Independence of America, declared as early as 4 July 1776, became a real possibility after the victory at Saratoga in October 1777 and the French alliance in February 1778. In 1779, Continental Congress appointed John Adams from Massachusetts as Minister Plenipotentiary for negotiating peace with Britain. In November, for his second time, Adams set out for France and in February 1780, after a difficult journey, he arrived at Paris, where Benjamin Franklin resided.
In June 1780, Adams received a cipher from James Lovell, his friend from Massachusetts and a prominent member of the Committee of Foreign Affairs.
This cipher, often called "Mr. Lovell's cipher", is notorious for puzzling Adams and other correspondents. Adams could not even decipher Congress' later instructions for peace negotiations!
An explanation of Lovell's cipher will be given in a separate article. Let us suffice to give the cipher table below.
Lovell's letter conveyed the keyword "CR", which is the first two letters of the surname of Richard Cranch (Abigail Adams' brother-in-law). Lovell refers to their mutual memory from January 1777 when they, both having been appointed as Massachusetts delegates to the Continental Congress, set out for Baltimore, where Congress convened at the time. It had been only a week since Washington finally put an end to his flight and could score a victory at Trenton. Before then, the British offensive had even threatened Philadelphia and made Congress move to Baltimore.
Apart from the hint of the keyword, Lovell provided virtually no information about the cipher itself. He mentioned "the Alphabet squared as on the other Side" but what he gave was only a fragment, in which only the first two columns were complete with numbers 1-27 and the 27 letters (a-z plus &) of the alphabet, respectively. Other than these, only the first four letters of the third and fourth columns were shown. Regarding how the "alphabet squared" and the keyword should be used, all Lovell said was:
Instead of explaining the cipher, Lovell dwelled on how Adams could select his own keyword.
While Lovell provided a "specimen", it did not help.
Indeed, Franklin himself was puzzled by Lovell's cipher. Back in 1777, Lovell sent Franklin his cipher (sample keyword: CHARDON) but Franklin never used it. Urged to use cipher by the French minister in Philadelphia, Lovell used cipher (keyword: COR) in a letter for Franklin of 4 May 1780, the day he communicated the cipher to Adams. But Franklin could not decipher it. (See another article.)
For some time, Adams was spared the trouble of the cipher. (Lovell's letters to Adams in July, September, and October were in the clear.) In the meantime, Adams offended the French foreign minister Vergennes by his letters in July and, after visiting Holland in August, he decided to reside permanently at Amsterdam. (See Adams Papers Digital Edition for details.)
The first extant letter in which Lovell used the CR cipher to Adams was that of 14 December 1780, in which Adams could decipher only some of the ciphered passages.
Adams seems to have consulted Francis Dana, who had come to Europe as his secretary and was then staying in Paris (December 1780 to April 1781) before his new mission to the court of St Petersburg.
When Adams received this, he asked for a key to the cipher.
Adams also consulted C. W. F. Dumas, an American agent residing in the Hague.
Indeed, Dumas had been writing in cipher to the Committee of Foreign Affairs from 1777 to 1779.
Unfortunately, Dumas' cipher (see another article) was nothing like Lovell's cipher.
Meanwhile, in Paris, Dana apparently requested materials from Franklin. Franklin provided Dana with Lovell's cipher (24 February 1780) and a passage in cipher from Lovell's letter of 4 May 1780 (see another article). Dana's first report was as follows.
Dana refers to Lovell's letter to himself of 6 January 1781, which was brought by Colonel John Laurens. For some time, Dana could not recall the name Brattle that would give the necessary keyword BRA. Now, Dana was personally involved in the trouble. Lovell's explanation in the letter was typically terse.
As it turned out, in two weeks, Dana worked out the cipher provided by Franklin (keyword COR) as well as that used in the letter to himself (keyword BRA).
In the meantime, Adams received a further letter from Lovell in cipher, dated 6 January 1781.
In this letter, Lovell referred to Adams' letter which offended the French court and said "I really am in a Disposition to wish that your Letter of July 27 had no being. I am so much pleased with the motive of it apparent in the 5th Paragraph that I doubly am grieved at the event." (The words in italics were enciphered).
Thus, the trouble continued for Adams.
While Dana was in Paris (December 1780 to April 1781), Dana used a small code in his correspondence with Adams in Holland.
The first use was a code name "Francisco" for Silas Deane in a letter from Dana to Adams on 1 January 1781. In the margin of the letter, Dana wrote "Mr. Searle will give you a key." James Searle was a special envoy for Pennsylvania to obtain a loan in Europe and was then visiting Amsterdam. The code list (see Adams Papers Digital Edition) was in Searle's hand.
The small code contained 28 names. While some names were given code names ("Francisco" for Deane, "Steady" for Adams, "Nestor" for Dumas), others received arbitrary symbols ("AZ" for Congress, "DD" for Franklin).
Its use can be seen in the following letters:
Dana to John Adams, 1 January 1781 (Adams Papers Digital Edition)
Dana to John Adams, 7 January 1781 (Adams Papers Digital Edition)
Dana to John Adams, [31 January 1781] (Adams Papers Digital Edition)
John Adams to Searle, 4 February 1781 (Adams Papers Digital Edition)
John Adams to Dana, 8 February 1781 (Adams Papers Digital Edition)
Dana to John Adams, 12 February 1781 (Adams Papers Digital Edition)
Dana to John Adams, 25 February 1781 (Adams Papers Digital Edition)
Dana to John Adams, 6 March 1781 (Adams Papers Digital Edition)
John Adams to Dana, 12 March 1781 (Adams Papers Digital Edition)
Weber (p.63) states that the code "AZ" for Congress was used in a letter of C.W.F. Dumas to John Paul Jones as early as October 1779 (cf. image here).
On 15 June 1781, Congress resolved to accept mediation of Russia and Austria and appointed additional four ministers plenipotentiary: Benjamin Franklin (in France), John Jay (in Spain), Henry Laurens (held prisoner in London), and Thomas Jefferson (Governor of Virginia until this month). It was in compliance with the wish of France that Adams' power as minister plenipotentiary was shared by four additional ministers. The instructions were sent to Franklin (19 June) and Adams (20 June) in cipher.
The new instructions included a most striking clause to the effect that the commissioners were instructed to undertake nothing in the negotiations for peace or truce without the French court's knowledge and concurrence, and ultimately to govern themselves by their advice and opinion. This was the result of the pressure from Luzerne, the French minister at Philadelphia, and Barbé-Marbois, his secretary. When reporting of the new instructions to Adams, Lovell deplored partly in cipher. Lovell's typical style of "enigmatic" writing is shown in the letter.
Lovell could at least console himself that New England fisheries ("haddock") would be safe as long as Adams retained his older instructions ("other parchments") to negotiate a commercial treaty with Britain but his hopes would be disappointed when it was revoked in July.
Cipher was used in this letter as well as in the instruction itself. Lovell knew of Adams' trouble with the cipher but he only provided a slightly more detailed explanation. Apparently, he thought Adams could not understand his hint of the keyword "CR".
Lovell's explanation was slightly more detailed than before. In particular, he described making two columns, one beginning with the first letter (C) and the other beginning with the second letter (R) of the keyword (CR), looking alternately into the columns, and finding a plaintext letter represented by the figure in the ciphertext. Adams, however, could not fully decipher the instructions. Considering that he could decipher some passages, Adams did understand the system and the keyword.
The following is the text of the new instructions, where the letters in blue are those Adams could read and the letters in red are those he could not.
Upon reviewing the original cipher, some errors are noted.
(1) Adams could correctly decipher the first few lines of the cipher "form & effect of the treaties subsisting between the said United States & his most Christian majesty and in which the said treaties shall not be". At this point, his deciphering stops. As it happened, two letters (corresponding to "le" in "left") were omitted in the ciphertext and this appears to be the immediate cause of Adams's failure of deciphering. When one comes across "ftin..." during deciphering, it would be natural to think that there is something wrong. Of course, a little patience would have revealed a longer sequence "ftintheirfull...", which might have taken him back to the correct track and the problem might have been limited to the two letters. But Adams apparently thought the cipher was garbled.
(2) There are other minor errors. In the fourth paragraph, "dis" as in "distance" is enciphered twice. There are instances such as 3 6 13 (thc), which should be 3 6 15 (the), and 3 16 4 1 13 (trucc), which should be 3 16 4 1 15 (truce). Of these, the latter belongs to a portion Adams could decipher but the last letter was left blank in the interlined plaintext by Adams.
(3) In Lovell's cipher, it is essential to regularly switch between the C column and the R column. However, Lovell omitted such switching once in "truce" and again in "concurrence". To illustrate the former instance of "truce", the R column is used twice in succession.
Though Adams gave up before he came across these errors, this exemplifies a major disadvantage of Lovell's cipher. If one correctly uses alternate columns, this ciphertext would read "tciot". With Lovell's cipher, one enciphering error may corrupt all the subsequent deciphering. It was his wife, Abigail Adams, who conveyed him a practical solution to this problem (see her letter of 17 June 1782 below).
In October 1781, Robert R. Livingston took office as the first Secretary of Foreign Affairs and took over Lovell's task in the Committee of Foreign Affairs. Livingston used Lovell's cipher in his first letters to John Adams but later realized the problems Adams had with the cipher.
Indeed, of the eight passages enciphered in a paragraph of his letter of 20 November 1781, Adams could decipher only the 2nd, 3rd, 5th, and part of the 6th. Again, the failure was caused by enciphering errors.
Lovell, who was still in Congress, possibly consulted by Livingston, sent a renewed explanation to Adams. (Lovell had used the cipher at least once more in his letter of 21 July 1781 to Adams.)
Lovell's explanation was no more than a repetition of previously provided ones. (Further, Lovell said "sixth", when he should have said "third". The keyword "CR" is the first third part of the name "Cranch".)
At the end of the year, Livingston himself felt a need to provide Adams with explanation and enclosed Lovell's explanation.
It was unfortunate for Adams that Lovell had a tendency to repeat rather than elaborate. (For Dumas' cipher mentioned in the letter, see another article.)
It was only on 14 February 1782 that Adams acknowledged news that Livingston took office of the Secretary of Foreign Affairs. Apparently, he shortly thereafter received Livingston's letter of 26 December.
His frustration is evident in stressing that he correctly comprehended the keyword but enciphering did not conform to the rules.
In the meantime, Dana went to Russia with Adams' fourteen-year-old son, John Quincy Adams. Dana sent from Russia a small numerical code to Adams probably in his letter of 8 September 1781, the first letter Adams received from Dana in Russia. It assigned numbers 1-22 such as 8 (Holland or Dutch), 18 (Adams), and 19 (Jay). In order to encode words not given in the small list, Dana also proposed to use a book code based on Entick's Spelling Dictionary (1772 edition), of which Dana said E. Jennings had a copy. (See Adams Papers Digital Edition for details.)
Adams used the small numerical code at least in his letter of 14 December 1781 to Dana (Adams Papers Digital Edition), which was a reply to Dana's letter of 8 September, and on 10 October 1782 (ibid.)
Dana used it at least on 17 December 1781 NS (ibid.), in which Dana said "I want to write to you upon a special matter which it wou'd not be prudent to do, till we have settled our Cyphers. Don't neglect the scheme [i.e., the book code] I sent you. I am convinced it wou'd with a little use, be attended with very little trouble to you." A small code of 22 names was not sufficient for encoding detailed accounts and probably Dana wanted to be sure that Adams had obtained the book from Jennings.
Dana further used it at least on 11 January 1782 NS (ibid.), 25 January 1782 NS (ibid.), 4 March 1782 NS (ibid.), 18 October 1782 NS (ibid.), and 15 October 1782 [26 October 1782 NS] (ibid.). On 18 October 1782, Dana remarked that "There is much curious History in this matter which I cannot go into for want of a Cypher between us." In the end, there is no evidence that the book code was ever used.
Dana appears to have received a full code from Livingston in August 1782 but he was not sure whether Adams had the same code (Dana to John Adams, 4 October 1782 NS) but, finding it "intolerably tedious", proposed a WAR cipher to Livingston and a WHO cipher to Adams. (See another article.)
Weber (p.72, n.38) states, quoting "AP, R358", that the code was also used in Jay to John Adams, 28 September 1782.
Back to the instructions for the commissioners, apparently, Franklin had no trouble in deciphering his instruction (Franklin to the President of Congress, 13 September 1781; see another article). When he learned he was appointed as an additional peace commissioner, he asked Adams in Holland about the proceedings up to then (Franklin to John Adams, 16 August 1781). Adams replied that there had been no move until in July he was invited to Paris by Vergennes and made comments on some articles (John Adams to Franklin, 25 August 1781). Adams considered that the preliminary condition for starting a negotiation was that a representative of America should be admitted to the peace conference. On the other hand, the British government was ready for reconciliation with Americans but refused to join a negotiation in which American independence should be recognized from the start. Thus, a peace conference in Vienna did not materialize in 1781.
In the meantime, Americans gained an important victory at Yorktown in October 1781. This made American independence inevitable. Adams was contacted by an agent of the British government.
According to Adams' report to Franklin (John Adams to Franklin, 26 March 1782), a messenger from the highest authority of the British government delivered a card from Digges, who desired a meeting with Adams. Adams agreed to meet him in the presence of his secretary as a witness and promised to communicate any message to Franklin in Paris and Vergennes. At the same time, he advised Digges to go to Paris and directly talk to Franklin. Digges appeared at the appointed time anyway and revealed that he was sent by no other person than Lord North, the British prime minister, to inquire Adams if he or any other had authority to treat with Britain for a truce. Adams answered that he came to Europe with full powers to make peace; that those powers had been announced to the public upon his arrival, and continued in force until the previous summer, when Congress sent a new commission, containing the same powers, to four persons; that he would not advise the King of England to think of a truce, because it would be but a real war under a simulated appearance of tranquillity, and would end in another open and bloody war, without doing any real good to any of the parties.
Adams said that any person authorized for negotiation should go to Franklin in Paris and if he would come to Adams, he could give him no opinion upon anything without consulting his colleagues. Of course, Adams could not talk because he did not fully know his instructions! In this report, Adams asked Franklin to send the plaintext of the instructions.
Now, his position as peace commissioner was not a mere title and Adams could no longer remain unconcerned about the undecipherable instructions.
Franklin, who had also received a letter from Digges, replied to Adams on 31 March. Because he received Adams' letter when he was writing another letter, he promised to send a copy of the instructions by the next courier. As to the offer from Digges, he told he would wait for some time. Lord North had suddenly resigned on 20 March and he wanted to see what tone would be taken by the new British ministry.
As promised, Franklin sent a copy of the instructions by the courier from Versailles (mentioned in Franklin to John Adams, 20 April 1782) and Adams acknowledged its safe receipt on 2 May.
The new instructions should have revealed to Adams the content which he could not decipher. That is, the commissioners were instructed to undertake nothing in the negotiations for peace or truce without the French court's knowledge and concurrence, and ultimately to govern themselves by their advice and opinion.
Adams' position was the same as Jay's: "to be honest and grateful to our allies, but to think for ourselves." But when Adams arrived at Paris at the end of October, he was told of an interpretation to this part of the instructions that he had not imagined. It was represented by some persons as subjecting the American commissioners to the French ministry, taking away from them all right of judging for themselves, and obliging them to agree to whatever the French ministers should advise them to, and to do nothing without their consent. Adams wrote indignantly about this to Livingston on 31 October 1782.
Upon receiving Adams' letter of 21 February 1782, in which Adams complained about the cipher, Livingston admitted that Lovell's cipher was incomplete.
Instead, Livingston sent Adams a new code. It was based on a regular code in which numbers up to 1011 are assigned to words or syllables. Such a code was less prone to errors than a polyalphabetic cipher such as Lovell's.
While Adams received the code in May 1783, about another year after Livingston first sent it, he did not use it when he wrote to Livingston about negotiations for a definitive peace treaty during June and July 1783. (See another article for Livingston's code.)
When Livingston received Adams' complaint of 21 February 1782, he conveyed it to Adams' wife, Abigail, in Massachusetts. It induced her to write to her husband about the knack of deciphering Lovell's cipher.
Originally, Abigail was as averse to cipher as her husband. When Lovell sent to John Adams in Paris the cipher on 4 May 1780, he also sent the cipher to his wife Abigail in Massachusetts. But Abigail did not like it.
Abigail not only professed her aversion to cipher but also chided Lovell, close friend of the Adamses, for his enigmatical way of writing.
Her husband had also confessed his reluctance to use a cipher. When he was about to return to America after his firs stay in France, he received from Arthur Lee, then in Paris, a proposal to use a cipher and answered "I am no Hand at a Cypher, but will endeavour, to unridel if you write in it." (John Adams to Arthur Lee, 24 March 1779)
Adams' aversion was known to his secretary Dana. When Dana sent Adams a small code (WE073) and a supplementing book code in September 1781, he wrote "[The book code] is at the same time, I think, equally easy [an]d attended with very little trouble. Those [cyphers?] J.L. [i.e., James Lovell] has sent you, are exceeding trou[bleso]me and tedious. I know you dislike [corresp]onding in Cyphers". (Dana to John Adams, [ca. 8 September 1781]) A year later, when Dana sent a cipher (WHO cipher, for which see other articles here and here) from Russia in October 1782, he wrote "I do not expect you will ever write me in cyphers, unless upon the most urgent occasions: nor shall I trouble you or myself in that way upon slight ones. 'Tis best to be armed at all points if possible. With this view alone, having a safe opportunity by your Son, I send you these Cyphers.." (Dana to John Adams, 18 October 1782 [29 October 1782 NS])
As noted above, Lovell started using the CR cipher for Adams in his letter of 14 December 1780. Lovell boasted his confidence in his cipher in a letter to Abigail, in which he probably enclosed a copy of his letter of 14 December.
It was customary that when a packet ship was about to be captured by the enemy, papers were thrown overboard to avoid seizure. Lovell felt frustrated that sometimes valuable letters were thrown away when only chased. But the danger was real. In September 1780, a ship which Henry Laurens, who was to seek aid from Holland, was onboard was captured by the British. Laurens threw papers overboard but the British sailors recovered them. The British found among them evidence of Dutch implications and used it as a pretext for declaring war against the Dutch in December 1780.
In the letter to Abigail, Lovell continued to complain he could not provide her with sensitive information because of Abigail's aversion to cipher and he told her that had it not been for her aversion, he could have enabled her to convey sensitive information to her husband and to read the enclosed letter in cipher.
At this point, Abigail was reconciled to use of ciphers.
Lovell responded by sending the "alphabet."
Lovell enciphered four passages in a letter to Abigail of 26 June 1781, in which he broke the news that four colleagues were appointed to her husband by the pressure from the allies. Abigail's brother-in-law, Richard Cranch, could successfully decipher these passages. Cranch's deciphering worksheet is printed in Butterfield and Friedlaender, Adams Family Correspondence. Cranch marked the four passages in cipher as A-D. The cipher of section B was "1 25 10 22 3 11 5 4 3". Applying the C and R columns of the cipher table of the CR cipher (see above) to this gives:
1 25 10 22 3 11 5 4 3
C & L X E M G F E
R O & L T A V U T
where the upper row is obtained by applying the C column and the lower row the R column. Letters in yellow, found by using alternate columns, indicates the plaintext "colleague".
On the other hand, Cranch's worksheet for this portion is as follows:
r & & l t a v u t
c o l x e m g f e
where yellow highlight is added to show the plaintext.
One can only speculate but perhaps Cranch first wrote "col" by correctly using alternate columns. Then, he felt it was bothersome to switch the columns every letter and start writing both the plaintext letters found from the C column (x e m g f e) and the R column (l t a v u t) on top of each other. After that, he could read the plaintext by picking up letters from alternate rows. (Cranch's worksheet chains letters l-l-e-a-g-u-e with a dash.) This speculation may be supported by the fact that, in the A section, the first four letters "beli" are written with a firm hand in one row and the following letters are similarly chained between two rows.
Lovell used cipher in his letters to Abigail on 13 July, 5 October, and on 4 December 1781. But even Lovell was sometimes tired of the cipher. On 29 November 1781, he declined to convey some sensitive information to Abigail because he was "not in the Humour to use Cyphers."
Abigail soon became accustomed to deciphering of Lovell's cipher. When she received Adams' complaint of 21 February 1782 to Livingston, she told her husband how to handle Lovell's cipher.
Abigail may have learned the trick from the efforts of Richard Cranch. This is illustrated with the previous example.
1 25 10 22 3 11 5 4 3
C C & L X E M G F E
R R O & L T A V U T
Instead of alternately using the C and R columns, the two possible letters that may be represented by each figure in the ciphertext are written on top of each other, regardless of which should be used. Either the upper or the lower letter is picked up to make a meaningful sequence. This way, even if the sender makes an error in switching the columns, the recipient might find out the plaintext. Such an advantage may be seen in the example of "truce" above and will be clearer in a longer ciphertext such as Jefferson's letter to short (see another article) and Dana's letter to Livingston (the cipher used in Dana's letter is not exactly Lovell's cipher but a similar polyalphabetic cipher with a nonce keyword "WAR").
Adams replied to Abigail's letter of 17 June on 17 August. (Abigail wrote two letters on 17 June but Adams' reference to "naval disaster" seems to imply Adams received at least the one quoted above.) In the reply, however, Adams did not mention the cipher. At the time, peace negotiations had progressed so far that Adams expected a full success.
Adams went over to Paris at the end of October and joined Franklin and Jay. A preliminary peace treaty was signed on 31 November 1782.
A previous work on the topic is:
'The Lovell Cipher and Its Derivatives' in Butterfield and Friedlaender (1973), Adams Family Correspondence, IV, Appendix p.393-399. (Adams Papers Digital Edition)