The Collation

Research and Exploration at the Folger

Decoding Early Modern Gossip

A guest post by Alicia Petersen

What comes to mind when you think of a coded letter? Political intrigue? Espionage? As the Folger Shakespeare Library’s 2014-5 exhibition Decoding the Renaissance: 500 Years of Codes and Ciphers highlighted, these guesses are pretty accurate. Particularly in the tense political climate surrounding the English Civil Wars (1642-1651), coded letters abounded in connection to plots and conspiracies. While the conflict’s two opposing factions, the Royalists and Parliamentarians, held divergent views on important religious and political issues, there were some similarities: both made use of coded correspondence.

However, in the Folger’s collection we find a different type of coded letter, one that uses code to hide gossip from prying eyes. Our letter (Folger MS V.b.333 (29)), which dates to c.1668, was exchanged between two elite women: Lady Mary Butler, Countess of Arran, and Lady Mary Goodricke, wife of Sir Henry Goodricke, Baronet of Ribston. Both women belonged to politically active families. Our letter writer, Mary Butler, had several infamous family connections. She was the granddaughter of George Villiers, Duke of Buckingham and favorite of James I.1 She was also the niece by marriage of Catherine Howard, Lady d’Aubigny,2 who was a famous female spy during the English Civil Wars. As for Mary Goodricke, her husband would later be involved in placing William and Mary on the English throne.3 Yet despite their political pedigree, our Lady Marys were not encoding political messages in their correspondence. This is what first interested me in Butler’s letter. Rather than using code in the context of politics or adultery, Butler encodes gossip. At first blush, gossiping in code may seem trivial. Yet Butler’s coded letter supports the view that gossip was an important social practice in 17th-century England.

During this period, gossip bore some resemblance to its modern counterpart. The 17th-century link between gossip and intimacy, for example, remains true today. Even now, we rarely exchange secrets with strangers. But 17th-century understandings of gossip also differed from our modern view. Historian Bernard Capp has shown that, in 17th-century England, the word gossip had several meanings, all of which were associated specifically with women. In its most positive form, gossip was used to denote a close, female friend, one who would assist a woman during childbirth. In this way, Capp connects gossiping to community building among women. But Capp also demonstrates that this very aspect of gossip, its ability to create female networks, made it a threat to England’s patriarchal social order. While gossip involved women’s role in determining socially-acceptable behavior, it also carried a negative connotation, alluding to women’s perceived penchant for criticism and meddling. Gossip was considered to be a problematic practice, and it received significant attention in satirical literature.4

The coded letter, for its part, was also treated with suspicion. This is unsurprising given that coded letters often featured prominently in plots hatched during the English Civil Wars. Particularly in the early years of the conflict, spies frequently corresponded in cipher and code. One such spy was Butler’s aunt, Catherine Howard, Lady d’Aubigny, who exchanged ciphered letters with King Charles I during the Civil Wars. Catherine would become notorious for her involvement in the 1643 Waller Plot. This plot was a failed attempt to rally Charles’ Royalist supporters in London. In this scheme, Catherine acted as Charles’ go-between, smuggling secret messages from the King to his supporters by concealing missives in her hair. Despite her efforts, the Plot ultimately failed. But the importance of coded letters in high-profile political dramas did not go unnoticed. The use of cipher was outlawed this same year.5

Catherine Howard, Lady d’Aubigny. This portrait, painted by Sir Anthony van Dyck in c. 1638, displays the beauty and glamor for which Catherine was known. Image courtesy of the National Gallery of Art, Washington, DC.

But it was not only the political elite who favored coded letters. Cryptography (code making and breaking) was a popular subject among the general reading public. A variety of cryptography manuals were published in the 17th century, even after 1643. From Samuel Morland’s A New Method of Cryptography (1666) to John Wilkins’ Mercury (1641), many resources were available to enable friends like Butler and Goodricke to communicate in code.6 The code Butler used (which was masterfully cracked by my fellow Yale graduate student Daniel Gaskell) features unmodified consonants, while vowels were replaced with a random consonant. The substituted consonant was generally a letter with an ascender or descender (think letters like p or b) which were then “hatched” one to five times. The number of hatch marks corresponded to a specific vowel: 1 for a, 2 for e, 3 for i, 4 for o, and 5 for u. This code has the potential to create a polyalphabetic system with 21 possible replacements for each vowel. But Gaskell found that the code’s unmodified consonants made it relatively easy to crack. Butler’s code was perhaps better suited to concealing gossip than conveying state secrets.

V.b.333 (29) p. 1

Chappellesard may the 19
Deare Cousen Mall,
I am uery sorry that you haue miscaried but doe
hope you will haue better lucke with the next
that I may be as you Intended; pray send me word
——————- how you doe and if
your are perfectly recouerd againe which I —–
am shure is wished to you by me; for I am still
the same to you I euer was, and shall be soe, I
desire to know whather you did euer receue a
letter from me whare in I sendtt to you for as
light lockes you Could gett for me, because if you
have sendt them thay haue — not come to my
hands yett; and I feare thay may be lost, I cannot
tell but that a longer letter would be troubellsome
to you att this time and therfore will conClude with
ashureing you that I am Deare Cousen Mall
your affectionat
cousen to seure you
M Arran
I heare that f: husband is comeing
(in code) to court with his wife and so

V.b.333 (29) p.2

[p. 2, continued in code]
by these meanes to raise thare forttune I doe
abhorr him for it and had rather it should
sink than to haue it gotten such a way as
that. It is a sad world now, and meane people
in it I can neuer thinke of him with pationce
when I remember how he has dishonored his
family with doeing soe …
[in Italic hand]
… my seruice to all att
yor house pray burne this and excuse me with
troubelling you with a long letter but I was not
able to hold

However, while they may not have been involved in political schemes, Butler and Goodricke undoubtedly felt they still had good reason to encode their messages. In her letter, Butler expresses a common anxiety among letter writers in this period: the fear that a letter will fall into the wrong hands.7 Butler inquires anxiously after a previous letter that seems to have gone astray. She says:

V.b.333 (29) p.1 detail

…I
desire too know whather you did euer receue a
letter from me whare in I sendtt to you for as
light lockes you Could gett for me, because if you
have sendt them thay haue — not come to my
hands yett; and I feare thay may be lost…

Here, Butler references a past request for a lock of Goodricke’s hair. But Goodricke either never received this request, or her locks were lost in the post. Either way, Butler has a cause for concern; earlier letters have disappeared. In fact, this was a common occurrence during this period. Letters were frequently lost, tampered with, or stolen while in transit.8 What’s more, Butler concludes by instructing Goodricke to “burne this [letter]”. This phrase appears frequently in early modern letters & was evidently often disregarded, as the letters are still with us. James Daybell has suggested that “burn this letter” was not meant to be taken literally but simply emphasized the need for privacy. Both code and “burn this letter” represent attempts to keep certain information confidential.9 We can see this in Butler’s letter: she wanted to restrict at least part of her letter to Goodricke alone, and writing in code enabled her to do so.

This is important because even if Butler’s letter had arrived to its intended destination intact, Goodricke would likely not be its only reader. It was common practice for letters to be read aloud and shared with the recipient’s family.10 Butler would have known that her words might be read by many. As a result, she used code to keep certain information private, known only to herself and Goodricke.

So what was Butler so keen on concealing? Butler’s coded message tells Goodricke of “F’s husband” who will soon be arriving:

V.b.333 (29) details of coded parts of p.1 and 2

(in code) to court with his wife and so
by these meanes to raise thare forttune I doe
abhorr him for it and had rather it should sink
than to haue it gotten such a way as
that. It is a sad world now, and meane people
in it I can neuer thinke of him with pationce
when I remember how he has dishonored his
family with doeing soe…

In Butler’s letter, we see both forms of gossip described by Bernard Capp. Butler is critical of F’s husband, who has violated the rules of polite society. As a result, Butler condemns his dishonorable behavior. But I would like to focus here on gossip’s role in building friendship and intimacy between Butler and Goodricke. As Capp has shown, gossiping was associated with female friendship in 17th-century England. In the case of Butler and Goodricke, gossip is made even more intimate by the women’s use of code. In order for Goodricke to understand Butler’s coded message, the code must have been previously agreed upon by the two women. It constituted a private language that only they shared, and it enabled them to exchange gossip. While any letter can serve as a point of connection between two distant parties, a letter containing coded gossip goes a step further in bonding writer and recipient. If Butler’s letter was circulated among Goodricke’s family, they would have been effectively excluded by their lack of familiarity with the code. By excluding others, the two women solidify their own bond.

This connection between intimacy and coded gossip is supported by other information that Butler’s letter gives us about the women’s relationship. As I mentioned, Butler’s letter references an earlier request for a lock of Goodricke’s hair. Gifts of hair, now associated with lovers, were also common tokens of friendship and kinship in 17th-century England. A loved one’s hair was often made into jewelry and worn as a memento. This trend is on display in William Larkin’s 1618 portrait of Lady Anne Clifford, who is sporting a hair necklace and earrings. Butler’s request for a lock of Goodricke’s hair reflects the affection the women shared and supports a reading of Butler’s gossip and code as further evidence of their close relationship.

William Larkin’s 1618 portrait of Lady Anne Clifford, who is wearing a necklace and earrings containing human hair. Photo courtesy of Wikimedia Commons/ National Portrait Gallery, London

In Butler’s letter, we see code being used in a different context: to communicate gossip rather than to conceal conspiracies or adulterous affairs. But that doesn’t mean that Butler’s coded message was insignificant. By gossiping in code, Butler amplifies the letter’s ability to facilitate intimacy between two women. Her coded gossip becomes an important part of the social fabric of 17th-century England.

Alicia Petersen is a PhD student in the History of Science and Medicine Program at Yale University. Her research interests focus on the history of the body and the history of the book in early modern Europe.

  1. Murtagh, Harman. “Butler, Richard, First Earl of Arran,” September 23, 2004. Oxford Dictionary of National Biography https://www.oxforddnb.com/view/10.1093/ref:odnb/9780198614128.001.0001/odnb-9780198614128-e-58099#odnb-9780198614128-e-58099.
  2. Mosley, Charles. In Burke’s Peerage, Baronetage & Knightage. Clan Chiefs, Scottish Feudal Barons, 107th ed., 1034–35. Stokesley: Burke’s Peerage & Gentry, 2003. https://www.burkespeerage.com/home.php.
  3. Folger Staff and Alison E. Bridger. “Guide to the Papers of the Goodricke Family of Ribston Hall, Nidderdale, Yorkshire, 1639?-1689.” Washington, DC, 2010. http://findingaids.folger.edu/dfogoodricke.xml#overview.
  4. Capp, Bernard. When Gossips Meet: Women, Family and Neighbourhood in Early Modern England. Oxford: Oxford University Press, 2007.
  5. Akkerman, Nadine. “Ciphered Pillow Talk with Charles I in Prison, 1646-9.” In Invisible Agents: Women and Espionage in Seventeenth-Century Britain, 27–63. Oxford, United Kingdom: Oxford University Press, 2018.
  6. Ellison, Katherine. “Millions of Millions of Distinct Orders: Multimodality in Seventeenth-Century Cryptography Manuals.” Book History 14, no. 1 (2011): 1–24. https://doi.org/10.1353/bh.2011.0001.
  7. Daybell, James. “Delivery, Reception and Reading.” In Women Letter-Writers in Tudor England, 127–50. Oxford: Oxford University Press, 2007.
  8. Koyama, Satoshi. “Between Love Letter and Newspaper—The Polish Royal Authority and News Media in the Sixteenth and Seventeenth Centuries.” ZINBUN 34, no. 2 (March 2000): 31–49. https://doi.org/10.14989/48779.
  9. Daybell, “Delivery, Reception, and Reading.”
  10. Daybell, “Delivery, Reception, and Reading.”

7 Comments


  • Thanks for a highly interesting post. It would be good to build more knowledge about the extent to which code(s) of this hatched ascender/descender sort were circulating. I wondered if you might be interested in a similar example in the inscription by one Joyce Swingfeild (on the flyleaf of a copy of de Scudery’s Curiae politiae (1654) in the Plume Library, Maldon) although it is simply playing on a way of writing her name — there’s a plate of this as Fig. 3.6 in David Pearson, Book Ownership in Stuart England (OUP 2021), p. 49. If you come across any similar examples I’d be fascinated to know of them. Tim Underhill, Cambridge UK.

  • Did you notice that the cancelled half-line in V.b.333 (29) appears to read ‘pray send me word ~~by the seruaunt in your letter~~’? I wonder why Mary didn’t want that to be plainly legible…

  • Tim Underhill – Fascinating – the Swingfeild inscription is clearly using the same principle of numbered marks to indicate vowels, but shows three different ways of doing it in the same inscription, none of which quite match how Butler does it. Thank you for drawing this to my attention!

    The idea of using 1 to 5 numbered dots or hatches to indicate the vowels a-e-i-o-u pops up in a lot of medieval manuscripts and even in some early diplomatic ciphers, especially in Spain. (See Meister, 1902, Die anfänge der modern diplomatischen geheimschrift.) It’s a simple enough idea – the sort of thing that would occur to many a bored child in penmanship class – that I expect it has many different points of origin. The Swingfeild inscription is the first time I have seen anything else resembling Butler’s version of this idea, however. Certainly, the Butler/Swingfeild ciphers seem to reflect more of the bored-child-in-penmanship-class influence than the influence of formal diplomatic cryptography.

  • An excellent piece, we are in Alicia Petersen’s debt for this close deconstruction of casual correspondence between two women, with its many covert signs & devices. Quick question: How certain is Alicia’s identification of “Cousen Mall” as Mary Goodricke? And re the letter’s geographical place of origin, identified in the letter as “Chappellesard”, is it not Chapelizod, Dublin? I mention this, as the letter suggested to me Mary (‘Mall’) Villiers Stuart, Duchess of Richmond, as recipient / addressee (“Cousen Mall”), whose several miscarriages (miscarriage, an immediate occasion of Goodricke’s letter) were rather well known & documented in various accounts of her life & activities, activities which included trips from London to Chapelizod, Dublin, to visit her daughter. So you see the (possible) intersections here. I imagine Alicia’s reading of the piece is perfectly acceptable & correct, but I had an entirely different ‘take’ on the piece, at first blush, owing to these other (perhaps merely coincidental) cross-references. All to say, heartily appreciated, we all enjoy these puzzles & challenges, they keep us sharp & on our toes, MEM / Princeton Research Forum, NJ. ///

  • I apologize for my delayed reply! I’m just now seeing these comments.

    Tim- I was not aware of Swingfield’s inscription, but I will certainly check it out! Thank you for the recommendation. I really stumbled upon this letter through archival happenstance so I’m not familiar with other coded letters outside of those already highlighted in secondary sources like Nadine Akkerman’s monograph.

    Elisabeth- I wasn’t able to make this out. Thank you for deciphering it for us! I could see how using a servant as a letter carrier would make sense in this situation. It was not unusual for elite women to send letters via their servants. As mentioned above, the English post could be unreliable; a servant was often a safer bet. This mode of delivery is logical in this case, given Mary’s aforementioned concerns regarding misplaced letters. However, since dispatching a servant with correspondence was common practice, I don’t know that Mary would have felt the need to hide it. Perhaps this was simply an error on the part of our letter writer, and she wished to cross it out. I’m not sure.

    Richard- thank you for recommending Philosophy Between the Lines. I’m adding it to my reading list!

    Maureen- this is as fascinating suggestion. My assertion that Cousen Mall was Mary Goodricke is based on the Folger’s identification of Goodricke as the letter’s recipient. It does seem likely to me that the Folger is correct since this letter was preserved in the Goodricke Family’s papers. But your idea is an intriguing one & certainly warrants further sleuthing on my part. Thank you!


Leave a Reply

  • (will not be published)