Codes and Ciphers of Thurloe's Agents

John Thurloe became head of intelligence in July 1653 (after the dismissal of the Rump Parliament by Cromwell) British Civil Wars and Commonwealth website) and became secretary of state when Oliver Cromwell established a Protectorate regime at the end of 1653. Thurloe built upon the intelligence network he took over from his predecessor, Thomas Scot, who managed intelligence of the Commonwealth in 1649-1653. The efficiency of Thurloe's espionage system led to a myth of "all seeing little secretary", which was often enough to spread dissension among the enemies (Marshall p.24).

The present article describes various codes and ciphers used in letters of Thurloe's agents printed in Thurloe State Papers (British History Online).

J. Peterson (1653)

In 1653, in the midst of the Dutch War (1652-1654), J. Peterson, long residing abroad, was employed as an agent for collecting information about Holland. He had been provided with a code of about 600 elements, including 34(a), e(185), 82(Flushing), 90(Texel), 109(fleet), 304(i), 352/366/380(the), 384(are), 464(they), 603(want).

This code has three merits that make it more difficult to break. First, frequent words or letters are assigned several numbers. Secondly, code numbers for single letters are distributed among those for words and syllables, while many ciphers used at the time reserved lower numbers for single letters (see, for example, various ciphers described herein below and Charles I's cipher in another article). Thirdly, assignment of words and syllables to code numbers is not alphabetical. A code with non-alphabetical ordering of elements is known as a two-part code because separate sheets must be provided for enciphering and deciphering. However, Peterson's code appears to have been arranged in a two-dimensional table and thus eliminated the need of two sheets. This aspect will be described with regard to Manning's code (see below), which may have been based on Peterson's code.

There are some four-digit codes such as 1000(East India) and 1005(Hague). In one of his first reports in June, he proposed some addition to the code.

in your counterparte add, for my friend at Ameth, in case yow have occasion to mention him 1003, and for my friend here Mr. Henry Collier 1004.
Jan Pieterson, 12/22 June 1653 (Thurloe State Papers, p.289)

Henry Manning (1655)

Henry Manning acted as a spy in Charles II's court in exile and was shot for the treachery.

In late autumn of 1654, Manning came to Cologne, where the King held a court in exile. He talked of a plan to capture Plymouth for the King but, as Hyde noticed some inconsistencies in his story, Manning was received with a cold civility. In February 1655, Charles secretly left Cologne for Middelburg, Zealand, expecting a royalist rising in England, but, learning that a premature rising was easily suppressed, he returned to Cologne in April (Brett, p.122).

Manning followed the King to the Low Countries and wrote his first letter to Thurloe dated 13 March 1655 NS, offering to provide intelligence about the royalist movement. In this and several following letters, he repeatedly asked for money, a cipher, and unsuspicionable cover addresses. To gain confidence, he even ran "a hazard of being ruined" by writing some particulars in plaintext. At last, Thurloe was convinced of the value of his information. Manning received Thurloe's cipher as of 16 April NS (Nicholas Papers p.151, 186, citing Egerton MS 2542 f.131, 168, 171; Greenspan p.114, citing Egerton MS 2542, f. 171) but he thought it insufficient.

I expect by the next, full directions in the same cipher I here send you; which realy, sir, not being well att this present, is not so full, I doubt, as I could wishe; but what there may want, as you direct, I shall make good in the next
Manning to Thurloe, Antwerp, 26 April 1655 NS

His cipher had some 600 elements including 41(b), 54(a), 196(e), 324/338(i), 372/386(the), 598(your) and some four-digit code numbers such as 1005(Charles Stuart) and 1016(Cologne). As with the case of Peterson's code, code numbers for single letters are distributed among those for words and syllables, which are arranged in non-alphabetical order. Further, many code numbers are exactly twenty greater than those in Peterson's cipher. Thus, Manning's and Peterson's ciphers would have been prepared by the same originator.

Presence of an assignment such as 259(rt) suggests bigrams are systematically assigned. As it turned out, the assignment is not completely random. If words and syllables are arranged alphabetically, the code numbers would form a regular sequence of step 14, which would result by regularly filling a table of 14 columns. The partially reconstructed first two columns are as follows.

Manning frequently pointed out to Thurloe that the royalists were passing through Dover with ease. He requested to have a cipher with the governor of Dover (Cal. SP 1655, p.192-193) and boasted he could have provided intelligence to allow the arrest of some royalists (Cal. SP 1655, p.235).

Though persons named in Manning's letters were later arrested in England, it is not clear how far the arrests should be accredited to Manning (Greenspan p.111). When Colonel James Halsall and others were arrested for a plot of assassinating Cromwell in November 1655, it was the information including letters and ciphers provided by Halsall's servant that led to the arrests (Greenspan p.112).

In November 1655, Hyde received warning from Antwerp that every post brought many letters for Manning and he had letters of credit upon a merchant of Antwerp for "good sums of money". When letters to and from Manning were intercepted, the suspicion was confirmed.

On 5 December NS, the servants sent by the King caught Manning writing with ciphers on the desk. Apparently, Manning was not careful in his use of ciphers. One colonel testified on examination that he had two or three times seen Manning working with ciphers when he shared lodgings with him in October. (Aubrey pp.102-103)

On 15 December NS, Manning was shot by the royalists near Cologne.


Eva Scott, The Travels of the King, pp.140-156

Clarendon, The History of the Rebellion and Civil Wars in England, xiv (Google Booksearch)

N. Greenspan, 'News, Intelligence, and Espionage at the Exiled Court at Cologne: the case of Henry Manning', Media History, Vol. 11, Issue 1 & 2 April 2005, pp.105-125

Philip Aubrey, Mr Secretary Thurloe: Cromwell's Secretary of State, 1652-1660

A.C.A.Brett, Charles II and His Court

William Stamford (1655)

William Stamford in Calais offered to discover a royalist plot in England in the spring of 1655. His cipher was as follows. (E=12/25)

Higher numbers are words and names such as 65(colonel), 67(England), 81(which), 82(from), 130(general rising), and 158(designs). Further, a circle-dot symbol like Θ represented "the lord protector".

His letters are found in Thurloe State Papers (13 March NS (Page 187 and Page 189), 30 March (Page 274), and 3 April (Page 340).

In the first of these, he observed: "if this correspondence continue betweene us, I should desire you to sende me another cipher, this which you have sent mee already being very imperfect." Indeed, Stamford's cipher which he says Thurloe had sent was inferior to Peterson's and Manning's in its seemingly small vocabulary and exclusive assignment of lower numbers to single letters. However, Stamford's code at least has a merit in that the code numbers for single letters are not quite alphabetical.

John Butler (1656)

John Butler, an informant in Holland, used a cipher, which seems to have randomly assigned numbers 1-60 and generally alphabetically assigned numbers 400-424 to represent single letters, in a letter of 22 September 1656 (cf. deciphered text and ciphertext (Page 575) in Thurloe State Papers). It had some codes for names such as 62(Spain) and 156(Charles). Other numbers such as 913, 350, etc. may be nulls. (E=3/5/405)


Another Informant

Another letter of intelligence written about the same time used the following cipher (To William Hanmer, 17 September 1656 NS, Thurloe State Papers, p.391).


Collision of "b" and "m" both enciphered as "1" may reflect a transcription error

Agent in The Hague (1) (1653-1658)

Thurloe State Papers includes many letters of intelligence from The Hague written either in English or French. Many of them use code for names (104 [the States General], 105 [the States of Holland, Hollanders, Holland], 130 [the Protector, Cromwell, England], 145 [the Orange party], 198 [Brussels]) and some words (124 [ambassador], 135 [money], 171 [peace], 179 [commerce, trade]).

Numbers less than 100 were reserved for single letters. (E=24/54/[82])

Interlined deciphering indicates significant regularity in the assignment of numbers (see letters in bold face), which allows reconstruction of missing parts. (As to the identification 60[i], there is counterevidence (60[o]) but regularity seems well established that this may be a transcription error. Identification "78[a]" is deduced from 37[s] 78 39 [u] 33[o] 86[i] 66[r], which is interlinearly deciphered as "know it".)

The above cipher is the same as the one used by General Blake (see below) in writing to Cromwell (separate from one he received from Thurloe).

An agent in Brussels named Burford also appears to have used this cipher in his letter of intelligence dated 16 January 1656 (Thurloe State Papers), in which an undeciphered passage "26 33 39 24 36 31 24 32 38" can be deciphered as "gouerment".

General Blake (1655)

In 1655, England took sides with neither of the belligerents in the war between France and Spain, though Cromwell had sent an expedition to the Spanish West Indies. Cromwell also sought to intercept the Plate Fleet bringing riches of the colonies to Spain.

Robert Blake, General at Sea, sent to the Mediterranean with a strong fleet to protect the English interests, was ready to fight either the French or the Spanish as necessary. He had also received secret instructions from Cromwell about the silver fleet and reported his will execute it in a letter to Cromwell almost entirely in cipher (12 June 1655 Thurloe State Papers, p.541). In his letter to Cromwell dated 4 July, he only enciphered his reference to the Plate Fleet (Thurloe State Papers, p.611).

The cipher portion appears to be the same as one used by an agent in The Hague (see above) (E=24/54/82), though not the assignment of higher numbers. On the same day (4 July), he acknowledged receiving a cipher from Thurloe.

Unfortunately for Blake, the Spanish Plate Fleet of 1655 reached home during his absence in England.

Edward Montagu (Mountagu), later 1st Earl of Sandwich (1656-1657)

Cromwell chose Montagu, who had no maritime experience, as a second-in-command for Blake. The fleet, with Blake and Montague onboard Naseby, sailed back to await the 1656 Plate Fleet in the spring of 1656.

Montagu used a cipher as follows in writing to Thurloe (E=18/42/56/93, THE=407). The same cipher was used when Cromwell wrote to Blake and Montague (9 June 1656, Thurloe State Papers, p.101; see also 13 September 1655, Thurloe State Papers, p.724) and, next year, in Cromwell's draft letter to Montagu concerning the Swedish affairs (9 October 1657, Thurloe State Papers, p.582).

Higher numbers include generally alphabetical list of words such as 105(and), 407(the), 482(year) and names such as 622(Brazil).

Richard White, a.k.a. George Pawly in Madrid (1656)

In 1655, England and Spain went into war. Spain was already at war with France and was reluctant to break with England. But Cromwell was determined to have France rather than Spain as a friend and the English expedition to the Spanish West Indies angered the Spanish.

In July 1656, there arrived in Madrid a Richard White, who was at best a mediocre intelligencer for Secretary Thurloe of the Cromwellian regime.

Richard White had been imprisoned in the Bastille as a spy for Spain but was released through exertions of Colonel Bampfield. Bampfield was then Thurloe's agent in Paris but when Lockhart was appointed as an official Parliamentary ambassador to Paris, he started to extend his activity (de Beer p.398-399; Apology p.173-175). Through his mediation, Richard White was employed by Thurloe as an agent to collect intelligence about Spanish affairs.

On arrival in Madrid, he showed the Spanish ministers his code with Thurloe as a token of England's willingness for peace negotiations (de Beer p.400).

As was often the case with spies, Richard's secret reports used false names and false cover addresses. He signed "George Pawly" (or its abbreviations) in letters to Bampfield or Thurloe and "Wescomb" or "Jo. Williams" in those to his brothers. The letters to Thurloe were generally addressed to Matthew Bonnell (de Beer p.399 n.6).

He used combination of three types of cipher when he sent intelligence from Madrid in 1656. De Beer describes them as "a clumsy cipher and a series of pseudonyms, the sort of code which only the half-educated would adopt".

Numerical Cipher

The first type is a common numerical cipher. An example is seen in his report of 17 August 1656: "27 19 26 12 18 11 28 21 31 11 28", which reads "New Foundland". While the printed version of Thurloe's State Papers assign letters as:
27 19 26 12 18 11 28 21 31 11 28
 n  e  w  f  o  u  n  l  a  n  d,
the following may be more reasonable in view of correspondences "28=d" and "11=n".
27 19 26 12 18 11 28 21 31 11 28
 n  e  w  f  u  n  d  l  a  n  d.

Higher numbers represent words in a roughly alphabetical order:

119 gallies
132 Indies
133 Jamaica
139 intelligence
151 London
153 letters
168 merchant
170 Madrid
175 the Duke of Modena
193 the Protector
200 Portugal

Simple Substitution Cipher

A simple substitution cipher was also used. For letters a-i and r-u, it was a Caesar cipher, in which each letter is replaced with the letter after the next in the alphabet. The substitution table can be reconstructed from deciphered letters as follows.

plaintext: abcdefghiklmnopqrstuwxyz

where ciphers for y and z are uncertain because of insufficient instances and those for k, w, x are conjectures. Identification poses some difficulty because of inaccuracy of the printed transcription (cf. de Beer p.399 n.5). When there are two ciphers "hink" and "hluk" for "fish" in print, this would imply a transcription error rather than variation of the cipher. Particularly, the cipher for "a" is often printed as "e".

The above table is consistent with a cipher "febgIt" translated as "hazard" in Thurloe State Papers when the cipher is corrected to "fcbIgt", which can be deciphered as "danger".

Word Substitution or Code Names

The type of cipher that was most frequently used by Richard White was substitution of words having the same initial letter. The vocabulary included not only code names for persons and places but also common nouns and verbs. Although spelling/transcription varies for some cases, the following list shows some of the codes.

Names include:

Allex : don Alfonso [don Alonso] de Cardenas (withdrawn Spanish ambassador in England)
Bartaine : admiral Blake
Borama : Bilboa
Clyrr : Cadiz
Compton : Prince of Conde
Creame : Cardinal Mazarin
Draffords : Dunkirkers
Egington : England
Fratford : France
Frinds : Flanders
Fruxe : France, the French, the French king
Hunts : Hollanders
Huply : Holland
Infidels : Irish
Inglands : Indies
Inrade : don John of Austria
Liver : don Louis [Lewis] de Haro (1598-1661; Spanish minister)
Marga : Madrid
Pantha : Portugal
Prant : Paris
Roba : Rome
Sligo : Spain
Smoke : San Sebastian
Sparker [Sparter] : King of Spain
Starmer : Scots king
Tribe : Thurloe

Other words include:

aux : agent, ambassador
ax : advise
axd : advised
axing : advising
buckler/butler : brother
bussune : business
candel/candle : condition
Cocke : colonel
croude/crowde : council (of state)
fish : fleet
gores : guns
gramp : galleon
Ix : I, me
kamfer/kainfer : king
kapp : know
lard : letter
mapp : man
mapps : men
max : money
monta : marry, match
montaage : marriage
nosegays : news
palme : peace
pearles : pistoles
perry : pope, papists
plate : pension
pott : pieces of eight
protrax : protector
stake : ship, merchantman
starkle : secretary
starr : ship
talke : take
tar : that
tarte : treat, treaty
water/wafer : war
wander : write
yx : you

Word Substitution between Brothers

Richard used a similar but different substitution vocabulary in letters to his brothers under the names of Wescomb and Jo. Williams:

9 August 1656, Wescomb to Andrew White in Paris (Thurloe SP; see images of p. 254, 255)

12 August 1656, Jo. Williams (Thurloe SP, see images of p. 264, 265)

30 August 1656, Wescomb (Thurloe SP, see images of p. 728, 729)

5 September 1656, Wescomb (Thurloe SP, see image of p. 346)

28 March 1657, Wescomb to Ignatius White in London (Calendar of the Clarendon State Papers)

In the following list, codes in red differ from those used with Thurloe and codes in blue are the same as or similar to those used with Thurloe.

Allen : don Alonso
Aylmer/Aylmour/Aylmor : [Andrew] White [then in Paris, going to Rome]
butlers : brothers
Corkia : Catalonia
Comfort/Comefort : the protector (Cromwell)
Mr. Cro. : Colonel Bampfield [The letter of 28 March 1657 uses "Brown".]
Edward : Charles Stuart
Egington : England
Frigia/Friga/Furgia : Flanders
Frux : the king of France [The letter of 12 August 1656 uses "Feinex"]
Ingland : Lord Inchiquin
Inkeepers : Irish
Jackson : White in Lambeth
Longge : don Lewis de Haro
match/mace/mach : money
Marsia/Mursia : Madrid
Mollineaux/Molneaux/Molleneux/Mollenax : Mazarin
nutmegs/nottmeggs : news
Peroy : the pope
pitch/pich/pick : peace
Prussia : Portugal
Rowland : Richard White
sadler : secretary
Spencer/Spenser : king of Spain

It was unfortunate for Bampfield that the White brothers were unreliable (Apology, p.174).

The letter of 28 March 1657 reported arrival of Henry Bennet at Madrid as the King's agent to obtain assistance for the royalist cause. Bennet, on his part, informed the Spanish minister don Luis of White's sending letters by an express, who was consequently stopped. Bennet translated and deciphered the intercepted letters. The Spanish wondered whether knowledge of their affairs by the French court was leaked by White or someone about don Alonso or their cipher was known at Paris. (Bennet to Hyde, 23 May 1657; Calendar of the Clarendon State Papers iii p.293)

The Spanish ordered Richard White not to meddle any more with things which did not belong to him. Richard came to Bennet, bringing all his letters and ciphers, and confessed that he was employed by Thurloe. (Bennet to Hyde, 23 May 1657; Calendar of the Clarendon State Papers iii p.324)

Word Substitution for '905' (Jermyn)

Another letter of 12 August 1656 (Thurloe SP, see image of p. 263), which is in the same hand as the letter by the name of Jo. Williams of the same date, used a similar cipher. Bampfield notes this was written to '905', which was a code for Jermyn he used with Thurloe.

In the following list, codes in red differ from those used with his brothers and codes in green are the same as or similar to those used with his brothers.

antagonist : ambassador
Coningsmarke : the protectour (Cromwell)
Our Edward : Char. Stewarte
Egington : England
Laurence : don Lewis
Phenix : the king of France
pitch : peace
Rawly : [Richard] White
Ravenna : Rome
Spencer : the king of Spayne

Similar Word Substitution by a Royalist Agent

Despite de Beer's disparaging remarks quoted above, there are precedents for such a pseudonym scheme.

A letter dated 2/12 April 1655 from Andrew Johnson (perhaps a pseudonym), who looked for a correspondent in London for the royalist secretary Nicholas, used similar word substitution (Nicholas Papers):

carrel : correspond
faith : faction
Mr. Koply : King
Lonley : (probably) London
Mr Newington : Nicholas
Padly : Paris

Elizabethan Example

Even as far back as the Elizabethan era, a similar scheme had been used. According to Butler (p.132), Public Record Office contains a very lengthy document containing several pages of such pseudonym words: "thankful" (representing "traitorous"), "suck" (representing "suborn"), "weight" (representing "war").


E. S. de Beer, 'The Marquis of Albeville and his Brothers', The English Historical Review, Vol. 45, No. 179 (1930), pp. 397-408

A.J. Butler (1901), 'Some Elizabethan Cipher-Books', Transactions of the Bibliographical Society, VI, pp.127-135 (Internet Archive)

*According to State Papers, Domestic Series, 1656-1657, p.306, a partial key to the cipher used by Richard White is in the Cypher Collection, Public Record Office, Vol.6, No.50.

William Lockhart (1656-1658)

William Lockhart was appointed by Cromwell as official ambassador to France in February 1656 and he was successful in reaching an agreement with Cardinal Mazarin to join forces against Spain (British Civil Wars and Commonwealth website).

He used a code (E=72, THE=403) of about 1000 elements, whereby "the" is encoded as "403" and there is a code number as high as 945 (Flanders).

As in the case of Peterson's cipher (see above) and Manning's cipher (see above), the code numbers for single letters are distributed among those for words and syllables and the non-alphabetical ordering of the elements seems to be the result of regularly filling a table of 12 columns. The partially reconstructed columns are as follows.

His letters partially in cipher include the following:

Lockhart to Thurloe, 8 August 1656 (Thurloe State Papers, see images of p.252, 253)

Lockhart to Thurloe, 12 August 1656 (Thurloe State Papers, see images of p.265, 266)

Lockhart to Thurloe, 7 February 1657 (Thurloe State Papers, see images of p.21, 22)

Lockhart to Thurloe, 4/14 April 1657 (Calendar of the Clarendon State Papers)

Lockhart to Thurloe, 31 May 1657 NS (Thurloe State Papers, see images of p.298)

Lockhart to Thurloe, 1 June 1657 (Thurloe State Papers, see images of p.301)

Lockhart to Thurloe, 2 June 1657 (Thurloe State Papers, see images of p.305)

Lockhart to Thurloe, 5/15 April 1658 (Thurloe State Papers, see images of p.52)

(Transcription of Thurloe State Papers is not always accurate in assigning the deciphering to code numbers. For example, 241 10 284 87 302 should be parsed as 241[Gr] 10[a] 284[ve] 87[l] 302[ing] rather than 241[Gra] 10[v] 284[el] 87[ing] 302[h].)

Blank Marshall (Michael Deane) (1656-1658)

In the spring of 1656, King Charles left Cologne and came to Brussels, the capital of the Spanish Netherlands, to obtain help from the Spanish Governor, Don John. Spain's entry into war with the Protectorate England made Spain a potential ally for the royalist cause. After an agreement was reached in April, Charles moved his court to Bruges.

Blank Marshall (pseudonym of Michael Deane) was an agent based in Bruges and spied on the King's court. His reports ranged at least from August of 1656 (Thurloe State Papers, p.293) to September 1658 (Thurloe State Papers, p.353).

His cipher regularly assigns four numbers to each consonant and five numbers to each vowel (E=18/19/20/21/21). A small number of codes were assigned to names such as 104(Charles Stuart), 111(Bruges), and 173(Ormond). (The original key is in Add MS 4166, f.117 (DECODE R4897).)

William Jephson in Sweden and Philip Meadowe in Denmark (1657-1658)

When war broke out between Sweden and Denmark in 1657 as a part of the Second Northern War (see Wikipedia), Cromwell sent envoys to the belligerant courts to mediate a peace.

Major-General Jephson was sent to the Swedish court. He used a code (THE=444) of about 600 elements in writing to Thurloe. There appear to be code numbers as high as 607/637 ("the Dutch ambassador"). (Apparently, even higher numbers 641 for "treat" and 825 for "year" are errors for 461 and 525, respectively.)

Numbers 1-104 appear to represent single letters (see below) (E=14/29/70/84). Higher numbers are assigned to a roughly alphabetical list of words such as 108(and), 444(the), and 525(year). Further numbers are for names.

Philip Meadowe was sent to Denmark. His code was as follows. As with Jephson's code, higher numbers are for words and names such as 559(Denmark) and 610(Dutch ambassador) (E=6/50/50). (The original key is in Add MS 4166, f.102-103 (DECODE R4890) below.)

Algernon Sidney (1660)

War between Denmark and Sweden threatened naval supplies for the maritime powers such as England and the Dutch. In 1659, England sent a delegation under Edward Montagu to mediate a peace. Death of Charles X of Sweden in February 1660 removed one major obstacle and the Treaty of Copenhagen was signed on 27 May 1660 by Denmark, Sweden, France, England, and Holland. (Wikipedia, s.v. Algernon Sidney, Second Northern War)

In a letter of 2/12 April 1660 (Thurloe State Papers, p.882), Algernon Sidney, a member of the delagation, used cipher in some of his complaints against the Dutch. The cipher appears to be a code of some 600 elements (THE=176?).

Colonel Bampfield [Bamfield, Bamfylde] (1655-1657)

Colonel Bampfield, once an agent of Charles I, failed to win the confidence of Charles II after the execution of Charles I and entered the service for John Thurloe in the spring or early summer of 1654 (see another article).

First Cipher

He used the following cipher in his report to Thurloe (E=22/27/30/32/92).

(The transcript in Thurloe State Papers includes some conflicting instances. The above table show such alternatives when it is hard to infer which is correct.) Higher numbers represented words in roughly alphabetical order such as 108 (abus[e]), 754/755/790/799 (the), and 820/829/831 (you) and there were still higher numbers for names as high as 1037 (Rome) and 1038 (the pope). At times, Bampfield felt need to assign a new code number.

7 [C] 26 [h] 15 [a] 36 [r] 41 [l] 20 [s] 70 [S] 37 [t] 30 [e] 73 [w] 17 [a] 64 [r] 80 [t], whose name pray add at 1040
A letter of col. Bampfylde, 2 November 1656 NS

For further precaution, Bampfield occasionally wrote in lemon juice (i.e., invisible ink) (e.g., 24 November 1655, 15 November 1656).

This cipher was used as late as 6 June 1657 NS in the following portion of a letter addressed to Simon Tanner, merchant in London, which was the cover address used when Bampfield wrote to Thurloe (A letter of intelligence from Bampfield, 21 July 1657).

I have likewise sownde meanes to intercept some letters goeing to 13[b] 30[e] 43[n] 22[e] 37[t] 90[-] 5[l?] 146[?] 962[Madrid].
A letter of intelligence from Bampfield, 6 June 1657 NS

(Henry Bennet had been sent to Madrid in 1657 as the King's agent to obtain assistance for the royalist cause.)

Second Cipher

The death of the Holy Roman Emperor Ferdinand III in April 1657 created a potential instability factor in Europe. Usually, the Electors chose a successor (King of the Romans) during the emperor's life but on this occasion, the elected King of the Romans had died before the emperor and candidates other than a Hapsburg coveted the crown of the Empire.

Bampfield offered to gather intelligence in Frankfurt, where the diet would be held for election. Since he did not receive definite instructions from Thurloe, he left Paris and embarked for England from Dieppe on 20 July. However, the vessel was caught by pirates from Ostend and Bampfield was left with no "money, friends, nor hardly clothes to my back".

Though Thurloe was displeased at Bampfield's attempt to return to England without leave, he confirmed Bampfield's mission in Germany. Thus, after visiting Paris for a short time, Bampfield went to Frankfurt.

Bampfield's reports from Frankfurt used a cipher different from the above (E=19/23/45/68/75).

Again, higher numbers represented words in roughly alphabetical order such as 107(about), 744/749 (the), and 802(yesterday) and there were still higher numbers for names such as 828 (Mayence).

Later Years

Apparently, Bampfield stopped writing to Thurloe after December 1657 partly because silence of Thurloe left him unsure whether his report from Frankfurt was needed (Apology p.182). Anyway, election of the Habsburg candidate Leopold was all but finalized early in 1658.

Bampfield contacted Thurloe again in the winter of 1658-59 (Apology p.184) but then the political situation in England changed. When the Protectorate after the death of Oliver Cromwell collapsed and the Rump Parliament was reinstated in May 1659, Thurloe was replaced by his predecessor Thomas Scot but Bampfield actively continued to work under Scot (Apology p.185; C.H.Firth, 'Thomas Scot's Account of His Actions as Intelligencer during the Commonwealth', English Historical Review (1897), p.122 (Internet Archive)).

Upon the Restoration in May 1660, Bampfield was put in the Tower for fourteen months. After release, he went to the Dutch republic. For some time he tried to endear himself to the Restoration government but his efforts were not rewarded. When he sent a report in June 1663, Bennet, now secretary of state, asked him bluntly if the report in cipher was worth the time required to decipher it (Apology p.189). Finally, Bampfield found employment under John de Witt, leader of the Dutch republic (Apology p.188).


John Loftis and Paul H. Hardacre ed., Colonel Joseph Bampfield's Apology "Written by Himself and Printed at His Desire" 1685

Agent in The Hague (2) (1658)

A letter from The Hague of 28 May 1658 NS (Thurloe State Papers; see image of Page 141) uses a cipher as follows (E=61/62/63/64/65/66/67/68).

Numbers 200-600 are codes of words and syllables in generally alphabetical order: 214(about), 303(day), 383(ho), 481(of), 554(the) and numbers above 600 are names such as 625(Charles Stuart), 677(Holland).

George Downing (1658-1660)

George Downing (1623-1684) is probably best remembered for founding Downing Street in the 1680s. His diplomatic career started during the Interregnum and in early 1658 he arrived at The Hague as resident. During the first few years in The Hague, he developed his talents as a spymaster and his character as a diplomat (Marshall, 265).

In many of his letters, he used a code (E=39/40/41/42/43/44/45, THE=468) of about 600 elements including generally alphabetical words and syllables such as 178(ab), 227(ba), 231(be), 232(been), 237(bi), 239(by), 241(bo), 242(bu), 244(but), 468(the), 524(us). (Somehow, the "u/v" section comes after the "w" section and the "y" section.) Higher numbers are for names and titles such as 527(Lord Nieuport), 534(Lord Protector), 535(States General), 536(King of Sweden), 572(Charles Stuart), and 585(Brussels).

Lower numbers were reserved for single letters. (The original key is in ADD MS 4166, f.115-116 (DECODE R4896).)

After the death of Oliver Cromwell, Secretary of State Thurloe was dismissed in mid-1659. When he was reinstated at the end of February 1660, Downing resumed writing to him with the same cipher.

While providing intelligence to Thurloe, Downing foresaw the King's restoration and managed to get reconciled with Charles II in April 1660. Thus, he was continued in his embassy in Holland at the Restoration. He remained in The Hague until the second Dutch War broke out in 1665.

He used code in writing to Lord Chancellor Clarendon. This code also includes letters and syllables in generally alphabetical order: 19(a), 29(c), 38(c), 41(d), 47(e), 98(n), 128(r), 128 (r), 271(ce), 366(ga), 395(he), 530(pe), 549(ra), 582(sl).

Henry Cromwell (1658-1659)

Henry Cromwell (1628-1674) was the fourth son of Oliver Cromwell and served as major-general of the forces in Ireland and lord deputy (later lord lieutenant) of Ireland. When he reported his arrival in Dublin in 1654, he used a cipher that he had sent to Thurloe.

I hope you received the character, which I sent at my comeing out of towne to you. Make use only of the upper clavis to uncypher the inclosed.
H. Cromwell to Thurloe, 8 March 1654 (Thurloe State Papers, see image of Page 149)

It was a simple substitution cipher according to the following table.

plaintext: abcdefghiklmnopqrstuwxyz

(Ciphers for x, z are conjectures.)

Apparently, Thurloe provided him with a new cipher, for H. Cromwell acknowledged receipt on 5 July 1655.

From 1658 to 1659, he received letters in cipher from Lord Fauconberg, who married Henry's sister in 1657. It was a numerical cipher as follows (E=11/13).

Here, ciphers for z are conjectures. Further, some names were represented by a capital letter such as A (Henry Cromwell), O (Lambert), V (Desbrowe/Desborough), Z (Protector).

Lord Broghill (1655-1660)

Roger Boyle, Lord Broghill from Munster, Ireland, assisted Cromwell's subjugation of Ireland in 1651. Broghill was elected for Cork, Ireland, in the First Protectorate Parliament (September 1654 to January 1655). (The Protectorate Parliament included representatives from Scotland and Ireland.) In March 1655, he was appointed Lord President of the Council for Scotland and resided in Edinburgh. In Scotland, Broghill assisted General Monck, commander of the forces in Scotland, who presided over a good espionage system in Scotland (Marshall, p.25).

In a letter of 22 September 1655 to Thurloe, he used the following cipher (E=18/42/56/93), with higher numbers assigned for words such as 217(general) and 256(intelligence). (March 2024: I noticed this is the same as Montagu's cipher above, many years after I wrote this.)

Broghill intercepted royalist correspondence between Colonel Borthwick, Earl of Glencairn, and Charles II and the ciphers used in them (Broghill to Thurloe, 27 November 1655; 18 December 1655). Carriers, Major Borthwick (brother to the Colonel) and Colonel Placketer, were cooperative.

Being an ardent supporter and friend of Cromwell, he was included in the inner cabinet of Cromwell's council when he returned to England (Wikipedia). He used intelligencers and received report in cipher (Broghill to Thurloe, 1 January 1656; 13 May 1656; 26 November 1656).

He was returned again to the Second Protectorate Parliament (September 1656 to June 1657, January to February 1658). When he was about to set out for Ireland during the adjournment, he sent Thurloe a cipher.

I heere inclosed present you with a cypher, which I shall make use of as occasion is offered.
Broghill to Thurloe, 3 August 1657

When Cromwell died in September 1658, Broghill supported his son Richard Cromwell. However, by April 1659, it was evident that the new protector could not control the Parliament and he left for Ireland. During 1659, he observed the turn of events. In April 1660, he was still hostile to Charles II. He wrote in cipher to Thurloe (who had been dismissed when the Commonwealth took place of the Protectorate but reinstated in February 1660) that he had entirely secured Munster against any that should be for the king or not for the council of state or parliament (Broghill to Thurloe, 24 April 1660).

The small cipher used here is as follows (E=17/18/19). Higher numbers include words and names such as 49(parliament), 51(council of state), 55(Broghill), and 122(the king).

In the end, Broghill saw the restoration inevitable and he changed his side in time.


British Civil Wars, Commonwealth and Protectorate 1638-60


Eva Scott, The Travels of the King, pp.140-156

When the Protectorate collapsed and Thurloe was replaced by his predecessor, Thomas Scot in 1659, he refused to hand over his codes and ciphers (British Civil Wars website). At the Restoration, however, he provided his expertise to the government in exchange for his release. The intelligence system of the Restoration government owed much to the Interregnum regime.

BL Add MS 4166

(March 2024) I noticed part of Thurloe's state papers are in Add MS 4166 (BL) (they were "In the possession of Joseph Jekyll, Esq." at the time of Thurloe State Papers in print). The catalogue describes the volume as "Letters of John Thurloe. Secretary of State, to Henry Cromwell, Major-General of the Forces in Ireland, 1655-1657; other state papers and cipher-keys from the 1650s; letter of Anne Thurloe, widow of John Thurloe, 1668." Most of the keys seem to belong to the Protectorate because they have an entry "Charles Stuart".

f.77-78 (DECODE R4880) (Henry Cromwell)

Thurloe to Henry Cromwell (3 February 1656), printed in Thurloe State Papers, vol.6 (British History Online).

The key is in f.118-119 (R4898). The vowels and six consonants are assigned numbers 1-6: 1/a:2/e:3/i:4/o:5/u:6/y, and 1/l:2/m:3/n:4/r:5/s:6/w (the other consonants are not enciphered!). A letter is represented by two figures, one on top of the other. The lower figure is even for vowels and odd for consonants. These figures can be chosen arbirarily when enciphering. The upper figure is a sum of the lower figure and the number assigned to the letter.

Superscript12210 14785812610
Base numberC 816Th105725727

This is a simple polyalphabetic cipher, whereby a letter is represented by a pair of figures, where only the difference of the figures is meaningful.

f.83-84 (DECODE R4881), f.85-86 (DECODE R4882) (Vincent Gookin)

Letters from Vincent Gookin (Wikipedia). The first is dated 3 February 1657 (British History Online) and the second is dated 27 January 1657 (British History Online).

Codes include 212(R.), 2999(your frends), 0330(your enimies), 1666(anabaptist), 2000(D.), 5005(H.H.), 5100(L.H.), 9000(L.D.).

f.87-88,89 (DECODE R4883, R4884)

Cromwell to General Blake (13 June 1655) (British History Online). F.89 corresponds to part of the plaintext of f.87v. The cipher is the same as the one for General Blake reconstructed above.

f.90-91 (DECODE R4885)

Cromwell to Blake and Montagu (9 June 1656) (British History Online). The cipher is the same as the one for Montagu reconstructed above.

f.92-93 (DECODE R4886)

An intercepted letter of Edward Hyde dated 1 November 1659 (British History Online). F.92 is in clear except for two code names "Mr. Mills" (Hyde) and "Edmundson" (decoded as "Ch[arles] St[uart]"). F.93 is full of undecoded code numbers, as follows.

I pray direct 966 232 to 925 794 920 932.
925 774 101 953 18 am 404 583 891 923 12 419 14 16 47 you may use 499 368 10 469 601 848 628 or the 890 800 37 that 691 5 772 25 as 670.
220 101 25 12 741 8 Mr. Perkins.
670 333 159 Mr. Tush.
773 15 772 25 Mr. Craft.
264 491 Mr. Knots.
724 890 863 19. Mr. Botts.
670 295. Mr. Dantrey.
670 60 101 Mr. Harle.
141 332 Flamton.
154 671 Bruton.
384 641 Barton.
101 933 811 Mr. Carle.
670 311 Mr. Fuller.
896 36 Mr. Sharpe.
237 406 Mr. Popple.
670 220 Mr. Hespin.
156 409 Bobley.
280 491 Mr. Suson.
194 332 Mr. Bring.
544 467 Mr. Day.
175 336 Mr. Cauton.
224 333 Mr. Bright.
267 660 Mr. Yole.
281 650 Mr. Rower.
193 450 Mr. Brave.
106 431 Mr. Prag.
484 182 Manton.
394 467 Mr. Switon.
239 451 Mr. Clun.
207 407 Cotton.
242 479 Mr. Lark.
599 5 769 740 Mr. Chance.
125 101 331 Dorme.
336 249 Mr. Shent.
526 492 Bosrey.
369 451 Mr. Payne.
325 337 Mr. Bray.
867 863 598 Mr. Frith.
101 953 740 Mr. Frith.
920 23 12 772 25 Mr. Holt.

An intercepted letter

Another short intercepted letter, not in Add MS 4166, is in Thurloe State Papers, vol.7 (British History Online).

Brussels, 3. April, 1660. [N.S.]
The last post failing us, makes me I know not what to think, war or peace. 71 55 45 62 47 8 and 4 must consult how to get 61 50 57 62 72 66 54 61 68 60, else you and many will fail. When I can take a resolution, I will send 45 58 44 50 72.

The following are keys.

f.95 (DECODE R4887)

A short code list, including entries for "Ignatius White", "Duke of Gloucester" (d.1660), etc.

f.96-99 (DECODE R4888)

A numerical cipher (E=23/53/81). Apparently in Dutch: 139(ko: van Spaigny[?]), 138(ko:van Schotlant), 148(Prins van Orange), 150(Stadthouder), 151(Intelligentie), 171(vreede) (annotated "peace" in a different hand); the latter part seems to be a fair copy.

f.100-101 (DECODE R4889)

A numerical cipher (E=25/26/27/28, THE=539): 523(Ch: Stuart), 231(Don Alonso d Cardelay[?])

f.102-103 (DECODE R4890)

A numerical cipher (E=6/50/68/97, THE=43*). "Sr Ph. Meadowes" (Wikipedia). This correspons to my reconstruction "Philip Meadowe's cipher" above.

f.104-105 (DECODDE R4891)

A numerical cipher (E=6/50/68/97, THE=4**). This seems to be the same as f.102-103 (DECODE R4890) above.

f.106-107 (DECODE R4892) ([Edward?] Winslow)

Symbols are letters with diacritics. "Character with Mr. [Edward?] Winslowe" (d.1655), who worked for Cromwell since 1646 (Wikipedia).

f.108-109 (DECODE R4893)

A numerical cipher (E=61/62/63/64/65/66/67/68, THE=554): 626(Ch: Stuart), 627(D. Glocester), 696(Amb^r Lockart). The endorsement does not make sense when read with this key: "308[deteyne] 131[o] 267[been] 104[k] 286[captaine] R. Copy of the Cypher given him may 15 1658" "Gardiner wnts[?] Monsr. Thomas Planipyn a la ... blauch a ... a London[?]".

f.110-111 (DECODE R4894) (James Sharp)

A list of code names, including Ch: Stuart (encoded as Conde). "For Mr. James Sharp (Wikipedia) at Edinburgh", "An ....."

f.113-114 (DECODE R4895)

A numerical cipher (E=39/40/41/42/43/44//45, THE=468).

f.115-116 (DECODE R4896) (George Downing)

A numerical cipher (E=39/40/41/42/43/44//45, THE=4**), "RESIDENT DOWNING". This corresponds to my reconstruction of "George Downing's Cipher (1658-1660) above.

f.117 (DECODE R4897) (Blank Marshall)

A numerical cipher (E=18/19/20/21/22) "Blank Marshall". This corresponds to my reconstruction of Blank Marshall's cipher (1656-1658) above, but lacks a number as high as 173.

f.118-119 (DECODE R4898) (Henry Cromwell)

Instructions for Henry Cromwell's cipher ("L H: Cypher / Lord Henry"), used in f.77-78 (DECODE R4880) (see above).

Even numbers, placed in the line and body of ye writing signify vowells viz. 1/a:2/e:3/i:4/o:5/u:6/y.(Figures are written over the letters in the manuscript.) But ye od numbers placed in like manner signify these letters following viz: 1/l:2/m:3/n:4/r:5/s:6/w So that onely these 12 letters are expressed by numbers, (signifying nothing where they are used otherwise) and all the rest retaine their proper significations The 6 vowells made by any even numbers as aforesaid are distinguished from each other by adding to ye said number, the other numbers respectively sett over them as above, & placing the sotall[?] overhead in the writing as for example, to express a sett any even number in ye line, as 2. then adde thereunto 1, wch makes 3. Wch 3 place over 2 thus 3/2. So that 3/2 signifies a, & doth[?] so 5/4, 7/6 &c. ....

f.120 (DECODE R4899)

A numerical cipher (E=24): 255(Tomlinson), 256(Blackwell). Many names are marked "C", "L", "Lt" (Lieutenant?), "Cp" (Captain?), "m", etc. The alphabet section is the same as Blake's cipher reconstructed above.

f.121-122 (DECODE R4900)

A numerical cipher (E=27): 82(Charles Stuart). "This Cipher is with Jack Crosby".

f.123-124 (DECODE R4901)

A numerical cipher (E=124/144/150/154): 79(L. Protector), 69(Ch: Steward). "Cypher with Mr. Barret".

f.124-125 (DECODE R4902)

A numerical cipher (E=41/42/43/44/45, THE=711): 848(L. Sydney), 853:(Committee of safety), 854(Councell of State), 855(Gen^ll Monk), 856(Ld. Lambert), 868(Ch. Stuart). The DECODE metadata attributes this to Algernon Sydney. This seems different from the one used in a letter of Algernon Sidney from 2/12 April 1660 mentioned above.

f.126-129 (DECODE R4903)

A numerical cipher, with code names (E=95/29): king of England (Mr. karny), Monck M. Gen^l (Mr Masin). "Lord Inchiquin (?Wikipedia) to his Lady" "Inchiquin Cypher with his Lady".

f.130-131 (DECODE R4904)

A numerical cipher, with code names (E=40/19); k. of England (Mr kirque), Monck M. Gen^l (Mr. Mark), similar in form to f.126-129(DECOD R4903).

f.132 (DECODE R4905)

A numerical code, including entries for K. Brittain, D. Buckingham, E. Bristol, L. Balcarres, C. Bampfeild, G. Cromwell, Munk, Sr[?] Robert Murray, D. Glocester.

f.133 (DECODE R4906)

A numerical code, including entries for Lord Lambert, P. Royalle, K. Britton, with substitution alphabet in graphical symbols.

f.134-135 (DECODE R4907)

A numerical cipher (E=15/16/17) in row-block array format (which I called two-dimensional format before), with syllables. Includes entries for Cromwell, Glocester, Lambert, Monke.

f.136 (DECODE R4908)

A short numerical code: 44(Ch: Stuart).

f.137 (DECODE R4909)

A numerical cipher (E=24/69/129/130/182/V), including entries for Barbados, Havanna, King of France, King of Spain, King of Portugal, King of Sweden, King of Polland, King of Scots, but not King of England.

f.139,139*,140 (DECODE R4910)

A numerical cipher (E=5/28).

f.141 (DECODE R4911)

A kind of music sheet cipher in French. Names are represented by a letter placed in a row of five horizontal lines like a staff in a music sheet. For example, "e" written above the first line ("premiere ligne") is "Charles Stuart", but "e" written between the first and second lines ("seconde ligne") represent "Roy de france". Letters written on a line are ignored.

Pour comprendre cette tablature qui nous doit servir de chifres il sera
noter qu'on escrit les lettres de deux sortes a scavoir sur la ligne
ou subs la ligne comme par exemples.

subs la ligne comme [staff example]sur la ligne mesme commes [staff example]
toutes les lettres qu on trouvera de ceste facon sus la ligne signifieront quelque chose toutes les lettres qu'on trouvera de ceste facon sur la ligne peront reputer* pour nulles.

par exemple [staff example]
de toutes ces lettres il n'y a que ce d qui signifie quelque chose du conformite
de la clef suivante et de notte[?] Brabant
[blank staff]
Clef des lettres au
nombre de six diversifie six fois
de[?] suitte de six diverses lignes
premiere ligne [staff example]
don juan (a)/ carastena (b)/ conde (c)/ D. de Clowstre (d)/ Charles Stuart (e)/ d. de jonq (f)

seconde ligne [staff example]
Protecteur (a) Turaines (b) Mazarin (c) Roy de Hongrie (d) de france (e) d Espagne (f)

troisiesme ligne [staff example]
duynquerque (a) Ostende (b) francfort (c) Bruxelles (d) Calais (e) Graveline (f)

quattriesme ligne [staff example]
flandres (a) Artois (b) Engleterre (c) Braband (d) France (e) Luczemburg (f)

Cinquiesme ligne [staff example]
Cavallerie (a) infanterie (b) argent (c) mutination (d) intelligence (e) Londre (f)

sixisme ligne [staff example]
Alumition de guerre (a) vaisseaux de guerre (b) nous unde....

DECODE Database:

Héder, M ; Megyesi, B. The DECODE Database of Historical Ciphers and Keys: Version 2. In: Dahlke, C; Megyesi, B (eds.) Proceedings of the 5th International Conference on Historical Cryptology HistoCrypt 2022. Linkoping, Sweden : LiU E-Press (2022) pp. 111-114. , 4 p. [pdf]

Megyesi Beáta, Esslinger Bernhard, Fornés Alicia, Kopal Nils, Láng Benedek, Lasry George, Leeuw Karl de, Pettersson Eva, Wacker Arno, Waldispühl Michelle. Decryption of historical manuscripts: the DECRYPT project. CRYPTOLOGIA 44 : 6 pp. 545-559. , 15 p. (2020) [link]

Megyesi, B., Blomqvist, N., and Pettersson, E. (2019) The DECODE Database: Collection of Historical Ciphers and Keys. In Proceedings of the 2nd International Conference on Historical Cryptology. HistoCrypt 2019, June 23-25, 2019, Mons, Belgium. NEALT Proceedings Series 37, Linköping Electronic Press. [pdf]

©2010 S.Tomokiyo
First posted on 11 July 2010. Last modified on 8 March 2024.
Articles on Historical Cryptography
inserted by FC2 system