w? D pi 3 S ne pi S
526.14 dl pi P 7 b D i b w dl D qa? x w dl x i b oc? dl ne w 3 i 9 2 c S at 3 o ne 7 o q pi 3 w?
526.15 w dl 9 oc? dl S oc? x n P pi 8 oc? dl - S x b 8 pi {my} w S 8 b i x i D di oc? x di pi
526.16 ne pi S q w box b pi S {my} w dl x oc? 2 7 4 x S P b i dl 8 w S {Admira}Decryption: READING.en.md
Decryption
524.99 par ce·
525.01 ours {Monseigneur} il {fec?}{ar} aIseaco[nn]oIstreDe [que] l
525.02 leFaconledItsaIgneur {fec} {saturnus} DesIrequoninter
525.03 prete soNInteNtiot {fec?} anturay seMbLable {pur} {my}
525.04 ieuouDraauouerdIrecteMentceFaitdeuant {sanc}
525.05 {Juno} ainsplustostsenecuserluypersuaDer
525.06 sIlpeutdenyauoirIamaIs{puis}penseMaIsconsI
525.07 DerelanaturedouceduIeuneprIncelanou[rr]Itu
525.08 redIceluyquIatousIoursesteencraIntereue
525.09 ·encetelleMent[que] apeinepo[uu]oitIlFaIreunpas
525.10 horsdesachambresanslecongedIceluyseIgne
525.11 ur{saturnus}sonFreretantsenFautquaunetelle
525.12 heureIleustpeusortIrDuchasteAuquiFerme
525.13 touTelanuItpareIllementnayaItonc[que]seuaucun
525.14 perso[nn]ageaupresdeluydu[que]lIleust {estre} {mim}
525.15 deconFiance[que]deprendreuneseMbLabLeresoLu
525.16 tIonaude[ss]eudesamagesteimperIaleauecla
525.17 quelleaucontraIreIl {fec?} certain {my3} apLusco[mm]u
525.18 nIquedepuIspeudeIours {my} nauoitdecoustSme
525.19 a[tt]eNdua·[ss]I[que]lesieurdemalstetsco[mm]e
525.20 depuisIls{fec?}uerifielauoItdemandeaumesme
525.21 {saturnus} noT [pour] go[uu]erneurainsiquaucunsdIsoIeNt
525.22 maisentierement[pour]seigNeurDesdItzpaysbaseN
525.23 traitoitauecsamagesteiMperialenonauecle
525.24 ditse{ucl}gNeurarchiduc {fec} {pur} Ieneuoyicyperso[nn]e
525.25 quimonstrededouterensaconscIence[que]·elleDecryption
526.01 chOsenayeestE [pour] pensLdiceluyseigneu
526.02 deliberLcOnclueensOncOnseilnoTObST
526.03 tOuteslesautrescOniecturesQuOnueil
526.04 ameneralencOntreRien [que] plusieurspreT
526.05 nt{my}nayeeuO[cc]asiOndauancerentelcasun
526.06 Frere [pour] despOuilersOnOnclesesneueu
526.07 ce [que] illeuraUartientdauantageientend
526.08 lieuQuOnaeua[ss]ezdepeineainciter [pour] ce
526.09 leDitseigneurarchiducle [que] lautremeNt
526.10 trOpgrandeuolOTtelOnadiOusteaQ[ss]i [QUE]
526.11 ditseigneur {saturnus} eustdesireleFairea{au}
526.12 ileustpeudepescherlessusditzgentilZh·
526.13 plusmatincO[mm]eileneutRienlEtemsles
526.14 nepartirentQuenuirOnlemidys[rr]mblablemE
526.15 endOnsOuspecOnysurce{my}escriuittOute
526.16 lesle[tt]res{my}enuOyaausprinces{Admira}Reading
# Ancel, letter of 22 October 1577, pp. 524 end-526: reading
Rests on: the contributed reading (`contributed/READING.en.md`, reading column; transcription and key output in
`contributed/transcription/` and `contributed/decoded/`), which applies Tomokiyo's key for BnF Cinq Cents de Colbert
398 to an image-adjudicated transcription. This file only puts the reading column into running text for the v3
verification (23 Oct 2026): hyphenated line ends joined; the contributor's marks dropped; plain-letter groups that
`contributed/OPEN.en.md` section 3 lists as nulls dropped; code words whose meaning a contemporary margin decipherment
on the calibration pages confirms written {code: meaning} (saturnus = l'Empereur, Juno = le Roy d'Espagne, my/my3 =
qu'il, my3θ = qu'ilz); other code words stay [code]; letters the contributor supplied in parentheses are bracketed
([?word], part[ial]) so they count as holes; "…" is an unreadable stretch. Spelling as the contributor gives it.
## Reading
par ce[s disc]ours, Monseigneur, il [?est] aisé a connoistre de que lle façon ledit seigneur {saturnus: l'Empereur} desire qu'on interprete son intention … vraysemblable [pur] {my: qu'il} ne voudra avouer directement ce fait devant {Juno: le Roy d'Espagne}, ains plustost s'en excuser, luy persuader s'il peut de n'y avoir jamais pensé ; mais consideré la nature douce du jeune prince, la nourriture d'iceluy, qui a tousjours esté en crainte et [re]verence, tellement que a peine pouvoit il faire un pas hors de sa chambre sans le congé d'iceluy seigneur {saturnus: l'Empereur} son frere ; tant s'en faut qu'a une telle heure il eust peu sortir du chasteau, qui ferme toute la nuit ; pareillement n'ayant oncque eu aucun personnage aupres de luy duquel il eust [mim] de confiance, que de prendre une semblable resolution au desceu de sa magesté imperiale, avec laquelle au contraire il est certain {my3: qu'il} a plus communiqué depuis peu de jours {my: qu'il} n'avoit de coustume ; attendu aussi que le sieur de Malstet, comme depuis il s'est verifié, l'avoit demandé au mesme {saturnus: l'Empereur}, non pour gouverneur, ainsi qu'aucuns disoient, mais entierement pour seigneur desditz pays bas, entraitoit avec sa magesté imperiale, non avec ledit seigneur archiduc [pur]. Je ne voy icy personne qui monstre de douter en sa conscience que telle chose n'aye esté pour pensé d'iceluy seigneur, deliberé et conclue en son conseil, nonobstant toutes les autres conjectures qu'on veuille amener a l'encontre, rien que plusieurs pretendent {my: qu'il} n'aye eu occasion d'avancer en tel cas un frere pour despouiller son oncle, ses neveu[x] ce que il leur appartient davantage. J'entend [au] lieu qu'on a eu assez de peine a inciter pour ce ledit seigneur archiduc, le que l'autrement … trop grande … ; l'on a adjousté aussi que [le]dit seigneur {saturnus: l'Empereur} eust desiré le faire ar[rester] il eust peu depescher lesdusditz gentilzh[ommes] plus matin, comme il ne feut rien ; le temps les ne partirent qu'environ le midy … en donnant soupçon ; sur ce {my: qu'il} escrivit toutes les lettres {my: qu'il} envoya aux princes.Translation
<!-- English translation of texts/p525-526_reading.md (Claude Opus 5.5, 23 Oct 2026), with the same holes. Code words resolved in the reading ({saturnus} = the Emperor, {Juno} = the King of Spain, my/my3 = "that he") are translated; unresolved code words stay [code]; "…" is an unreadable stretch. -->
# Ancel to [the French court], Prague, 22 October 1577, pp. 524 end-526: English translation
By these [?words], Monseigneur, it [?is] easy to recognise in what way the said lord, the Emperor, wishes his
intention to be interpreted … likely [pur] that he will not wish to admit this deed directly before the King of Spain,
but rather to excuse himself, and to persuade him, if he can, that he never thought of it. But considering the gentle
nature of the young prince, and his upbringing, who has always been in fear and [re]verence, so much so that he could
hardly take a step out of his room without the leave of that lord the Emperor his brother — so far was he from being
able to leave the castle, which is shut all night, at such an hour — and likewise never having had anyone about him in
whom he had [mim] confidence, so as to take such a resolution without the knowledge of His Imperial Majesty, with whom
on the contrary it is certain that he has communicated more in the last few days than he used to; considering also
that the Sieur de Malstet, as has since been verified, had asked the Emperor himself for him, not as governor, as
some said, but outright as lord of the said Low Countries, and was negotiating with His Imperial Majesty, not with the
said lord Archduke [pur]. I see no one here who shows any doubt in his conscience that such a thing was not thought
out by that lord, deliberated and concluded in his council, notwithstanding all the other conjectures that one may
wish to bring against it, except that several claim that he had no occasion in such a case to advance one brother in
order to strip his uncle, his nephew[s], of what belongs to them more. I understand [instead] that it took a good
deal of trouble to incite the said lord Archduke to this, the that otherwise … too great …; it is also added that if
the said lord the Emperor had wished to have him stop[ped], he could have despatched the said gentle[men] earlier in
the morning, which was not done; the time, they left only about noon … giving suspicion; whereupon he wrote all the
letters that he sent to the princes.2 Nov 1577, p. 539
The Emperor does all he can to persuade the Spaniards that Matthias's departure happened without his knowledge (contributor's summary).
Transcription
# BnF Cinq Cents de Colbert 398, p.539 (Gallica btv1b525218133 f573), Ancel?, 2 Nov 1577. No contemporary decipherment seen.
# images: gallica/p539_f573_x5y47.jpg, x50y47, x5y60, x50y60 (+ x6y40, x37y40 for preceding clear text)
# preceding clear text: "... Chambellan du sr Archiduc, lequel on dit estre demouré a sa maison, voyant qu'il avoit pareillement {pibdu} {fee} {pemee}"
# Adjudicated by image against an independent Codex transcription (second/p539_pass1.txt), 2026-10-20
# Adjudicated by image a second time, against three further independent Codex readings (second/p539_pass3[abc].txt), 2026-10-20
539.01 x oc? x S | S 4 up w th | di b oc? P | 3 i w x S | i x g w b | {Monseigneur} | plus | {mine}
539.02 {my3} th | S oc fl di | dl 7 x oc? i w dl D | 8 h 4 b g w | 9 pi | S w o pi | 4 x 8 oc? x P | h 4 S D w b
539.03 dl oc? fl oo o S D 7 dl di | plus | {vost} | {saturnus} | {sanc} | x w x i ne q pi | plus | ne oc? dl | 8 b oc? 2 w
539.04 ne w | 8 oc? dl D b 4 i b pi | c | p 4 i D | di oc? x di pi | ne 4 | 9 w 3 oc? dl S D b 4 di i oc? dl | {my3} | ne x 2
539.05 ucl | P oc? hash i o ne w | ros | P w b S x 4 9 w b | ne w S | pi S P 4 g dl oc? ne th | plus | D w ne
539.06 9 w P 4 b D | ucl? | 4 x w dl x | 4 | S oc? dl | 9 pi | S 8 w x | {sec} | {estre} {Car} {ioura} {qu} {il} {est}Decryption: READING.en.md
Decryption
539.01 uOus sauez trOp mieus iuger {Monseigneur} [que] {mine}
539.02 {my3}z sont nauOient charge de sebe aucOup haster
539.03 nOnobstant [que] {vost} {saturnus} {sanc} ueuille [que] lOn crOye
539.04 le cOntraire fait tOute la demOnstratiOn {my3} luy
539.05 [est] pO[ss]ible [pour] persuader les espagnOlz [que] tel
539.06 depart [EST] auenu a sOn de sceu {sec} {estre}{Car}{ioura}{qu}{il}{est}Reading
# Ancel, letter of 2 November 1577, p. 539: reading
Rests on: the contributed reading (`contributed/READING.en.md`, reading column; transcription and key output in
`contributed/transcription/` and `contributed/decoded/`), which applies Tomokiyo's key for BnF Cinq Cents de Colbert
398 to an image-adjudicated transcription. This file only puts the reading column into running text for the v3
verification (23 Oct 2026): hyphenated line ends joined; the contributor's marks dropped; plain-letter groups that
`contributed/OPEN.en.md` section 3 lists as nulls dropped; code words whose meaning a contemporary margin decipherment
on the calibration pages confirms written {code: meaning} (saturnus = l'Empereur, Juno = le Roy d'Espagne, my/my3 =
qu'il, my3θ = qu'ilz); other code words stay [code]; letters the contributor supplied in parentheses are bracketed
([?word], part[ial]) so they count as holes; "…" is an unreadable stretch. Spelling as the contributor gives it.
## Reading
vous savez trop mieux juger, Monseigneur, que [mine] {my3θ: qu'ilz} sont [?] n'avoient charge de se beaucoup haster, nonobstant que … {saturnus: l'Empereur} vueille que l'on croye le contraire ; fait toute la demonstration {my3: qu'il} luy est possible pour persuader les Espagnolz que tel depart est advenu a son desceu.Translation
<!-- English translation of texts/p539_reading.md (Claude Opus 5.5, 23 Oct 2026), with the same holes; [mine] is an unresolved code word, [?] an unread sign. --> # Ancel, p. 539: English translation You know far better how to judge, Monseigneur, that [mine] they are [?] had no instructions to make much haste, although … the Emperor wishes it to be believed otherwise; he makes every show that he can to persuade the Spaniards that this departure happened without his knowledge.
14 Dec 1577, pp. 561-563 read in part
Varambon, sent by Don John, has waited three weeks for leave to levy troops; Borgia says the Emperor must back his brother in earnest, as the Catholic King will defend his right; the Estates will hardly grant extraordinary aid (contributor's summary).
Builds coherent sentences, but under the read bar of 95% of words read as sense (95% here); the holes are marked in the reading.
Transcription
# BnF Cinq Cents de Colbert 398, p.561 (Gallica btv1b525218133 f595), 14 Dec 1577
# NOTE: the flat L-like sign that the first pass wrote as L (pp) reads g in seigneur, Borgia, gendarmerie, magesté; replaced L -> g on this page.
# Transcribed from tiles gallica/tiles/c398_f595_r{1..7}c{1,2}.jpg (symbol transcription only, not deciphered); halves stitched by cross-correlation (c2 offset x=+1229, row pitch about 630 px), overlap dropped. 19 cipher lines (561.01-561.19), then nothing below (r7 = bottom of leaf, bare paper).
# Plain text at top of page was not transcribed in detail (about 7 lines of letter in French, dated 14 Dec 1577); cipher block starts below it, no "# clear:" lines recorded for it (page header region: not legible enough at tile scale).
# Conventions as p525/p526. i (hook, curled-left top) vs dl (tall upright loop) judged on stitched 2x crops; doubtful ones carry "?". "S" = open arc + y sign. "L" = flat/short Lambda-like sign (pp in key; l.09, .14, .17, .19, some doubtful), "at" = circled "au" sign (l.18). "?" alone = cluster not resolved (l.03 end, l.19). Small curl "z?" (l.18) = unmapped ʒ-like shape without the arc. Punctuation dots omitted. Plain-letter words in braces are approximate readings ({manue}, {bel?}, {sec?} uncertain).
# Adjudicated by image against an independent Codex transcription (second/p561_pass1.txt), 2026-10-20 (p/phi and 3/z not changed on shape alone: kept first reading + "?"; deleted token on l.11 idx36)
# Adjudicated by image a second time, against three further independent Codex readings (second/p561_pass3[abc].txt), 2026-10-20
561.01 i w dl 4 2 P w x w dl D pi dl 9 b pi 8 w | {my3} | {sec?} | th 4 x b oc dl D di b 4 i D pi 3 4 i S i 8 w
561.02 ne? x 2 | {manue} | 4 9 i D 7 x dl P w b S oc qq 4 g? w qa x i ne 7 x i S mu? D w ne x 2 P 4 b ne
561.03 7 dl D 9 w ne 4 b 8 h i 9 x 8 3 4 D h i 4 S | {my3} | dl w phi 4 x D P 7 S 9 oc b w S dl 7 x ?
561.04 dl D | plus | {saturnus} | 3 4 b 8 h pi S i phi b 4? i 9 w 3? pi dl D 8 oc dq w i ne 4 8 oc dq 4 dl 8 w
561.05 {my3} | ucl | dl w 8 pi hash 4 i b pi | {my3} | S pi b w S oc ne x w 9 pi | plus | ne P 4 b D 2 i ne n? oc x
561.06 9 b 4 pi S D b w oc x 9 x T? i D | {Juno} | oc x 9 x 9 i D S w i g dl w x b 4 b 8 h i 9 x 8 S i?
561.07 9 x P b w 3? i pi b | {my3} | ne w 3? oc dl S D b w P 4 b ne w S pi qx w D th 9 oc qq 4 dl D P w b 3? i?
561.08 hash i oc dl 9 w ne pi x w b g pi dl S 9 pi g x w b phi? pi P 4 b ne 7 ne w 3? 4 g dl pi 8 oc dq w
561.09 ne pi dl 4 S x Ldot? ne i w ne pi 3 4 b qa? x i th 9 pi x 4 b w 3? o oc dl 4 x dl oc 3? 9 w 9 oc 3? i? w h
561.10 4 dl 9 7 x S D b i 8 h w oc qx b 4 dl D 3 w S phi pi 9 w o 4 i ne q pi b 8 4 x D i oc dl c? S 4
561.11 D i S? P? 4 i b pi w dl D i w b pi 3? w dl D 4 D oc x di? 8 w qa? x i 4 w S D pi oc b 9 oc qq w
561.12 D oc x 8? h 4 dl D ne 7 g? w dl 9 4 b 3? pi b i w pi S 9 w x S 9 pi b dl i w b pi S 9 i w D pi S
561.13 i? 3? P w b i 4 ne pi S D w dl x w S 4 S P i b pi c b 4 D i S o oc dl w 8 pi | plus
561.14 oo? o S w b x 4 dl D ne w 9 i D S w i g? dl pi x b | {saturnus} | dl w ne pi P w x D 9 i b w 8 D pi
561.15 3 w dl D b pi P? x S pi b 4 i dl S i | {my3} | {bel?} | S P w 8 i 4 ne pi 3? pi dl D 8? oc dl D pi dl x w S
561.16 b w 8 pi th? 9 pi S S x S 9 i D w n 9 i pi D w S 8 pi dl oc? dl oo o S D 4 dl D S 4 3 4
561.17 g? w S D pi i? 3? P w b i 4 ne w 4 pi dl D b w D pi dl x i? 8 2 P ne x S 9 w D b oc i S n w
561.18 3 4 i dl? pi S ne w 9 i D 3 4 b al? x i th? S pi b w z? pi box 4 dl? D 4 ne 7 at i x oc? 9 x
561.19 9 i D S i w x b o oo b g? i 7 {roc} ne x 2 9 oc qq pi b phi pi S P oc dl S pi c oc b w STranscription
# BnF Cinq Cents de Colbert 398, p.562 (Gallica btv1b525218133 f596). Faint note in the top margin, about "14 Janv... 1577" (not verified).
# Transcribed from tiles gallica/tiles/c398_f596_r{1..7}c{1,2}.jpg (symbol transcription only, not deciphered). 26 text lines, nothing below l.26 (r7 tiles empty).
# Light mirror-writing and the long swash curves ("(" arcs) are verso bleed-through or flourishes and are ignored. The glyph "y-like with a swash arc" is S (circled s), not a word gap.
# i (dark, short hook) vs dl (tall loop with a high swash) was decided by the tall swash above x-height; "i?" / "dl?" where doubtful. "4" (plain cross) vs "oc" (cross with tapered V foot and left barb) decided by shape and context; "4?" / "oc?" where doubtful.
# qx = "q with a crossbar" (l.10 "7 qx 4 i b w", l.18 "4 qx w 8 D", l.19 "oc qx i 8 i w b"): by context it reads ff (affaire / affection / officier), not l. qq = box-like "q" (nn, l.18 "bonne"). fl = flat-like glyph (dark square with a tall stem, l.23, between i and g). VM? = cross sitting in a cup (l.3, l.18), may be Vostre Majeste. "|" in l.6 and l.24 stands for a full stop (dot) after plain text/cipher.
# Lines run to the right edge and are cut by the binding shadow; the last token is "?" or token? where a glyph is partly visible. Lines 7, 10, 11 wrap hidden text (l.10 "...semble m" + l.11 "nant" = maintenant).
# clear: l.6 ends with the plain word Monseig(neur) (cut at the edge). l.5 ends with the nomenclature letters "ros" (pour), as in p539 l.5. l.8 and l.25 plain "saturnus". l.9 plain "Juno". l.10 and l.18 plain "my". l.20 and l.22 the letters "ucl" (est). l.24 starts with plain "bien experimente." (reading of the first word uncertain) then cipher after the dot.
# Adjudicated by image against an independent Codex transcription (second/p562_pass1.txt), 2026-10-20
# Adjudicated by image a second time, against three further independent Codex readings (second/p562_pass3[abc].txt), 2026-10-20
# The cross-over-box sign written gn in second image adjudication is the double-letter mm sign of the key (dq); renamed dq? (context: "comme").
562.01 ne w S D i w dl D pi dl ne oc dl g x w x b ne pi P ne x S plus ne q pi P w x D?
562.02 D b pi P 4 b D i 8 w ne x 2 o oo b g i 4 7 9 i D plus S i ne w 3 P pi b?
562.03 x D S oc x S D w dl i b S oc dl 4? b pi VM? pi i ne q x 2 S w b 4 o w S oc i dl 9 pi
562.04 p 4 i b w 4 o oo dl pi S 8 i w dl D 9 4 x D 7 dl D plus ne w b oc 2 8 4 D h
562.05 p w b 7 D oc x D w S ne pi S P b oc x i S i oc dl S P oc hash i o ne w S | ros?
562.06 x pi b c 3 4 i dl D w dl i b S oc dl 9 b oc i D | P 4 b 8 w ne 7 | {Monseig}
562.07 x oc x S P oc m w th i x g pi b pi dl qa x w ne th D w b 3 pi S n w D b oc ?
562.08 9 i D S w i g dl pi x b | {saturnus} | ne w qa x pi ne 8 oc? dq? pi 4 x 8 x dl S P b?
562.09 3 w dl D dl 4 x oc i D P oc hash i o ne pi w S D i 3 pi qa x i 8 w ne x 2 | {Juno} | ?
562.10 P b w dl 9 b pi 8 w D 7 ffx 4 i b w P 4 b ne w o oc x D | {my} | S w 3 o ne pi 3?
562.11 dl 4 dl D x oc x ne oc i b p 7 i b pi 4 i dl S P ne x S D oc S di pi S P w b oc i? ?
562.12 x oc 2 7 dl D S oc dl P 4 2 th qa x 4 S i pi dl P b oc 2 w i ne S pi 8 oc dl D?
562.13 i D plus ne 4 b 8 h i 9 x 8 3 4 D h i 4 S 2 p x S D n oc dl g oc m w b?
562.14 c dl w P pi x D oc dl w dl 8 oc b pi S 4 x oc i b 9 pi qa x w ne 8 oc S D w dl?
562.15 D oc x b dl w b 4 8 4 b i ne w dl D b pi D i pi dl D ne pi S w S P 4 g dl oc ne?
562.16 o w ne q pi S P 7 b oc ne q pi S c p 7 i D dl w 4 dl di 3 oc i dl S S pi 8 b pi?
562.17 P ne x S i w x b S P pi D i di pi S 8 h oc S pi S qa x i 9 w S 8 oc m b? ?
562.18 o oc qq w 4 ffx w 8 D i oc dl | {my} | 4 9 pi p 7 x oc b i S w b n oc dl p b w VM? w 8 ?
562.19 S pi P b pi S w dl D w qa x pi ne plus oc ffx i 8 i w b qa x i ne w x pi x ?
562.20 7 ne q w b D b oc m w b i ne | ucl | dl 8 oc dl D i dl 4 dl D b pi 3 x dl w b pi 9 w ?
562.21 8 w S P 4 hash w th c oc dl ne x 2 p oc x b dl i D oc x D b pi 8 w ne 7 4 b | ucl?
562.22 S oc dl x oc 2 4 g w 8 pi qa x i | ucl | 4 9 x w dl x 4 P ne x S i w x b S c ?
562.23 4 x D b w S dl 4 g x w b pi S 7 x dl i fl? g w dl i w x b dl w 7 P oc ne i D 4 dl?
562.24 {bien} {experimente?} | 8 w S i oc x b S 8 2 P 4 b w i ne q pi 3 w dl D i ne 4 ?
562.25 8 h 4 dl D pi w dl ne 7 8 h 4 3 o b w 9 x 9 i D n w i g dl w x b | {saturnus} | x?
562.26 S oc dl 4 x qa w S 4 x 7 dl D 7 g w 9 pi S w n P 4 g dl oc ne th 3? pi ne 7 qa x ?Transcription
# BnF Cinq Cents de Colbert 398, p.563 (Gallica btv1b525218133 f597)
# Transcribed from tiles gallica/tiles/c398_f597_r{1..7}c{1,2}.jpg (re-cut at 2x from gallica/full/c398_f597.jpg). Symbol transcription only, not deciphered; 15 cipher lines.
# i (hook) vs dl (tall delta) judged at 2x crops; doubtful readings carry "?". Bare "?" = glyph with no key entry: stem + upper-right loop + lower-left bowl (563.01, .06, .10; context says r) and a three-bar stem with a top hook (563.05; context says e). fl? in .13 = heavy b-shaped glyph. {x} = small two-stroke plain x; c = small one-stroke plain c (null). "S" is the open-arc + y sign; the thin vertical line crossing it is the tail of a long s above.
# clear (before the block): none; the cipher starts at the top of the page.
# Adjudicated by image against an independent Codex transcription (second/p563_pass1.txt), 2026-10-20
563.01 ne w 9 i D S w i g dl pi x b | {saturnus} | 3? oc dl S D b 7 ? w 8 pi x oc i b x dl S i dl g x ne i w
563.02 b P ne 7 i S i b c D oc S di 4 P b w S ne pi dl x oc 2 4 7 ne 4 b 8 h i? 9? x 8 w b dl pi S D w
563.03 ros | ne x 2 p? 7 i b w 7 x oc i b P 4 b D 9 x P 4 hash w D pi 3? S | {Je} {Vous} {say?}
563.04 8 w S 3? pi dl x pi S P 4 b D i 8 x ne 7 b i D w th | {Monseigneur} | 4 box w dl 9 oc dl D | plus
563.05 i w S 4 8 h w 4 hash w x b ne? 3? pi dl D qa x pi ne q al? S pi b 7 ne 4 9? w b dl i w b pi
563.06 ? pi S oc ne x D i oc dl 9? x 9 i D S w i g dl w x b | {saturnus} | ne w qa x pi ne 7 ne 4 x w b i
563.07 D w 8 oc dq? w i x g w dl D P ne x S i w x b S n 4 g w S P w b S oc qq 7 g w S dl 4
563.08 7 x 8 x dl w 3? pi dl D o w S oc i dl 9? pi 3? 4 m 4 i S w i dl D w ne q i g w dl 8 pi c 3? oc i dl S
563.09 9 pi g x w at pi 7 x w 8 ne w 9 i D S pi i g dl w x b | {Juno} | dl pi S D 7 dl D di? b oc P
563.10 o i w dl P oc x b x w x 9? w oc i dl oc dl 8 pi S c w S D 7 dl D 9? i ffx i 8 mu ne w 9 4 box i b w? VM?
563.11 ne w S pi S D 7 di th 9? pi ne w 3? P i b pi 7 | plus | ne | plus | S w 8 oc x b S pi {x?} D b 4
563.12 oc b 9? i dl 4 i b w x pi x 3? w S phi w qa? x oc dl fl w P w x D o oo D w dl i b 9 w x S ne w x
563.13 b S D 7 {x?} pi S oc b 9? i dl 4 i b pi S n i 8 w dl {ucl} P oc hash i o ne w ne pi S w ne pi 8
563.14 D w x b S 9? w S 4 {x?} w c o b 4 dl 9 pi o oc x b g ne w S qa? x pi ne th D oc? x D w S oc oc? i S
563.15 dl w S pi x oc x 9? b oc dl D 9? w o pi 4 x 8 oc x P 4 ffx? oc dl o ne i b
# clear: De Hongrie nous n'avons rien entendu depuis mais debonne? et ne se parle-
# Adjudicated by image a second time, against three further independent Codex readings (second/p563_pass3[abc].txt), 2026-10-20
# The cross-over-box sign written gn in second image adjudication is the double-letter mm sign of the key (dq); renamed dq? (context: "comme").Decryption: READING.en.md
Decryption
561.01 ienaypeuentendrece {my3} {sec?} zauronttraitemaisice
561.02 Luy {manue} aditaunperso[nn]aGequilauisIteluyparl
561.03 antdelarchiducmathias {my3} nemautpasdoresnau·
561.04 nt [que] {saturnus} marchesimrAideMentco[mm]eilaco[mm]ance
561.05 {my3} [est] nece[ss]aire {my3} seresoluede [que] lpartyilSou
561.06 draestreouduDit {Juno} oududitseigneurarchiducsI
561.07 dupreMier {my3} leMonstreparleseletzdo[nn]antperMI
561.08 [ss]iondeleuergensdeguerMeparlaleMagneco[mm]e
561.09 lenasu[PP]lielemarQuizdeuareMbonaunoMdedoMIeh
561.10 andaustricheolrantmesmedebaillercautionsa
561.11 tiSPaireentiereMentatouTceQuiaesteordo[nn]e
561.12 touChantlaGendarMerieesdeusdernieresdietes
561.13 IMperialestenuesaspireratisbonece [que]
561.14 ObseruantleditseiGneur {saturnus} nelepeutdirecte
561.15 mentrePuserainsi {my3} {bel?} specialeMentContenues
561.16 receZdessusditesdietescenOnobstantsama
561.17 GesteIMperialeaentretenuIcyplusdetroisse
561.18 maiNesleditmarAuiZsereRe[tt]aNtala[rr]iuOdu
561.19 ditsieurborGia{roc}luydo[nn]ermesponseoresDecryption
562.01 lestientenlongueurleplus[que]llepeuT
562.02 treparticeluyborgiaadit[que]silempeR
562.03 utsoustenirsonAre[VOSTRE MAJESTE]eilluyserabesoinde
562.04 faireabonescientdautant[que]leroycath
562.05 feratouteslesprouisionspo[ss]ibles [POUR]
562.06 uermaintenirsondroit parcela {Monseig}
562.07 uouspo[uu]eziugerenquelztermessetro·
562.08 ditseigneur {saturnus} lequelcO[MM]eaucunspR
562.09 mentnauoitpo[ss]ibleestimequiceluy {Juno} ·
562.10 prendreceta[ff]aireparlebout {my} sembleM
562.11 nantuouloirfaireainsplustostesperoI·
562.12 uoyantsonpayzquasienproyeilseconT
562.13 it[que]larchiducmathiasyfustsongo[uu]eR
562.14 nepeutonencoresauoirdequelcosteN
562.15 tourneracarilentretientlesespagnoL
562.16 bellesparollesfaitneantmoinssecrE
562.17 plusieurspetiteschosesquidesco[uu]R·
562.18 bo[nn]ea[ff]ection {my} adefauorisersonfre[VOSTRE MAJESTE]ec·
562.19 sepresentequel[que]o[ff]icierquileueu·
562.20 allertro[uu]eril [est] ncontinantremunerede·
562.21 cespa[ss]ezonluyfournitoutrecelaar [EST]
562.22 sonuoyagecequi [est] aduenuaplusieurs·
562.23 autresnagueresauniNgenieurneapolitaN
562.24 {bien}{experimente?} cesiourscypareillementila·
562.25 chanteenlachambreduditseigneur {saturnus} U
562.26 sonauqesauantagedesespagnolzMelaqu·Decryption
563.01 leditseigneur {saturnus} Monstra·eceuoirunsingulie
563.02 rplaisirtostapreslenuoyaalarchIDucerneste
563.03 [pour] luyFaireauoirpartdupa[ss]eteMs {Je}{Vous}{say?}
563.04 cesMenuesparticularitez {Monseigneur} a[tt]endont [que]
563.05 iesachea[ss]eurLMentquellAseralaDerniere
563.06 ·esolutionDuditseigneur {saturnus} lequelalaueri
563.07 teco[MM]eiugentplusieurssagesperso[nn]agesna
563.08 aucuneMentbesoinDeMa[uu]aiseintelligenceMoins
563.09 degue[rr]eauecleditseigneur {Juno} nestantTrop
563.10 bienpourueuDeoinoncesestantDi[ff]icileda[tt]irE[VOSTRE MAJESTE]
563.11 lesestatzDeleMpirea [que] l [que] secourse{x?}tra
563.12 orDinaireueuMesmeQuonnepeutbotenirdeusleu
563.13 rsta{x?}esorDinairessicen{ucl}po[ss]ibleleselec
563.14 teursDesa{x?}ebrandebourglesQuelztOutesoOis
563.15 neseuouDrontDebeaucoupa[FF]onblirReading
# Ancel, letter of 14 December 1577, pp. 561-563: reading
Rests on: the contributed reading (`contributed/READING.en.md`, reading column; transcription and key output in
`contributed/transcription/` and `contributed/decoded/`), which applies Tomokiyo's key for BnF Cinq Cents de Colbert
398 to an image-adjudicated transcription. This file only puts the reading column into running text for the v3
verification (23 Oct 2026): hyphenated line ends joined; the contributor's marks dropped; plain-letter groups that
`contributed/OPEN.en.md` section 3 lists as nulls dropped; code words whose meaning a contemporary margin decipherment
on the calibration pages confirms written {code: meaning} (saturnus = l'Empereur, Juno = le Roy d'Espagne, my/my3 =
qu'il, my3θ = qu'ilz); other code words stay [code]; letters the contributor supplied in parentheses are bracketed
([?word], part[ial]) so they count as holes; "…" is an unreadable stretch. Spelling as the contributor gives it.
## Reading
Je n'ay peu entendre ce {my3: qu'il} … traité, mais iceluy [manue] a dit a un personnage qu'il a visité, luy parlant de l'archiduc Mathias, {my3: qu'il} ne faut pas doresnav[ant] …nt que {saturnus: l'Empereur} marche si froidement comme il a commencé ; {my3: qu'il} est necessaire {my3: qu'il} se resolve de quel party il voudra estre, ou du dit {Juno: le Roy d'Espagne} ou du dit seigneur archiduc ; si du premier, {my3: qu'il} le monstre par les effetz, donnant permission de lever gens de guerre par l'Alemagne, comme l'en a supplié le marquiz de Varembon au nom de dom Jehan d'Austriche, offrant mesme de bailler caution satisfaire entierement a tout ce qui a esté ordonné touchant la gendarmerie es deux dernieres dietes imperiales tenues a Spire [et] Ratisbone ; ce que observant, ledit seigneur {saturnus: l'Empereur} ne le peut directement refuser, ainsi {my3: qu'il} [bel] specialement contenues […] esdessusdites dietes. Ce nonobstant sa magesté imperiale a entretenu icy plus de trois semaines ledit marquiz, se remettant a l'arrivée du dit sieur Borgia luy donner response … les tient en longueur le plus qu'il peut. [D'au]tre part iceluy Borgia a dit que si l'Empereur veut soustenir son frere, il luy sera besoin de faire a bon escient, d'autant que le roy cath[olique] fera toutes les provisions possibles pour [conser]ver/maintenir son droit. Par cela, Monseig[neur], vous pouvez juger en quelz termes se tro[uve] [le]dit seigneur {saturnus: l'Empereur}, lequel, comme aucuns pr[e]tendent, n'avoit possible estimé qu'iceluy {Juno: le Roy d'Espagne} prendroit cest affaire par le bout {my: qu'il} semble m[ain]tenant vouloir faire, ains plustost esperoi[t], voyant son payz quasi en proye, il se cont[entero]it que l'archiduc Mathias y fust son gouver[neur]. [On] ne peut on encores savoir de quel costé [il] tournera, car il entretient les Espagnol[z] [de] belles parolles, fait neantmoins secre[tement] plusieurs petites choses qui descouvr[ent] [la] bonne affection {my: qu'il} a defavoriser son fre[re]… se presente quelque officier qui le veu[ille] aller trouver, il est incontinant remuneré de ces passez, on luy fournit outre cela ar[gent]… son voyage, ce qui est advenu a plusieurs autres nagueres, a un ingenieur neapolitan bien experimenté ; ces jours cy pareillement il a chanté en la chambre dudit seigneur {saturnus: l'Empereur} … [une chan]son au desavantage des Espagnolz … ledit seigneur {saturnus: l'Empereur} monstra recevoir un singulier plaisir, tost apres l'envoya a l'archiduc Erneste pour luy faire avoir part du passetemps. Je vous say ces menues particularitez, Monseigneur, attendant que je sache asseurement quelle sera la derniere resolution dudit seigneur {saturnus: l'Empereur}, lequel a la verité, comme jugent plusieurs sages personnages, n'a aucunement besoin de mauvaise intelligence, moins de guerre avec ledit seigneur {Juno: le Roy d'Espagne}, n'estant trop bien pourveu de finances, estant dificile d'attirer les estatz de l'empire a quelque secours extraordinaire, veu mesme qu'on ne peut obtenir deux leurs taxes ordinaires, si ce n'est possible les Electeurs de Saxe, Brandebourg, lesquelz toutesfois ne se voudront de beaucoup affoiblir.Translation
<!-- English translation of texts/p561-563_reading.md (Claude Opus 5.5, 23 Oct 2026), with the same holes; bracketed part-words completed in the reading are translated as words; unresolved code words stay [code]; "…" is an unreadable stretch. --> # Ancel, pp. 561-563: English translation I have not been able to learn what he … negotiated, but this [manue] said to a person he visited, speaking to him of Archduke Matthias, that from now on it is not right …that the Emperor should proceed as coldly as he has begun; that it is necessary that he decide which party he will be of, either of the said King of Spain or of the said lord Archduke; if of the first, that he show it by deeds, giving permission to raise soldiers throughout Germany, as the Marquis of Varambon has begged him in the name of Don John of Austria, even offering to give security to comply fully with everything that was ordained concerning men-at-arms in the last two imperial diets held at Speyer [and] Regensburg; which being observed, the said lord the Emperor cannot refuse it directly, since he [bel] specially contained […] in the aforesaid diets. Notwithstanding this, His Imperial Majesty has kept the said Marquis here more than three weeks, putting off giving him an answer until the arrival of the said Sieur Borgia … he keeps them waiting as long as he can. [On the] other hand, this Borgia has said that if the Emperor means to support his brother, he will have to do it in earnest, since the Catholic King will make every possible provision to [pre]serve/maintain his right. From this, Monseigneur, you can judge in what position the said lord the Emperor finds himself, who, as some claim, had perhaps not thought that the King of Spain would take this affair by the end that he now seems to wish to, but rather hoped that, seeing his country almost a prey, he would be content to have Archduke Matthias as his governor there. [One] cannot yet know which way [he] will turn, for he entertains the Spaniards with fine words, yet secretly does several little things that reveal the good will he has to favour his brother… if some officer presents himself who wishes to go and join him, he is at once rewarded for his past services, he is furnished besides with [money]… his journey, which has lately happened to several others, to a very experienced Neapolitan engineer; these last days likewise he sang in the said lord the Emperor's chamber … [a so]ng to the disadvantage of the Spaniards … the said lord the Emperor showed that he took singular pleasure in it, and soon after sent him to Archduke Ernest to let him share the entertainment. I tell you these small particulars, Monseigneur, while waiting to know for certain what will be the final resolution of the said lord the Emperor, who in truth, as several wise persons judge, has no need at all of ill understanding, still less of war, with the said lord the King of Spain, not being too well provided with finances, and it being difficult to draw the estates of the Empire into any extraordinary aid, seeing that one cannot even obtain their ordinary taxes from two of them, except perhaps the Electors of Saxony and Brandenburg, who however will not want to weaken themselves much.
21 Dec 1577, p. 565
Varambon is back from Archduke Charles's court with nothing; the Emperor waits for the King of Spain's [Juno] answer (contributor's summary).
Transcription
# BnF Cinq Cents de Colbert 398, p.565 (Gallica btv1b525218133 f599), 21 Dec 1577
# Transcribed from tiles gallica/tiles/c398_f599_r{1..7}c{1,2}.jpg (re-cut at 2x from gallica/full/c398_f599.jpg). Symbol transcription only, not deciphered; 16 cipher lines (565.01 is only the tail of the plain opening line).
# "?" alone = glyph with no key entry (stem + upper-right loop + lower-left bowl); token+"?" = uncertain reading. {nez?} = a short plain word (read "nez"/"mes"), {manua?} = plain word that recurs in 565.02 and 565.14. fl = the b-shaped glyph with a box/pointed foot (Nflat, n). qa = the "a with curled foot" glyph (read as q: marquiz, lesquelz, mesme qu'on). "S" is the open-arc + y sign.
# clear (before the block): "21. decemb. 1577." / "Monseigneur, Vous aurez veu par ma decembre? du xxm.e? de ce moys"
# Adjudicated by image against an independent Codex transcription (second/p565_pass1.txt), 2026-10-20
# Adjudicated by image a second time, against three further independent Codex readings (second/p565_pass3[abc].txt), 2026-10-20
565.01 8 4? oc w
565.02 | {fec} | {saturnus} | D i w dl di w dl ne oc dl g x pi x b ne w P ne x S | {my} | P w x D | {mais?} | ne | {manua?}
565.03 9 w S P 4 g dl pi dl w x oc n ne 7 dl D n i | plus | P b oc 3? P D w 3? pi dl D n w 9 pi 8 ne 7 b w b
565.04 S x b ne 7 P w b 3? i hash i oc dl 9 pi S P? w x ffx S 9 pi g w dl S T w | guod? | ros? | ne 7 qa? x
565.05 pi ne q w n oc ne q i 8 mu D w b P b i dl 8 i P 4 ne w 3? pi dl D ne w 3 4 b qa? x i th 9 w x 7 b w 3? o
565.06 oc dl 7 w n D pi 9 pi P w S 8 h w P 4 b 9 w 8 4 P 4 b 9 oc 3? i w h 4 dl 9 4 x S D b i 8 h w
565.07 3 4 i dl D w dl 7 dl D w S D 7 dl D b w D oc x b fl w i 8 w ne x 2 3? 4 b qa? x i th 9 pi ne 4 8 oc x b D
565.08 9 pi ne 7 b 8 h i 9 x 8 6 h 4 b ne w S oc x i ne w S D oc i di 7 ne q w dl 7 g x w b pi S
565.09 i w P x i S o i w dl x oc x n 4 hash w x b pi b | {mesme?} | th dl oc dl D pi dl g 4? b w b i w dl
565.10 oo o D w dl x dl 2 3? w S phi? pi q w x b 7 w S D pi 9 oc qq w 4 x 8 n dl w i dl D w dl D i oc dl
565.11 9 oc o D w dl i b T 7 x 4 dl D 4 g w 8 2 4 P b w S n w b pi 3? w box 7 dl D | {saturnus}
565.12 7 ne 4 b w S P oc dl S w | {my} | 7 box w dl q 9 x | {Juno} | S x b 8 pi | {my} | ne x 2 7 b w
565.13 S 8 b i di 9 pi P x i S ne w P 4 b D w 3? pi dl D 9 w ne 4 b 8 h i 9 x 8 3? 4 D h i 4 n
565.14 o i w dl qa? x oc dl 4 hash w x b pi | plus | ne w S x n 9 i D | {manua?} | P w x D 9 w S i 7
565.15 7 x oc i b pi x | {(ainsi} {quoy} {m'a} {dit)} | 4 x i dl? 9 x 9 i D S w i g dl pi x b VM oc? 2 S 4 dl
565.16 3 4 i S D b w 9 pi 8 w | {my} | 9 oc i D di b 4 i D w b | {fec} | {vib?}
# clear: Mardy dernier arriva icy le sr de la Croix envoye par monsieur le ducDecryption: READING.en.md
Decryption
565.01 cAoe
565.02 {fec} {saturnus} tientenlongueurleplus {my} peut {mais?} l {manua?}
565.03 despagneneuoslantsi [que] proMpteMentsedeclarer
565.04 surlaperMi[ss]iondesPeu[ff]sdegensde [GUERRE] [POUR] laQu
565.05 ellesolliciterprincipaleMentlemarQuizdeuareMb
565.06 onaestedepeschepardecapardoMiehandaustriche
565.07 maintenantestantretourneiceluyMarQuizdelacourt
565.08 delarchiduccharlesouilestoitallenagueres
565.09 iepuisbienuousa[ss]eurer {mesme?} znontengArerien
565.10 obtenunyMesMeleuraestedo[nn]eaucsneintention
565.11 dobtenirdauantagecyapressereMe[tt]ant {saturnus}
565.12 alaresponse {my} a[tt]enldu {Juno} surce {my} luyare
565.13 scritdepuisleparteMentdelarchiducMathias
565.14 bienQuona[ss]eure [que] lesusdit {manua?} peutdesia
565.15 auoireu {(ainsi}{quoy}{m'a}{dit)} auiNduditseigneur[Vostre Majeste]Oysan
565.16 maistredece {my} doittraiter {fec} {vib?}Reading
# Ancel, letter of 21 December 1577, p. 565: reading
Rests on: the contributed reading (`contributed/READING.en.md`, reading column; transcription and key output in
`contributed/transcription/` and `contributed/decoded/`), which applies Tomokiyo's key for BnF Cinq Cents de Colbert
398 to an image-adjudicated transcription. This file only puts the reading column into running text for the v3
verification (23 Oct 2026): hyphenated line ends joined; the contributor's marks dropped; plain-letter groups that
`contributed/OPEN.en.md` section 3 lists as nulls dropped; code words whose meaning a contemporary margin decipherment
on the calibration pages confirms written {code: meaning} (saturnus = l'Empereur, Juno = le Roy d'Espagne, my/my3 =
qu'il, my3θ = qu'ilz); other code words stay [code]; letters the contributor supplied in parentheses are bracketed
([?word], part[ial]) so they count as holes; "…" is an unreadable stretch. Spelling as the contributor gives it.
## Reading
…cane {saturnus: l'Empereur} tient en longueur le plus {my: qu'il} peut, mais l[e] [manua] d'Espagne ne voulant si promptement se declarer sur la permission des levées de gens de guerre, pour laquelle solliciter principalement le marquiz de Varembon a esté depesché par deça par dom Jehan d'Austriche ; maintenant estant retourné iceluy marquiz de la court de l'archiduc Charles, ou il estoit allé nagueres, je puis bien vous asseurer [mesme] … n'a encore rien obtenu, ny mesme leur a esté donné aucune intention d'obtenir davantage cy apres, se remettant {saturnus: l'Empereur} a la response {my: qu'il} attend du {Juno: le Roy d'Espagne} sur ce {my: qu'il} luy a rescrit depuis le partement de l'archiduc Mathias, bien qu'on asseure que le susdit [manua] peut desja avoir eu [ainsi quoy m'a dit] avis dudit seigneur … maistre de ce {my: qu'il} doit traiterTranslation
<!-- English translation of texts/p565_reading.md (Claude Opus 5.5, 23 Oct 2026), with the same holes; [manua] and [mesme] stay as in the reading; the bracketed "ainsi quoy m'a dit" is the contributor's uncertain run, translated in brackets. --> # Ancel, p. 565: English translation …cane the Emperor keeps it waiting as long as he can, but the [manua] of Spain not wishing to declare himself so promptly on the permission for the levies of soldiers, to solicit which the Marquis of Varambon was chiefly sent here by Don John of Austria; now that this Marquis has returned from the court of Archduke Charles, where he had lately gone, I can well assure you [mesme] … has still obtained nothing, nor has any intention even been given them of obtaining more hereafter, the Emperor deferring to the answer he awaits from the King of Spain on what he wrote to him after the departure of Archduke Matthias, although it is asserted that the aforesaid [manua] may already have had [as he told me] word from the said lord … master of what he is to negotiate
Key: key.tsv
06 Sources
- BnF, Cinq Cents de Colbert 398, Gallica btv1b525218133: p. 523 view 557, p. 525 view 559, p. 539 view 573, pp. 561–563 views 595–597, p. 565 view 599.
- Harumi-Matsuba, vulcob1575-public (CC BY 4.0 data, MIT code), and issue #53, 20 October 2026.
- S. Tomokiyo, “French Ciphers for German (and Eastern) Affairs”, Cryptiana, 15 March 2026: the key.
- Prior work found by the contributor: G. Labouchère, “Guillaume Ancel, envoyé résident en Allemagne”, Revue d’histoire diplomatique 37 (1923), the clear text only; D. Fontvieille on Vulcob’s cipher correspondence (Heidelberg, April 2024), working from the margin decipherments.
The contributor’s transcription, key, reading and calibration files, the re-measure and the notes are in targets/ancel1577/.