Adrian of Utrecht to Charles V (1520) — Key from external plaintext

At a glanceKey from external plaintext · complete

The letter: Adrian of Utrecht, Cardinal of Tortosa and governor of Castile for Charles V (sixteen months later Pope Adrian VI), writing from Valladolid on 9 September 1520. He signs “Car.l dertusen”.
Where: Archivo General de Simancas, Patronato Real, PTR,LEG,2,1,19, f. 1r (PARES 12987122). Danvila printed the letter in Memorial histórico español XXXVI (1897) without the cipher paragraph: “no ha podido traducirse”.
The cipher: The Cardinal’s “dotted trigraph” code: consonant-vowel-consonant groups followed by a dot stand for words and syllables (gig. que, leb. no, sa. de), with a homophonic alphabet of signs. It differs from the code of the joint letter of December 1521, which has its own page.
How it was deciphered: Key K7 was rebuilt from cipher and plain pairs of the Cardinal’s other letters of 1520–21. One measurement on a fresh pair named before download (PTR,LEG,2,1,99, 5 December 1521) gave 0.882 of characters against at best 0.176 for 200 shuffled keys, and 0.770 of words.
What it says: The Junta of the Comunidades means to gather all the alcabalas and all the towns and places that had recently been the Crown’s, from the grandees, and it moves not only against Charles but against the great lords.
Still open: The first cipher word (six signs) and the group fud. are unresolved. The images were not seen here.
Credit: Brice Dudout rebuilt the key, re-read doubtful signs blind, ran the validation and read the paragraph, working with Claude Opus 5.5 agents.

01 The letter

In August and September 1520 the Junta of the Comunidades, meeting at Ávila, seized Tordesillas (29 August) and Queen Joanna. Adrian, at Valladolid, tried in vain to draw it to the town; the mission of the commander Hinestrosa is told in plain text in the letter. Before the cipher paragraph comes the plain sentence “el fin que sobresto tienen sabrá V. Al. por la cifra que va dentro desta”, so an enclosed cipher sheet existed. Its contemporary decipherment survives separately (PTR,LEG,2,1,107) and is not the text of this paragraph.

Almost all the Cardinal’s cipher letters of 1520–21 are printed deciphered by Danvila from the contemporary transcriptions. Still without plain text: this paragraph and a struck-out cipher paragraph of 21 October 1520.

02 The key and its test

K7 was built from the pairs PTR,LEG,2,1,23 and 32 (Bergenroth, Calendar of State Papers, Spain, Supplement, 1868), 2,1,30, 61 and 88 (Danvila), and 2,1,68, 70, 71, 73 and 85 (contemporary transcriptions 2,1,69, 72 and 83). A code group is kept only when attested twice. Sign shapes were re-read blind on three plates, which split confused homophones. Earlier keys failed the criterion fixed in advance: K3 (0.851 of characters but 0.541 of words on 2,1,68), K4 (0.783) and K6 (0.813); their pairs then went into training. Only K7 was frozen before its test pair was named.

03 The reading

The paragraph, with probable words in braces and restored words in brackets: “… esto, dará … a estos [tales], que no {solamente} se mueven {contra} [vuestra] majestad, {mas} aun contra todos los grandes, de los cuales {entienden} {primeramente} [juntar] todas las {alcabalas}, y {después} todas las [villas] {y} lugares que de {poco} {tiempo} en {fueron} de la {corona} real.”

Abolishing or reducing the alcabalas and returning to the Crown the towns alienated to lords were among the Comuneros’ demands. Of 46 cipher words, 28 are certain, 12 probable, 4 restored and 2 unresolved: 44 of 46, or 95.7%, so the 95% bar is met narrowly. The translation justifies every probable, restored or unresolved word. The full Spanish text and an English translation are under Full texts.

04 What remains uncertain

The first cipher word (six signs, one uninventoried shape) and the group fud. are unresolved.
Signs still outside the key: the first signs of [tales] and [juntar], the second sign of [villas], and three more.
Twelve probable words rest on one sign or group attested once.
The transcription was made by AI and the sign shapes re-read by the contributor from plates; no person checked them against the images here.

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.

Adrian of Utrecht to Charles V, Valladolid, 9 September 1520 (AGS PTR,LEG,2,1,19), cipher paragraph

On 9 September 1520 Adrian of Utrecht – Cardinal of Tortosa, regent of Castile for Charles V, and sixteen months later Pope Adrian VI – wrote from Valladolid while the Junta of the Comunidades was settling at Ávila. In the middle of the letter are six lines in cipher. Danvila, who printed the letter in 1897, noted only: “Sigue un párrafo cifrado que no ha podido traducirse”. The cipher is the Cardinal’s dotted code; we rebuilt it from his other letters of 1520–1521 whose plain text is known, validated it on a letter held out, and applied it to this paragraph. It tells what the Junta meant to do to the grandees: take from them the alcabalas and the towns that had belonged to the Crown.

Transcription

transcription.txt

# AGS PTR,LEG,2,1,19, f. 1r, the six cipher lines after "por la cifra que va dentro desta | quanto mas".
# Sign labels re-read on the image against a sign-shape atlas (Oct-Nov 2026): g9 (9 = a) vs gr (ꞡ = r); Fx, Qx, Zx = shapes
# not inventoried elsewhere (first word, l. 4 verb, "‹?›ales"); dho = ð with o (= u, cf. 2,1,61 "levantado").
# One token = one sign or one dotted code group (groups end with "."). ASCII sign labels:
# Étiquettes ASCII des signes (lecture sur images PARES ≈1200 px, agrandies ×2) : X. = groupe de code pointé tel qu'écrit (gig. leb. …) ; signes : tt (double t / H barré), F (f bouclé), m, mo (m+o), th (θ / ϑ), 4, oo, g (g/9), qb (q bouclé ꝗ), q, qx (q à croix dessous), cf, Yo (y + o souscrit), Z (ʒ/z), Zp (z à point/trait au-dessus), Bx (B barré), ch (c♯), h (ɧ), p, ga (g♯), dh (ð), oto, M, & , eo (e+o), ro (⊥ + o dessous), r (⊥), ao (a+o), ai (a♯), L (ʆ / 6 renversé), n, ni (n♯), al (α), 3, 7, H (ꟗ / d), e, y, b, c, x, 2, tri (▽), 8, inf (∞), div (÷), hash (#), dho (ð+o)
L1: tt Fx m mo th 4 roc. oo g9 gr cf fud. 4 roc. Yo Zx Bx ch h p
L2: gig. leb. ded. mud. di. qb ga dh oto M & vuh. eo ro 4 qx m. mu. nah. yae.
L3: vuh. q phi g9 cab. Yo nof. gag. su. so. p gof. raf. zu. & ao phi ai L dh r 4
L4: mud. Qx dho Z q 4 gr ceb. Yo nad. cf ch Zp g9 ga 4 ni Bx Yo al su. gad.
L5: ceb. qx nad. oto ti b g9 p n neg. raf. gig. sa. hag. cu. zi. qec. r hash & so. g9
L6: dh div phi hash & 4 fed.

Decryption

decryption.txt

# Line by line: transcription, then the mechanical decryption with key K7 (key_K7.json, SHA-256 8ca16a1d...).
# _x_ = code group; ‹x› = sign or group not in K7 or ambiguous; "1x:" = value attested once.

L1: tt Fx m mo th 4 roc. oo g9 gr cf fud. 4 roc. Yo Zx Bx ch h p
	p ‹Fx› u i c a _esto_ d a r a _‹fud.›_ a _esto_ s ‹Zx› a l e s

L2: gig. leb. ded. mud. di. qb ga dh oto M & vuh. eo ro 4 qx m. mu. nah. yae.
	_que_ _no_ _1x:sola_ _mente_ _se_ m u e u e n _con_ t ‹ro› a ‹qx› _‹m.›_ _magestad_ _1x:mas_ _ahun_

L3: vuh. q phi g9 cab. Yo nof. gag. su. so. p gof. raf. zu. & ao phi ai L dh r 4
	_con_ t r a _todo_ s _los_ _gran_ _des_ _del_ s _qual_ _es_ _1x:entende_ n p r i ‹L:r/t› e r a

L4: mud. Qx dho Z q 4 gr ceb. Yo nad. cf ch Zp g9 ga 4 ni Bx Yo al su. gad.
	_mente_ ‹Qx› 1x:u n t a r _toda_ s _las_ a l ‹Zp› a u a l a s y _des_ _1x:pues_

L5: ceb. qx nad. oto ti b g9 p n neg. raf. gig. sa. hag. cu. zi. qec. r hash & so. g9
	_toda_ ‹qx› _las_ u ‹ti› z a s 1x:y _lugar_ _es_ _que_ _de_ _1x:poco_ _1x:tiempo_ _en_ _1x:fue_ r o n _del_ a

L6: dh div phi hash & 4 fed.
	e o r o n a _real_

Reading

reading.md

# Established text (Spanish)

{x} probable; [x] restored from the sense; ‹…› unresolved. Key K7 (validated 2 Nov 2026).

Context in plain text: En Plazencia tanbien han quitado al corregidor la vara […] y dixieronle que si muchas mas recibiera por vra. Mat. todas ie las tomaran | el fin que sobresto tienen sabra v. al. por la cifra que va dentro desta | quanto mas

Cipher: ‹…› esto, dara ‹…› a estos [tales], que no {solamente} se mueuen {contra} [vuestra] magestad, {mas} ahun contra todos los grandes, dels quales {entienden} {primeramente} [juntar] todas las {alcavalas}, y {despues} todas las [villas] {y} lugares que de {poco} {tiempo} en {fueron} de la {corona} real.

Translation

translation.md

# English translation

“… and the more ‹…› this, the more ‹…› it will give to these [men], who rise not {only} {against} [Your] Majesty {but} also against all the grandees, from whom they {intend} {first} to [gather] all the {alcabalas}, {and then} all the [towns] {and} places which {lately} {were} of the royal {Crown}.”

## Justification of the probable, restored and unresolved words

- dara (certain): o̥o = d, 9 = a, ꞡ = r, cf = a, all firm values of key K7; “esto dará”.
- {solamente}: ded. = “sola”, a single attestation, followed by mud. = -mente (certain).
- {contra} (l. 2): vuh. = con and ε̥ = t certain, then a ⊥ with o below that is not in the key, and 4 = a: “cont‹?›a”.
- {mas}: nah. = mas, a single attestation.
- {entienden}: zu. = “entende”, a single attestation (PTR,LEG,2,1,73, “da allá en que entender”), + & = n.
- {primeramente}: p r i ‹?› e r a + mud. (-mente); the 4th sign (ʆ) is r or t in the key, not m.
- {alcavalas}: 8 of 9 signs have firm values; the 3rd (z with a stroke above) is not attested for c.
- {despues}: su. = des (firm), then gad. = pues, a single attestation.
- {y} (l. 5): the sign n = y has a single attestation.
- {poco} {tiempo}: hag. and cu., one attestation each, in the letter of 15 March 1521 (PTR,LEG,2,1,68) and its contemporary transcription (2,1,69): “hay pocos o ningunos que en este tiempo quieran comprar”.
- {fueron}: qec. = “fue”, a single attestation (2,1,73, “ha dias que fue a Burgos”), + r, o, n.
- {corona}: 5 of 6 signs certain; the key gives e for the 1st (ð), read c (shape close to θ = c); sense fixed by “real”.
- [tales]: “a estos ‹?›ales”; -ales certain, “estos tales” a common phrase; the first sign is an uninventoried shape.
- [vuestra]: “contra ‹q m.› mu.”; mu. = magestad (certain), elsewhere always preceded by a “vuestra” group (cim.); the two signs before it are outside the key.
- [juntar]: key K7 reads “‹?› u n t a r” (ðo = u, ʒ = n, q = t, 4 = a, ꞡ = r); the 1st sign is an uninventoried shape (Brice Dudout’s re-reading: “an odd g”). The restoration [quitar] of the first version rested on a wrong label (the last sign is ꞡ = r, not ꝗ; the 2nd is ðo = u) and does not fit “‹?› u n t a r”: it is dropped.
- [villas]: the key reads “u ‹?› z a s”: the 2nd sign is unknown and b is z in K7, not ll; “villas y lugares” imposes the word. Marked as a restoration, never counted certain.
- Unresolved: the first word (6 signs, p ‹?› u i c a; “publica” possible, not shown) and the group fud., unattested. The phrase “de poco tiempo en fueron” is awkward; the group zi. = en is certain.

Key: key_K7.json

05 Sources

AGS, Patronato Real, PTR,LEG,2,1,19, PARES 12987122 (not reproduced; no open licence stated).
A. Danvila, in Memorial histórico español XXXV–XXXVIII (1897–98); G. A. Bergenroth, Calendar of State Papers, Spain, Supplement (1868).
Brice Dudout, Europe chiffrée, Tortosa 1520.

Working files: targets/adrian1520/ in the repository.

Adrian of Utrecht to Charles V, 9 September 1520