Adams Family Correspondence, volume 4

Appendix
The Lovell Cipher and Its Derivatives

The earliest instance so far located in the Adams Papers in which a passage in cipher appears is in a letter from James Lovell to JA of 14 Dec. 1780, a copy of which (also in the Adams Papers) Lovell enclosed in his letter to AA of 19 December.1 The presumption raised by the absence of any accompanying key or explanation that there had been some earlier communication in reference to the cipher, and perhaps use of it, is borne out by two letters of Lovell written six months before.2 In the letter to JA, Lovell had written that the cipher was one which had already been “communicated to Doctr. Franklin and which will serve great numbers with equal safety.” Lovell's letter to AA informed her that the cipher was being communicated to her both for her own use in letters to JA and so that she would be able to decode letters written to her in it. No instances have been found in the Adams Papers in which JA or AA employed the cipher themselves, and both continued to experience difficulty in decoding it.

A reluctance to employ the cipher and even to attempt to decode it is attributable to an aversion or hostility to the clandestine implications and attendant ambiguities of secret writing. AA had been clear on the point, and associated her husband with her view, in declining to give her attention to the cipher proffered by Lovell: “I . . . thank you for your alphabeticall cipher tho I believe I shall never make use of it. I hate a cipher of any kind and have been so much more used to deal in realities with those I love, that I should make a miserable proficiency in modes and figures. Besides my Friend is no adept in investigating ciphers and hates to be puzzeld for a meaning.”3

AA, however, with time became “more reconciled to ambiguity and ciphers, than formerly”4 when confronted by repeated instances of intercepted letters and by Lovell's affirmation in his letter of 19 Dec. 1780 that information of importance to JA and to her was being denied her because Lovell judged it safe to communicate it only in 394code: “If you had not bantered me so more than once about my generally-enigmatic manner, and appeared so averse to cyphers I would have long ago enabled you to tell Mr. A some Things which you have most probably omitted, as well as to satisfy your Eve on the present Occasion.”5

Lovell responded to the modification in AA's dislike of ciphers “from the necessity of them” by enclosing another “Alphabet i.e. key for your use” in his letter to her of 30 Jan. 1781.6 Some months afterward, when faced with the cipher in a letter, she managed to read it with substantial help from Richard Cranch.7 Much later, upon having sight, through Lovell, of JA's letter of 21 Feb. 1782 to Secretary Livingston,8 and learning from it that JA continued to be unable to crack the code, AA undertook to induct her “dearest Friend” into its mysteries, reporting, “I have always been fortunate enough to succeed with it.”9

Lovell seems to have been the deviser of the cipher, which he employed in writing to Franklin, to Dana, and to others, as well as to the Adamses. At least it was called “M. Lovell's Cypher,” and Edmund Randolph and Madison acknowledged him as their instructor. The cipher's first recorded use was in a letter from Lovell to Horatio Gates, 1 March 1779; it had currency at least until 1784 in private communications among those in government, and seems to have been at least informally adopted for official use.10 The President of Congress, Samuel Huntington, wrote to JA in the cipher, as did R. R. Livingston after he became Secretary for Foreign Affairs.11 There were variations of Lovell's cipher, other than changes in the key-word, in use among America's representatives abroad. Dumas employed one with which Franklin was familiar, and Francis Dana sent a more complicated system based upon the same principles as Lovell's to JA from Russia.12

Lovell's cipher was of the type built upon the substitution of numerical equivalents for agreed-upon letters from a key-word or phrase. 395Each of the letters adopted for use served in turn as the initial element in alphabets arranged on a sheet in parallel vertical columns. A column of corresponding numbers placed alongside these columns of alphabets provided the equivalents for the letters of encoded words, the substitutions being made alternately from two alphabets, or if more than two then in strict rotation, forward or backward as desired. Since the alphabets normally included the ampersand as a final element, the numbers used in substitution were from 1 to 27. In the cipher's purest form the numbers 28 and 29 were used at the beginning of a passage as indication that the substitutions were being made in the normal or in a reverse order, and the number 30 was used as a blind. However, more frequently all three numbers (28–30) served as blinds. Any uncoded word or words broke the continuity, the next succeeding coded passage beginning with that alphabet used at the outset of the first encoded passage, unless 28 or 29 signaled reversal.

The key letters which Lovell used and instructed others to use in communications with JA, Franklin, and Dana were the letters c and r, which he derived from and clued to (at least for JA) the name Cranch. In writing to Elbridge Gerry, Lovell used as key letters e and o, representing the second and third letters “of the maiden Name of the Wife of that Gentleman from whom I sent you a Little Money on a Lottery Score.”13 Other keys, all employing more than two alphabets, were used in ciphers of the same type by Madison and Randolph (Cupid), Jefferson and William Short (Nicholas), and Lovell with other correspondents.14

The difficulties in decoding experienced by JA, as well as by Franklin and Dana, can be attributed in part to their receiving instruction in the cipher exclusively from a distance. What was easy for domestic users by explanation and demonstration close at hand proved formidable when communicated by mail, which had to conceal as well as explain. This may account for Lovell's adopting for his tutees abroad a simplified form of the cipher in which only two letters were used as the key.

A more serious hazard was that Lovell, the expositor, was gifted with neither precision nor lucidity, and having once formulated his directions, was given more to repetition and even playful variation than to real clarification. Lovell was frequently chided, particularly by AA, for his natural bent toward obfuscation: “If Mr. L——I will 396not call me Sausy I will tell him he has not the least occasion to make use of them ciphers himself since he commonly writes so much in the enigmatical way that nobody but his particular correspondents will ever find out his meaning.” She followed this statement with a sentence she decided not to include in the recipient's copy, attributing to JA similar sentiments about Lovell's style: “I have seen my friend sometimes rub his forehead upon the receipt of a Letter, walk the room —What does this Man mean? who can find out his meaning.”15

Lovell's first instructions to JA, communicated in May 1780, were that the “Mode ... is the Alphabet squared . . . and the key Letters are the two first of the Surname of the Family where you and I spent the Evening together before we set out from your House on our Way to Baltimore. . . . Make use of any of the perpendicular columns according to your key Letters.” To this he appended a vertical column of numbers, 1–27, a second alphabetical parallel column beginning with a and ending with &, a third and fourth column beginning with b and c respectively and carried only through the fourth letter, the rest of the “Alphabet squared” indicated by dots of elision. Aside from indicating that the key letters or key-word could be altered at will and suggesting the means to communicate the new key, there was nothing more. Explaining the cipher just afterward to AA, Lovell, less fearful of interception, added no help beyond using for his illustrative columns two alphabets in which the first letters were c and r. When AA six months later asked for help, he responded only by enclosing the same two alphabetical columns.16 His instructions to Franklin in Paris, which Franklin on request sent to Dana, differed in no essential, concentrating on the selection of the key-word or letters.17 Not until June 1781, when he wrote to JA, “I suspect that you did not before understand it from my not having said supped in Braintree,” did he undertake clarification. This time his column of numbers extended to 30, the numbers 28, 29, and 30 “to be used as Baulks in the Beginning and End or within your Words”; and apparently for the 397first time explained the “rule of Sequence”: “Make 2 Columns of Letters. . . . Begin your 1st Column with the first letter and your second Column with the 2d letter of the Family Name formerly referred to. Go on to &, then follow a b &c. &c. &c. Look alternately into the Columns.”18 When Secretary Livingston unhappily concluded that JA had not understood his earlier letters in cipher, written without awareness of the difficulty, he had Lovell enclose still another explanation: “You are to form Alphabets equal in number and of the same commencement and Range, as the Letters of the first sixth part of the family Name where you and I supped last with Mrs. Adams, and you are to look alternately into these constructed Alphabets opposite to my figures, for the Elements to spell with, some figures however I may have used as Baulks.”19 Nearly a year later Lovell tried again, but in the same language.20 Livingston, meanwhile, had apparently decided to resolve the problem by adopting a different cipher: “I am sorry for the difficulty the cypher occasions you, it was one I found in the Office, and is very incomplete. I enclose one that you will find easy in the practice, and will therefore write with freedom.”21 AA, during the same period, in a brave but misguided moment had decided to try her hand at explaining: “Take the two Letters for which the figure stands and place one under the other through the whole Sentance, and then try the upper Line with the under, or the under with the upper, always remembering, if one letter answers, that directly above or below must be omitted, and sometimes several must be skiped over.”22

More than two years after his first attempt at explication, Lovell wrote to JA: “I have not to this day Information that you comprehend the Cypher which I have very often used in my Letters.”23 Livingston, noting JA's lack of response, concluded earlier that JA did not comprehend “the cyphers. ... I had them from the late committee of foreign affairs, tho' they say they never received any letters from you in them.”24 JA's own allusions to the cipher not only confirm fully the doubts felt in Philadelphia, but also convey an unconcern that must have had an effect there: “Your Plan of a Cypher I cannot comprehend—nor can Dr. F. his”;25 “I have Letters from the President and 398from Lovell, the last unintelligible, in Cyphers, but inexplicable by his own Cypher—some dismal Ditty about my Letters of 26th July—I know not what.”26 “I am on this Occasion as on all others hitherto utterly unable to comprehend the Sense of the Passages in Cypher. ... I have been able sometimes to decypher Words enough to show, that I have the Letters right; but upon the whole I can make nothing of it, which I regret very much upon this Occasion, as I suppose the Cyphers are a very material part of your letter.”27

The frustrations attendant upon the efforts of Congress' committee and of the Secretary for Foreign Affairs to have their representatives abroad master and use the cipher derived not from JA alone. Almost a year after JA reported Franklin's inability to comprehend it, Franklin himself wrote to Dana, “If you can find the Key and decypher it, I shall be glad, having myself try'd in vain.”28 Dana, in turn, seems to have had his own difficulties. To him Livingston wrote, “I need not tell you how impatient I shall be to hear that this has reached you, since I cannot use my cipher, till I receive a line from you written in it, nor can I write with freedom to you, till I have a cipher.”29

As for JA, while it is true that he gave an impression of insouciance in reporting his inability to comprehend the encoded messages, evidence exists that he made some effort to master Lovell's instructions. To Lovell's iteration of the clue to the key letters as the source of his difficulties, JA, with asperity, wrote, “I know very well the Name of the Family where I spent the Evening with my worthy Friend Mr.—— before We set off, and have made my Alphabet accordingly. . . . The Cypher is certainly not taken regularly under the two first Letters of that Name.”30 In the Adams Papers in JA's handwriting and endorsed “cypher” by him, undated, is a complete “Alphabet squared” with a vertical numerical column, 1–27, at left. The square is without indication that the columns in which c and r are the initial letters are more important than any others. A second attempt in JA's hand, also surviving in the Papers, and illustrated in the present volume, suggests one reason why he remained unenlightened. Across the top of the sheet is an alphabet including the ampersand, at the left is a vertical column of numbers, 1–30, paralleled by three columns of alphabets with initial letters a, c, and r. Failing to understand or to heed 399Lovell's belated explanation that the numbers 28–30 were “baulks,” JA utilizes these numbers in the c and r columns to begin a new alphabetical cycle. Thus, in the one column the equivalent of c is not only 1 but 28, that of d is 2 and 29, of e, 3 and 30; in the second column the same holds true for r, s, and t. The application of such a system to the material in code could only produce results unsatisfactory to all.

A further account of the kinds of difficulties that recipients experienced in decoding the Lovell cipher is presented in the Descriptive List of Illustrations in the present volume, Nos. 3 and 4. The discussion there should be read in conjunction with what has been developed here, and the pertinent facsimile illustrations themselves examined.

Examples of other ciphers constructed on systems other than numerical substitution do exist in the Adams Papers. On the various types in use during the Revolutionary period, the discussions of the subject by Burnett and Boyd should be consulted.31

1.

P. 36, above.

2.

To JA, 4 May; to AA, post 4 May 1780, both in Adams Papers.

3.

To Lovell, 11 June 1780, vol. 3:363, above.

4.

To same, 3 Jan. 1781, p. 57, above.

5.

P. 36, above.

6.

Adams Papers. The new key is not now, however, with Lovell's letter.

7.

Lovell to AA, 26 June 1781, p. 163, above; the undated fragmentary sheet containing Cranch's efforts to decode the ciphered passages is illustrated in the present volume.

8.

LbC, Adams Papers; printed in JA, Works , 7:521–530; Wharton, ed., Dipl. Corr. Amer. Rev. , 5:192–199.

9.

To JA, 17 June 1782, p. 327, above.

10.

Franklin to Dana, 2 March 1781, Adams Papers; Edmund C. Burnett, “Ciphers of the Revolutionary Period,” AHR , 22:331 (Jan. 1917); Jefferson, Papers, ed. Boyd, 7:149, 237, 451.

11.

5 July, 20 Nov. 1781, both in Adams Papers; Livingston's letter is printed without indication that passages are in cipher in Wharton, ed., Dipl. Corr. Amer. Rev. , 4:849–851.

12.

Livingston to JA, 26 Dec. 1781, Adams Papers, printed in Wharton, 5:73–74; Dana to JA, 18 Oct. 1782, Adams Papers.

13.

5 June 1781, MHi: Gerry-Knight Collection. Other examples of the cipher in the same collection are in letters from Lovell of 17 June, 13 July 1781.

14.

Burnett, AHR , 22:331.

15.

To Lovell, 11 June 1780, Dft, vol. 3:363, above. The Adamses' judgment of Lovell's deficiencies is amusingly echoed by CFA when in arranging the family's papers he came to read Lovell's letters: “A man whose situation gave his letters unusual interest. Yet he is so crackbrained that his prose is hardly intelligible and his cypher utterly unreadable. This is a great pity. Such half disclosures of the course of things are worse than none at all” (Diary, entry for 3 Jan. 1835).

16.

30 Jan. 1781>, Adams Papers. It should be noted, however, that at the end of his letter to AA of 8 Jan. he said flat out, without alluding to cipher or key: “This Evening four Years ago I passed with you at your Brother Cranche's” (above, p. 63).

17.

Franklin to Dana, 2 March 1781.

18.

21 June 1781, Adams Papers; printed in Burnett, ed., Letters of Members , 6:124–125.

19.

Livingston to JA, 26 Dec. 1781, with enclosure of same date signed by Lovell and attested by L. R. Morris, Secy., “By Order Mr. Livingston,” Adams Papers.

20.

To JA, 30 Nov. 1782, Adams Papers.

21.

To JA, 30 May 1782, Adams Papers; printed in Wharton, ed., Dipl. Corr. Amer. Rev. , 5:459–460.

22.

To JA, 17 June, p. 327, above.

23.

30 Nov. 1782.

24.

To JA, 26 Dec. 1781.

25.

To Lovell, 24 June 1780, LbC, Adams Papers.

26.

To Dana, 12 March 1781, LbC, Adams Papers; printed in JA, Works , 7:377–378; Wharton, ed., Dipl. Corr. Amer. Rev. , 4:284–285.

27.

To Livingston, 21 Feb. 1782.

28.

2 March 1781. Franklin did, however, in the same letter write correctly two short passages in the cipher.

29.

To Dana, 10 May 1782, in Wharton, ed., Dipl. Corr. Amer. Rev. , 5:411–414.

30.

To Livingston, 21 Feb. 1782.

31.

AHR , 22:329–334; Jefferson, Papers, ed. Boyd, 6:x–xi.