At a glanceKnown key applied · partial (96% of the cipher)
- The letter
- Cardinal Marino Ascanio Caracciolo, governor of Milan, to Charles V, 14 November 1537, in Italian.
- Where
- AGS Estado, ff. 110r–111r; DECODE catalogued it as not deciphered. The four DECODE images are out of archive order.
- The cipher
- Cifrario 38: an alphabet of invented signs with homophones, a syllabary using superscript vowel digits, nulls and a few code words.
- How it was read
- Wanruo Luo reconstructed Cifrario 38 in her 2021 thesis, and Satoshi Tomokiyo noted it fits. Applied unchanged it decrypts all eleven cipher lines and the cancelled line. His four lines are confirmed with two corrections.
- What it says
- The imperial agent Marnol’s dealings with the Swiss and the Grisons as the Dauphin crosses into Italy; the envoy Panizone; and Marnol now seeing “la malignita et perfidia” of the Swiss.
- Still open
- About ten signs stay doubtful, none where they change the sense.
- Credit
- The key is Wanruo Luo’s; Satoshi Tomokiyo first noted it fits and read four lines.
01 The letter and the key
The four DECODE images are out of archive order. P4 is f. 110r (scan 0408) and opens “Sacra Cat.ca & Ces.a M.ta”. It is in clear, on the Dauphin and the Grand Master crossing with munitions, the King at Briançon, and “Mons.or di Marnol” with the Colonel of the Isola working to recall the Swiss in French service. P1 is f. 110v (scan 0409), with clear text on the Grisons and then the eleven cipher lines. P3 is f. 111r (scan 0410). It has one cipher line struck through, then the clear close, dated “In M.lo ali xiiij de Nov.e M. D. xxxvij” and signed “il Car.l Caracciolo”. P2 is the address leaf, with the docket “del Cardenal Caracciolo xiiij de Nov.e 1537”.
Cifrario 38 (Luo 2021, p. 657, from AGS Estado leg. 38 doc. 190) has three parts.
- An alphabet of invented signs with homophones. a has six signs, and e, i, l, n, o, p and u have two to four each.
- A syllabary. A consonant’s base sign takes a superscript digit for the vowel: 5 = a, 6 = e, 7 = i, 8 = o, 9 = u. For example, 2 is the base for s, so 2⁷ = si.
- Nulls, marked by the superscripts 2, 3 and 4, and a few code words.
The first cipher line shows the system at work: 2⁷ a̱o⁸ 3⁶ + 9 ṉo⁵ θ 4⁷ a+⁵ ṉo⁶ = si do ve a p ra c ti ca re.

02 The decipherment
Line by line, as the cipher spells it. [..] marks signs not resolved, and square brackets around words mark a doubtful decryption.
1 solo si dovea practicare per Marno a se volevano concordarse
2 como erano Suizari, et in tal caso ho scripto piu vol-
3 te a Marno a che manda il Panizone in mio nome et pagato
4 quale a fe? resolui, perche io serria? contento
5 dare a epsi Grisoni como a[..] Elvetii la tracta
6 del frumento gratis, facendo capituli convenienti.
7 Fin qui a epso Marno a non e parso [..] dicti Griso-
8 ni; harriano 'l facto como [..] Suizari. Epso Marno a
9 se accorge hora de la malignita et perfidia de
10 Suizari et [..] et soi ministri. Se deve ne recor-
11 dare, e quante volte io ha dicto che epsi Suizari far-
(f. 111r, cancelled) riano quello hora fan po[..]
In substance, Caracciolo reports on Marnol’s work with the Grisons. Marnol was to find out whether they would come to terms as the Swiss had. Caracciolo had written to him several times to send Panizone in Caracciolo’s name and to pay him. The Grisons were to be given, as the Swiss had been, the tratta of grain free of duty, on suitable articles. So far it had not seemed good to them, and they would have done as the Swiss did. Marnol now saw the malice and bad faith of the Swiss and their ministers. Caracciolo reminds the Emperor how often he had said what the Swiss would do “hora”. The last words are in the line the writer cancelled.
03 Prior work and method
Tomokiyo’s page on the Spanish ciphers of Charles V, in its section on Luo’s thesis, names the key and gives a rough decipherment of lines 1–4 (“… pagabo quale e a e f ressolvi perche io se r ri a conbento”). That note was found before the work and led to the key. Here the letter was deciphered line by line from the DECODE images, and the key needed no change. Line 3 gives pagato and line 4 contento, where Tomokiyo had pagabo and conbento. Luo’s thesis does not edit this letter.
04 The second pass, and what remains uncertain
The first pass left about fifteen percent open. A second pass at 4× with contrast on every line settled most of it with four rules:
- A superscript belongs to the sign before it. In line 5, 3⁶ 4⁷ et decrypts as ve ti i, so the word is “Elvetii”.
- A bare ♀ is the letter r, and ♀ with a superscript is the p-syllable. This gives “harriano” in line 8 and “parso” in line 7.
- The g-shaped sign with a bar is an f homophone, which gives “frumento” in line 6.
- Dotted signs are nulls.
About ten of roughly 330 signs (an estimate, not a sign-by-sign count) stay doubtful:
- Line 4: “a fe”, decrypted letter by letter, and “serria”, where the sign taken as n stands where an a is expected.
- Line 5: two a-homophones after “como a”.
- Lines 7 and 8: one sign each.
- Line 10: “2 a” after “et”.
- The cancelled line: one syllable after “po”.
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.
R9966, Milan 14 Nov 1537 read in part
Cardinal Caracciolo, governor at Milan, to Charles V: Marno(l), the imperial agent with the Swiss, was to treat with the Grisons only if they would come to terms as the Swiss had; Caracciolo would give them the duty-free grain export gratis under fitting articles, and Marno now sees the malice and perfidy of the Swiss and their ministers.
Builds coherent sentences, but under the read bar of 95% of words read as sense (94% here); the holes are marked in the reading.
Transcription
# Copied verbatim from NOTES.md ("The cipher and the key"), 22 Oct 2026. Only cipher line 1 was ever transcribed sign by
# sign; lines 2-11 and the cancelled line were read straight from the images (reading.txt) and have no sign transcription.
# A superscript digit belongs to the sign before it (5 = a, 6 = e, 7 = i, 8 = o, 9 = u).
Line 1: 2⁷ a̲o⁸ 3⁶ + 9 n̲o⁵ θ 4⁷ a+⁵ n̲o⁶ ♀⁶ ♀
o̲⁵ ♀ ꝑ⁸ a 2⁶ 3⁸ P⁶ 3⁵ ꝑ⁸ a+⁸ t a+⁸ ♀ a̲o⁵ ♀ 2⁶ = si dove a practicare per Marno a se volevano concordarse.Decryption
Cardinal Marino Ascanio Caracciolo (Milan) to Charles V, 14 November 1537 AGS Estado leg. 1184 fol. 110 (DECODE R9966), cipher passage on AGS scan 0409 (DECODE image P1), lines 1-11, plus one cancelled line at the top of scan 0410. Key: Luo 2021 Cifrario 38 (Charles V - Caracciolo, 1530-1538). Read line by line from the images; second pass at 4x with autocontrast on every line. [..] = signs not resolved; word? = doubtful reading. A superscript digit belongs to the sign before it. (clear) ... perche no hano alcuna pretensione co questo stato. [struck: Ne hano grela alcuna] (cipher) 1 solo si dovea practicare per Marno a se volevano concordarse 2 como erano Suizari, et in tal caso ho scripto piu vol- 3 te a Marno a che manda il Panizone in mio nome et pagato 4 quale a fe? resolui, perche io serria? contento 5 dare a epsi Grisoni como a[..] Elvetii la tracta 6 del frumento gratis, facendo capituli convenienti. 7 Fin qui a epso Marno a non e parso [..] dicti Griso- 8 ni; harriano 'l facto como [..] Suizari. Epso Marno a 9 se accorge hora de la malignita et perfidia de 10 Suizari et [..] et soi ministri. Se deve ne recor- 11 dare, e quante volte io ha dicto che epsi Suizari far- (cancelled line, f. 111r) [null] riano quello hora fan po[..] [nulls] (clear) Il partir de la S.ra Duchessa fu a tempo ... Unread or doubtful, by sign: line 4 'a fe' (+ 9 hooked-9, literal, sense doubtful) and 'serria' (the sign read as t = n stands where 'a' is expected); line 5 the two a-homophones after 'como a' (bottom-T and T); line 7 one bare r-sign before 'dicti'; line 8 one bare 2 before 'Suizari'; line 10 '2 a' after 'et'; the cancelled line one syllable after 'po'. About 10 of roughly 330 cipher signs.
Reading
<!-- Reading of DECODE R9966 (Claude Opus 5.5, 23 Oct 2026): the cipher passage in reading.txt (Luo 2021 Cifrario 38, read line by line from the images) put into sentences with the clear words around it (in ⟨ ⟩): line numbers dropped, line-end splits joined (vol-te, Griso-ni, recor-dare, far-riano), punctuation as in reading.txt with one stop added. Spelling as the key gives it. [..] = signs not resolved; a doubtful word ("fe?", "serria?") is [?word]; "Marno a" is the writer's name for Marnol as the key gives it. The cancelled line at the top of f. 111r is given at the end. -->
# Cardinal Marino Ascanio Caracciolo to Charles V, Milan, 14 November 1537 (AGS Estado leg. 1184 fol. 110)
## Reading
⟨... perche no hano alcuna pretensione co questo stato.⟩ Solo si dovea practicare per Marno a se volevano concordarse
como erano Suizari, et in tal caso ho scripto piu volte a Marno a che manda il Panizone in mio nome et pagato quale a
[?fe] resolui, perche io [?serria] contento dare a epsi Grisoni como a[..] Elvetii la tracta del frumento gratis,
facendo capituli convenienti. Fin qui a epso Marno a non e parso [..] dicti Grisoni; harriano 'l facto como [..]
Suizari. Epso Marno a se accorge hora de la malignita et perfidia de Suizari et [..] et soi ministri. Se deve ne
recordare, e quante volte io ha dicto che epsi Suizari farriano quello hora fan po[..] ⟨Il partir de la S.ra Duchessa
fu a tempo ...⟩Translation
<!-- English translation of texts/R9966_reading.md (Claude Opus 5.5, 23 Oct 2026), with the same holes; clear words in ⟨ ⟩. --> # Cardinal Marino Ascanio Caracciolo to Charles V, Milan, 14 November 1537: English translation ⟨... because they have no claim on this state.⟩ It was only to be negotiated through Marno[l] whether they wished to come to terms as the Swiss had; and in that case I have written several times to Marnol that he should send Panizone in my name, and paid, which [?] I resolved, because I [?would be] content to give these Grisons, as [to] the [..] Swiss, the export of grain free, making suitable articles. Up to now it has not seemed [good] to the said Marnol [..] the said Grisons; they would have done it as [..] the Swiss. The said Marnol now realises the malice and perfidy of the Swiss and [..] and their ministers. He should remember how many times I have said that these Swiss would do what they are now doing po[..] ⟨The departure of the Lady Duchess was in time ...⟩
05 Sources
- Archivo General de Simancas, Estado leg. 1184 fol. 110, scans 0408–0411: DECODE R9966 (login).
- Wanruo Luo, El lenguaje cifrado de Isabel de Portugal (1530–1539), doctoral thesis, Universitat de València, 2021 (RODERIC), Cifrario 38, p. 657.
- Satoshi Tomokiyo, cryptiana.web.fc2.com, “Spanish ciphers during the reign of Charles V”, section on Luo (2021).
Notes, reading and profile: targets/caracciolo1537/ in the repository.
