At a glanceCiphertext-only break · complete
- The documents
- The copybook of lieutenant colonel Bernardo Taulet, commanding the Jalet (Falset) canton and the Priorato columns during the January 1866 rising: his daily reports to the military government of Tarragona, with a cipher telegram received from Reus on 7 March 1866; and four leaves of the letters he received in January 1866.
- Where
- Archivo Histórico Nacional, Madrid, DIVERSOS-COLECCIONES,193,N.99 (PARES 5928106) and N.94 (PARES 5855640). Catalogue entry 2444. No DECODE record and no edition were found.
- The cipher
- Each letter is replaced by a two-digit number between 10 and 92, with several numbers for the common letters and no word divider. The cipher runs stand inside clear Spanish sentences.
- How it was deciphered
- Brice Dudout’s agents rebuilt the key from the runs of the reports, using the clear words around them, and tested it twice under rules fixed in advance. The key read the held-out runs, but the margin over shuffled keys (0.277, then 0.261) stayed under his 0.30 rule, so he calls the key partial.
- What it says
- Taulet meets the brother-in-law of the rebel Saguete and waits to be told where to go; the Minister of War orders that “no se haga ninguna concesión a Escoda” and that Escoda and Sagueta be taken at all costs.
- Still open
- The key is not validated by the contributor’s rule, and two numbers seen only in N.94 (12 = y, 78 = u) are untested. Nineteen words are probable: one number corrected by sense, or a number not in the key read by sense. The project did not check the transcription against the images.
- Credit
- Transcription, key, tests, reading and translation by Brice Dudout (Europe chiffrée, with Claude Opus 5.5 agents), pull request #376 of 10 November 2026. The catalogue entry came from Daniel Bourdeau’s sweep of PARES.
01 The documents
Unit N.99 of Diversos-Colecciones 193 in the Archivo Histórico Nacional holds 54 images: orders received from the military government of Tarragona (views 2–13 and 44–50), the copybook of the daily reports nos. 1–67 of lieutenant colonel Bernardo Taulet (views 14–45, 14 January to 2 March 1866), and a cipher telegram received from Reus on 7 March 1866 (views 51–53). Taulet commanded the Jalet (Falset) canton and the columns of the Priorato during the January 1866 rising. In report 13 (24 January, view 21) he cannot read a dispatch “por carecer de la clave para descifrar” and asks for the key; numbers appear in his own reports from report 18 (27 January). No decipherment is written on any leaf.
Unit N.94 (PARES 5855640) holds letters received by Taulet in January 1866. The contributor used four of its views (43, 87, 78 and 26) as new material to re-test the key; they had not been seen while the key was built.
No image is shown on this page. The contributor included no image, crop or thumbnail, and the PARES images were not reachable from the session that integrated the pull request. A run of the cipher is printed in numbers in the strip under the title.
02 The cipher
A homophonic substitution of letters by 35 two-digit numbers: a 21 62, b 14, c 81 82 87, d 61, e 11 33 84, f 25, g 19 60, h 70, i 41 92, j 56, l 13 18, ll 16, m 38, n 29, ñ 15, o 36 64, p 72, r 27 77, s 10 50 80, t 88, u 32 55, v 20. There is no word divider, and small words are often left in clear. N.99 has 33 cipher runs of 370 numbers: S01–S30 in reports 18 to 55 (views 25–42) and S31–S33 in the telegram (views 52–53).
The short run at the end of each report, just before “Dios guarde”, is the day of the week. Fifteen of the sixteen agree with the 1866 calendar; the first, S01 in report 18 of Saturday 27 January, reads domingo, one day ahead, which the contributor explains as a report closed after midnight.
03 How it was deciphered
The key. The contributor’s agents transcribed the 33 runs and froze the transcription by SHA-256. The key was built by sense on S01–S32 only, from the clear context of each report and the repeated end-of-report runs read as days of the week, and frozen in turn.
First test. Under a protocol written and hashed beforehand, the last run, S33 (51 numbers, the end of the telegram), was decoded once with the frozen key. It reads a Cambrils con su primer batallón dé cuenta de haberse alojado. The measure is the share of letters covered by Spanish words: 0.981, against 0.704 for the best of 200 shuffled keys, a margin of 0.277. The rule fixed in advance asked for 0.85 and a margin of 0.30, so the key was declared partial.
Re-test on new material. On 10 November 2026 the same frozen key and script were measured once on the cipher of N.94 views 43, 87, 78 and 26 (325 numbers in 17 runs), pre-registered before any number was transcribed: 0.960 against 0.699, a margin of 0.261, under the rule again. Two numbers, 12 and 78, are not in the key and were scored as uncovered.
The contributor notes that the shuffled keys cover about 70% of the letters whatever the length (0.704 on 51 numbers, 0.699 on 325), so the 0.30 margin is close to unreachable for this system with this measure; he reports the rule as written. The project re-ran both measurements from the frozen files, with identical figures, and decoded the transcription again with the key, with the same result as the contributor’s decryption.
The extent. Every cipher word reads as Spanish in its sentence, so more than 95% of the cipher gives sense and the extent is complete. The contributor submitted it as partial because the key failed his rule; the site records the state of the key apart, as partial.
04 The reading
Counts by cipher word, from the contributor’s reading files. A word is certain when the key gives every letter, probable when one number had to be corrected by sense or is not in the key.
| Document | Words | Certain | Probable | Unresolved |
|---|---|---|---|---|
| N.99, reports and telegram | 76 | 71 | 5 | 0 |
| N.94, letters received | 68 | 54 | 14 | 0 |
| Total | 144 | 125 | 19 | 0 |
Report 21 (about 30 January 1866). Clear words in angle brackets, a probable word in braces:
ahora ⟨que⟩ son las ocho de la noche se me presenta el cuñado de Saguete con el pliego ⟨de V.E.⟩ fecha ⟨de ayer⟩ hemos ⟨quedando⟩ entendi{d}os ⟨y espero⟩ su aviso ⟨para⟩ acudir al sitio ⟨que⟩ me diga.
Now that it is eight in the evening, Saguete’s brother-in-law comes to me with Your Excellency’s dispatch dated yesterday; we have come to an understanding, and I await his word to go to the place he tells me.
In report 22 (31 January) most people say that “Sagueta n{o} está ⟨en el⟩ país” and that many believe he went to Madrid or Barcelona to obtain the pardon. The telegram from Reus of 7 March orders Taulet to march with his first battalion to Cambrils and report when quartered. In N.94 the Minister of War, through the Captain General, orders that no concession be made to Escoda, that he be judged by the laws if taken, and that “a toda costa sean aprendidos no solo Escoda {y} Sagueta sino los demás dispersos”; another leaf gives troop figures: a strength of 242, drummers included, from which each of three columns can have only 81. Saguete or Sagueta, a rebel of the rising, is not identified in the sources.
05 What remains uncertain
- The key is not validated by the contributor’s rule (margins 0.277 and 0.261 against 0.30). Both held-out portions are used up, and N.99 holds no other unseen cipher; a further test needs fresh cipher of the same command and a null other than lexicon coverage.
- Two values, 12 = y (five times) and 78 = u (once), were derived from N.94 after the measurement and are untested (
key_derived_n94.tsv). - Nineteen probable words. In N.99: entendi{d}os (81 for 61), n{o} (15 for o), sába{d}o (“sabaos”), domi{n}go (a doubtful 21 or 29) and mar{c}he (21 for c). In N.94: orde{n}, con{o}cimiento, {y} (four times), {q}ue, j{u}zgue, le{y}es, igualmen{t}e, reci{b}ido, {v}einte, o{c}ho and cu{m}plimentará.
- Declared deviations: some values of the learning runs were seen by eye before the transcription was frozen, and clear words of N.94 views 43 and 87 were seen after the key was frozen (
validation.txt). - One AI transcription, from the PARES images. The project did not check it against the images, and did not survey the other views of N.94 for more cipher.
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.
reports 18-55 (27 Jan - 14 Feb 1866) and telegram from Reus (7 Mar 1866): cipher runs in clear reports
Taulet, hunting the rebels of the January 1866 rising in the Priorato, reports that the brother-in-law of the rebel Saguete has brought him a dispatch and that they have come to an understanding; most people say Sagueta has gone to Madrid or Barcelona to obtain a pardon. The end-of-report runs give the day of the week. A telegram of 7 Mar 1866 orders Taulet to march to Cambrils with his first battalion and report when quartered.
Transcription
# taulet1866 - transcription of the cipher runs (AI transcription from the PARES images, 9 Nov 2026; byte-identical data lines to frozen/transcription.txt, SHA-256 0c211566...). # Format: run<TAB>T|H<TAB>numbers. n? = doubtful; a|b = variant reading (a kept). T = learning set (S01-S32); H = held-out run S33 (measured once). # Images: AHN DIVERSOS-COLECCIONES,193,N.99, PARES 5928106 (views 25-42: reports; views 52-53: telegram). No image reproduced here. S01 T 61.64.38.92.29.60.64 S02 T 61.64.38.92.29.60.64 S03 T 21.70.64.27.21 S04 T 10.64.29.13.21.10.64.81.70.64.61.11.13.21.29.64.81.70.11.10.11.38.11.72.27.11.10.11.29.88.21.11.13.81.55.15.21.61.64.61.11.10.21.19.55.11.88.11.81.64.29.11.13.72.13.92.11.60.64 S05 T 25.11.81.70.21 S06 T 70.11.38.64.10 S07 T 11.29.88.11.29.61.92.81.64.10 S08 T 10.55.21.20.92.10.64 S09 T 21.81.55.61.92.27.21.13.10.92.88.92.64 S10 T 38.11.61.92.60.21 S11 T 38.21.27.88.11.10 S12 T 38.21.10.21.25.92.27.38.21.29 S13 T 50.62.19.55.11.88.21.29.15.84.10.88.21 S14 T 72.21.41.50 S15 T 38.55.82|87.70.36.10.61.11.84.16.64.10 S16 T 21.38.62.61.77.41.61.36.14.21.27.81.11.13.64.29.21 S17 T 64.14.88.11.29.84.27.11.13.92.29.61.55.13.88.64 S18 T 38.92.11.27.81.64.13.11.10 S19 T 56.55.33?.20.84.10 S20 T 20.92.11.27.29.11.10 S21 T 10.21.14.21.64.10 S22 T 61.64.38.92.21|29.60.36 S23 T 13|18.55.29.11.80 S24 T 38.21.27.88.11.10 S25 T 56.55.11.20.84.10 S26 T 20.92.11.27.29.84.10 S27 T 10.21.14.62.61.64 S28 T 13.55.29.11.10 S29 T 38?.21.27.88?.11.10 S30 T 38.92.11.27.81.64.13.84.10 S31 T 21.13.81.64.27.36?.29.11.13.88.62.32?.13.84.88 S32 T 38.21.15.62.29.21.29.55.11.20.84.38.21.27.21.70.11 S33 H 21.81.62.38.14|16.27|22.41.13.10.81?.36.29.50.55.72.77.41.38.84.27.14|16.62.88.21.16|14.64.29.61.84.81.55.11.29.88.62.61.11.70.62.14|16.11.27.50.84.21.13.64.56|36.62.61.36
Decryption
# taulet1866 - line-by-line decryption with key.tsv (raw output of frozen/decode_T.py; '?' would mark a number not in the key: none occurs).
# The variant a of a|b is used. No correction here; the reading (reading.md) marks the 5 probable words in { }.
S01 T domingo
S02 T domingo
S03 T ahora
S04 T sonlasochodelanochesemepresentaelcuñadodesagueteconelpliego
S05 T fecha
S06 T hemos
S07 T entendicos
S08 T suaviso
S09 T acudiralsitio
S10 T mediga
S11 T martes
S12 T masafirman
S13 T saguetanñesta
S14 T pais
S15 T muchosdeellos
S16 T amadridobarcelona
S17 T obtenerelindulto
S18 T miercoles
S19 T jueves
S20 T viernes
S21 T sabaos
S22 T domiago
S23 T lunes
S24 T martes
S25 T jueves
S26 T viernes
S27 T sabado
S28 T lunes
S29 T martes
S30 T miercoles
S31 T alcoroneltaulet
S32 T mañananuevemarahe
S33 H acambrilsconsuprimerbatallondecuentadehabersealojadoReading
# taulet1866: reading of the cipher runs (Spanish), 9 Nov 2026
Copybook of colonel Bernardo Taulet (Regimiento de León), commanding the Jalet (Falset) canton and the Priorato columns,
Jan-Mar 1866. AHN, DIVERSOS-COLECCIONES, 193, N.99 (PARES 5928106).
Marks: clear words of the copybook are given only as context in ⟨ ⟩ (not counted). In the cipher: {x} = probable (a
one-number slip of the cipherer or copyist, corrected by sense; the justification follows each passage); [x] = restored;
‹…› = unresolved. There are no restored and no unresolved words in this unit.
Classes (Daniel's rule). **Certain** = every letter given by the key, applied without change. **Probable** = one number
corrected by sense. The key itself is **partial** by our own protocol (held-out margin 0.277, under our 0.30 rule; see
validation.txt), so we classify the unit "read in part" although every cipher word is read by sense.
## Counts (cipher words)
| passage | certain | probable | restored | unresolved |
|---|---|---|---|---|
| report 21 (S03-S10, view 26) | 26 | 1 | 0 | 0 |
| report 22 (S12-S17, view 27) | 15 | 1 | 0 | 0 |
| day words at the end of the reports (S01, S02, S11, S18-S30) | 14 | 2 | 0 | 0 |
| telegram from Reus, 7 Mar 1866 (S31-S33, views 52-53) | 16 | 1 | 0 | 0 |
| **total, 76 words** | **71** | **5** | **0** | **0** |
Read share: certain only 71/76 = **0.934**; certain + probable (Daniel's rule) 76/76 = 1.000. We keep "read in part"
because the key fell short of our validation rule, not because words are open.
The five probable words: entendi{d}os (S07), n{o} (S13), sába{d}o (S21), domi{n}go (S22), mar{c}he (S32).
## Report 21 (view 26, Jalet, about 30 Jan 1866): S03-S10
ahora ⟨que⟩ son las ocho de la noche se me presenta el cuñado de Saguete con el pliego ⟨de V.E.⟩ fecha ⟨de ayer⟩ hemos
⟨quedando⟩ entendi{d}os ⟨y espero⟩ su aviso ⟨para⟩ acudir al sitio ⟨que⟩ me diga.
- entendi{d}os: the cipher gives 81 (= c) "entendicos" where 61 (= d) is required.
## Report 22 (view 27, Jalet, 31 Jan 1866): S12-S17
⟨Los⟩ más afirman ⟨que⟩ Sagueta n{o} está ⟨en el⟩ país ⟨creyendo⟩ muchos de ellos ⟨marchó⟩ a Madrid o Barcelona ⟨con
objeto de⟩ obtener el indulto.
- n{o}: the cipher gives 15 (= ñ, as in cuñado, mañana) "nñ" where an o (36/64) is required; "no está en el país" is
imposed by the sense ("most claim that Sagueta is not in the country, many believing he went to Madrid or Barcelona").
## Day words at the end of the reports (just before "Dios guarde")
Each short run is the day of the week of the report, and all agree with the 1866 calendar: a check word.
| run | report (view), date | reading |
|---|---|---|
| S01 | 18 (25), 27 Jan (Saturday) | domingo (one day ahead: the report was perhaps closed after midnight) |
| S02 | 19 (25), 28 Jan (Sunday) | domingo |
| S11 | end of report, 30 Jan (Tuesday) (27) | martes |
| S18 | 22 (27), 31 Jan (Wednesday) | miércoles |
| S19 | 23 (27), 1 Feb (Thursday) | jueves |
| S20 | 24 (28), 2 Feb (Friday) | viernes |
| S21 | 30 (30), 3 Feb (Saturday) | sába{d}o |
| S22 | 31 (30), 4 Feb (Sunday) | domi{n}go |
| S23 | 33 (31), 5 Feb (Monday) | lunes |
| S24 | 34 (32), 6 Feb (Tuesday) | martes |
| S25 | 37 (34), 8 Feb (Thursday) | jueves |
| S26 | 39 (34), 9 Feb (Friday) | viernes |
| S27 | 41 (35), 10 Feb (Saturday) | sábado |
| S28 | 50 (40), 12 Feb (Monday) | lunes |
| S29 | 51 (40), 13 Feb (Tuesday) | martes |
| S30 | 55 (42), 14 Feb (Wednesday) | miércoles |
- sába{d}o (S21): written 10-21-14-21-64-10 "sabaos", ending 64-10 where 61-64 (d-o) is required; corrected by the day
(3 Feb 1866 was a Saturday; S27 has the regular form 10-21-14-62-61-64 sábado).
- domi{n}go (S22): the fifth number is doubtful in the image (21|29). The reading kept in the frozen transcription, 21,
gives "domiago"; the variant 29 gives "domingo", as do the day (4 Feb 1866, Sunday) and S01/S02. Counted probable
because the retained reading needs the correction.
## Telegram (views 52-53; Reus, 7 Mar 1866, forwarded to colonel Taulet)
⟨El Excmo. Sor. General Comandante General de esta división, en telegrama que acabo de recibir ahora que son las 10 y
cuarto de la noche, me dice = "Traslade V.S.⟩ al coronel Taulet ⟨por parejas de Guardia Civil, lo siguiente =⟩ mañana
nueve mar{c}he ⟨V.S.⟩ a Cambrils con su primer batallón, dé cuenta de haberse alojado." ⟨Dios gue. a V.S. m. a. Reus 7
Marzo de 1866.⟩
- mar{c}he (S32): the cipher gives 21 (= a) "marahe" where a c (81/82) is required.
- "mañana nueve": either "tomorrow at nine" or "tomorrow, the ninth"; left as written.
- S33 ("a Cambrils con su primer batallón dé cuenta de haberse alojado", 11 words) is the held-out run: it was decoded
only once, by the frozen key, in the single measurement. All 11 words are certain.
- A clear telegram of the same day (view 51, Tarragona, general Manso de Zúñiga) orders Taulet's battalion to Vilaseca,
"donde quedará alojado": a neighbouring order with another destination. It was not used for the key or the measurement.Translation
# taulet1866: English translation of the cipher runs (with the clear context), 9 Nov 2026
Clear text of the copybook in ⟨ ⟩; probable words (one number corrected by sense) in { }. See reading.md.
- Report 21 (about 30 Jan 1866): Now ⟨that⟩ it is eight in the evening, Saguete's brother-in-law comes to me with the
dispatch ⟨of Your Excellency⟩ dated ⟨yesterday⟩; we have ⟨come to an⟩ {understanding}, ⟨and I await⟩ his word ⟨to⟩ go
to the place ⟨that⟩ he tells me.
- Report 22 (31 Jan 1866): ⟨The⟩ greater part claim ⟨that⟩ Sagueta is {not} ⟨in the⟩ country, many of them ⟨believing
that he went⟩ to Madrid or Barcelona ⟨in order to⟩ obtain the pardon (indulto).
- End of the daily reports, 27 Jan - 14 Feb 1866: the day of the week (Sunday, Sunday, Tuesday, Wednesday, Thursday,
Friday, {Saturday}, {Sunday}, Monday, Tuesday, Thursday, Friday, Saturday, Monday, Tuesday, Wednesday).
- Telegram (Reus, 7 Mar 1866): ⟨His Excellency the General commanding this division, in a telegram I have just received,
now at a quarter past ten at night, tells me: "Forward⟩ to colonel Taulet ⟨by pairs of the Civil Guard the following:⟩
tomorrow at nine (or: tomorrow, the ninth) ⟨Your Lordship⟩ will {march} to Cambrils with your first battalion, and
report that you are quartered there."letters and telegrams received, Jan 1866 (views 43, 87, 78, 26): the re-test portion, cipher runs in clear letters
A letter from Albarca (28 Jan 1866) promises to carry out a superior order and report the result, in answer to a dispatch received at 9 in the evening. A telegram of the Minister of War, forwarded by the Captain General, orders that no concession be made to the rebel Escoda, that he be judged according to the laws if arrested, that the prisoners be judged too, and that not only Escoda and Sagueta but all the other scattered men be arrested. An acknowledgement of the 28th, and a strength of 242 men, drummers included, so that each of three columns can only have 81.
Transcription
# taulet1866 - N.94: transcription of the cipher runs of the re-test portion H (AI transcription from the PARES images, 10 Nov 2026; data lines byte-identical to retest/transcription.txt, SHA-256 934efc1c...). # Format: run<TAB>H<TAB>numbers. n? = doubtful. All 17 runs are H (held-out, measured once with the frozen key). Comment lines in [ ] give the clear words written between runs. # Images: AHN DIVERSOS-COLECCIONES,193,N.94, PARES 5855640, views 43, 87, 78, 26. No image reproduced here. # view 43 (letter from Albarca, 28 Jan 1866; signer read as Rafael Ramis, uncertain) R01 H 11.29.88.11.27.21.61.64.92.10.11 # [cumplimentará] R02 H 10.55.10.55.72.11.27.92.64.27.64.27.61.11.27 # [quedando en] R03 H 61.21.27.21 # [V.S.] R04 H 81.64.29.61.81.92.38.92.11.29.88.64.61.11.13 # [inmediato] R05 H 27.11.10.55.13.88.21.61.64 R06 H 11.10.81.55.21.29.88.64 # [debo decir a V.S.] R07 H 11.29.81.64.29.88.11.10.88.21.81.92.64.29.21.10.55.64.25.92.81.92.64 # [de este día] R08 H 27.11.81.92.14.92.61.64 # [a las 9] R09 H 61.11.13.21.29.64.81.70.11 # view 87 (telegram of the Minister of War, forwarded by the Captain General) R10 H 29.64.10.11.70.21.60.62.29.41.29.60.55.29.21.81.64.29.81.11.10.92.64.29.21.84.10.81.36.61.62.12.19.55.11.50.84.56.78.10.60.55.11.81.36.29.21.27.77.84?.60.13.36.62.13.21.10.13.11.12.84.50.10.41.25.55.84.27.62.21.72.27.84.29.61.41.61.64 # [Los prisioneros] R11 H 10.11.27.21.29.92.60.55.62.13.38.11.29.38.84 # [juzgados] R12 H 72.27.64.81.55.27.11 # [V. E. ? que a toda costa] R13 H 10.11.21.29.21.72.27.11.29.61.41.61.36.50.29.64.10.36.13.64.11.10.81.64.61.62.12.10.21.19.55.11.88.62.10.92.29.64.13.36.50.61.11.38.21.10.61.41.50.72.84.27.10.64.50 # view 78 R14 H 27.11.81.92.61.92.61.64.11.13.55.11.92.29.88.11 # [y] R15 H 64.70.64.10.11.81.55.29.72?.13.92.38.11.29.10?.88.21.27.21 # view 26 (added by the pre-registered rule: views 43+87+78 give 292 numbers, under 300) # [y siendo esta la de] R16 H 61.64.10.81.92.11.29.88.64.10.81.55.21.27.11.29.88.21.12.61.64.10 # [incluzos los Tambores, solo podrá componerse cada una de estas tres columnas de] R17 H 64.81.70.11.29.88.21.12.55.29.64
Decryption
# taulet1866 - N.94: line-by-line decryption of the 17 runs of H with the frozen key (frozen/KEY.tsv, unchanged; retest/decode.txt, SHA-256 11341d6d...).
# '?' = a number not in the frozen key: 12 (5 times) and 78 (once); see key_derived_n94.tsv. No correction here; the reading (reading_n94.md) marks the 14 probable words in { }.
R01 H enteradoise
R02 H susuperiororder
R03 H dara
R04 H condcimientodel
R05 H resultado
R06 H escuanto
R07 H encontestacionasuoficio
R08 H recibido
R09 H delanoche
R10 H nosehaganingunaconcesionaescoda?guesej?sgueconarregloalasle?essifueraaprendido
R11 H seranigualmenme
R12 H procure
R13 H seanaprendidosnosoloescoda?saguetasinolosdemasdispersos
R14 H recididoelueinte
R15 H ohosecunplimenstara
R16 H doscientoscuarenta?dos
R17 H ochenta?unoReading
# taulet1866: reading of the cipher runs of N.94 (Spanish), 10 Nov 2026
Letters and telegrams received by colonel Bernardo Taulet, January 1866. AHN, DIVERSOS-COLECCIONES, 193, N.94 (PARES
5855640), views 43, 87, 78 and 26: the held-out portion H of the re-test (325 numbers, 17 runs R01-R17; see
validation.txt and retest/). Decoded with the frozen key (frozen/KEY.tsv), unchanged.
Marks: clear words of the documents are given only as context in ⟨ ⟩ (not counted). In the cipher: {x} = probable (a
number corrected by sense, or a number not in the frozen key read by sense; the justification follows each passage);
[x] = restored; ‹…› = unresolved. There are no restored and no unresolved words here.
Classes (Daniel's rule). **Certain** = every letter given by the frozen key, applied without change. **Probable** =
one or more numbers corrected by sense inside the word, or a number absent from the frozen key (12, 78) read by sense.
## Counts (cipher words)
| passage | words | certain | probable | restored | unresolved |
|---|---|---|---|---|---|
| letter from Albarca, 28 Jan 1866 (view 43, R01-R09) | 22 | 20 | 2 | 0 | 0 |
| telegram of the Minister of War (view 87, R10-R13) | 33 | 27 | 6 | 0 | 0 |
| acknowledgement (view 78, R14-R15) | 6 | 2 | 4 | 0 | 0 |
| troop figures (view 26, R16-R17) | 7 | 5 | 2 | 0 | 0 |
| **total** | **68** | **54** | **14** | **0** | **0** |
Read share: certain only 54/68 = **0.794**; certain + probable 68/68 = 1.000. N.94 stays "read in part" with the
whole target because the key fell short of our validation rule again (margin 0.261 < 0.30).
The fourteen probable words: orde{n}, con{o}cimiento (view 43); {y}, {q}ue, j{u}zgue, le{y}es, igualmen{t}e, {y}
(view 87); reci{b}ido, {v}einte, o{c}ho, cu{m}plimentará (view 78); {y}, {y} (view 26).
Two counting slips of retest/clair.txt are corrected here (the totals 54/14 were right): orde{n} is probable but was
not braced there, and the number 12 occurs 5 times, not 4.
## Letter from Albarca (view 43, 28 Jan 1866): R01-R09
⟨…⟩ enterado i se ⟨cumplimentará⟩ su superior orde{n}, ⟨quedando en⟩ dar a ⟨V.S.⟩ con{o}cimiento del ⟨inmediato⟩
resultado. Es cuanto ⟨debo decir a V.S.⟩ en contestación a su oficio ⟨de este día⟩ recibido ⟨a las 9⟩ de la noche.
- "enterado i se": the cipher has 92 (= i) for the conjunction y, a common spelling of the time; counted certain.
- orde{n} (R02): the last number is 27 (= r), "order", where 29 (= n) is required.
- con{o}cimiento (R04): 61 (= d), "condcimiento", where 64 (= o) is required.
- The signer is read as Rafael Ramis (uncertain); not used for the key or the measurement.
## Telegram of the Minister of War, forwarded by the Captain General (view 87): R10-R13
⟨…⟩ no se haga ninguna concesión a Escoda {y} {q}ue se j{u}zgue con arreglo a las le{y}es si fuera aprendido.
⟨Los prisioneros⟩ serán igualmen{t}e ⟨juzgados⟩. Procure ⟨V.E. … que a toda costa⟩ sean aprendidos no solo Escoda {y}
Sagueta sino los demás dispersos.
- {y}, le{y}es (R10), {y} (R13): number 12, absent from the frozen key; y by sense (see key_derived_n94.tsv).
- {q}ue (R10): 19 (= g), "gue"; the frozen key has no q, and "Escoda y que se juzgue" is imposed by the sense.
- j{u}zgue (R10): number 78, absent from the frozen key, u by sense; the z is written with 10 (= s), kept as written.
- igualmen{t}e (R11): 38 (= m), "igualmenme", where 88 (= t) is required.
- "aprendido(s)" = aprehendido(s), arrested: spelling of the original, counted certain.
- The clear words between R12 and R13 are partly doubtful in the image ("V. E. ? que a toda costa"); see validation.txt.
- Escoda and Sagueta (the "Saguete/Sagueta" of the N.99 reports) are rebels of the January 1866 rising; not identified
further here.
## Acknowledgement (view 78): R14-R15
⟨…⟩ reci{b}ido el {v}einte ⟨y⟩ o{c}ho, se cu{m}plimentará ⟨…⟩
- reci{b}ido (R14): 61 (= d), "recidido", where 14 (= b) is required.
- {v}einte (R14): 55 (= u), "ueinte", for v (u/v).
- o{c}ho (R15): "oho", the c is missing.
- cu{m}plimentará (R15): "cunplimenstara", 29 (= n) for m and an extra s; that s is our doubtful reading 10? of a
mark at the start of a line of the image (see validation.txt).
## Troop figures (view 26): R16-R17
⟨y siendo esta la de⟩ doscientos cuarenta {y} dos ⟨incluzos los Tambores, solo podrá componerse cada una de estas
tres columnas de⟩ ochenta {y} uno.
- {y} (R16, R17): number 12, y by sense.
- The figures agree with each other: 242 men, including the drummers, make three columns of about 81 (3 x 81 = 243).
Internal consistency only, not a validation.Translation
# taulet1866: English translation of the cipher runs of N.94 (with the clear context), 10 Nov 2026
Clear text of the documents in ⟨ ⟩; probable words in { }. See reading_n94.md.
- Letter from Albarca (28 Jan 1866): ⟨…⟩ Informed, your superior {order} ⟨will be carried out⟩, ⟨and I undertake to⟩ give
⟨Your Lordship⟩ {notice} of the ⟨immediate⟩ result. This is all ⟨I have to tell Your Lordship⟩ in answer to your
dispatch ⟨of this day⟩, received ⟨at 9⟩ in the evening.
- Telegram of the Minister of War, forwarded by the Captain General: ⟨…⟩ no concession is to be made to Escoda, {and}
{let} him be {judged} according to the {laws} if he is arrested. ⟨The prisoners⟩ will {likewise} ⟨be judged⟩. ⟨Your
Excellency⟩ is to see to it ⟨… that at all costs⟩ not only Escoda {and} Sagueta but also the other scattered men are
arrested.
- Acknowledgement: ⟨…⟩ {received} on the {twenty} ⟨-⟩ {eighth}, it will be {carried out} ⟨…⟩
- Troop figures: ⟨and this being that of⟩ two hundred and forty-two ⟨including the drummers, each of these three
columns can only be made up of⟩ eighty-one. (In both numbers the Spanish "y", "cuarenta {y} dos", "ochenta {y} uno",
is probable.)Key: key.tsv
06 Sources
- Archivo Histórico Nacional, Madrid, DIVERSOS-COLECCIONES,193,N.99: PARES description 5928106, 54 images (reports views 25–42, telegram views 52–53); N.94: PARES description 5855640, views 26, 43, 78 and 87. The images are not reproduced here.
- Brice Dudout, the target folder: transcriptions, key, frozen protocol and measurements (
frozen/,retest/), readings and translations; pull request #376. - Catalogue entry 2444 (Daniel Bourdeau’s sweep of PARES, AHN Diversos-Colecciones, vetted 9 November 2026), now removed from the catalogue.
The notes, the key, the transcriptions, the readings, the translations and the frozen test files are in targets/taulet1866.
Cite this page
Brice Dudout, “Bernardo Taulet’s copybook and letters received, Jalet (Falset), January–March 1866”, Cyphersolver, posted 10 November 2026, https://cyphersolver.com/taulet1866.html.