Editor’s Introduction: The DMNES staff are super delighted to host a three-part guest blog by Dr. Anna Dorofeeva. Dr. Dorofeeva is a historian specialising in Western Latin book history and culture, and her current work focuses on ciphers and cryptography in medieval manuscripts; you can follow her on Twitter at @LitteraCarolina. In this series of posts, she talks about how personal names were rendered in code form in the Middle Ages.
Secret names: Cracking the medieval code (Part 1)
Dr. Anna Dorofeeva, ZKS Barker Junior Research Fellow, Durham University
Medieval people often needed to write names down. They were important for witnessing documents, recording real estate, noting who had borrowed which book or owned which slaves, and ensuring that communities of people were remembered. And sometimes, names were written down using ciphers.
This was especially common between the eighth and eleventh centuries, when much writing was done in monasteries by both clergy and lay people. In later centuries, such ciphers were dismissed as the result of bored monks playing around. But we’ve recently begun to uncover the importance of marginalia – codes, notes and signs – for understanding the social spaces of monasteries and their wider communities. In such marginalia, we can hear the echo of the words and names of people from the distant past.
So why did these people deliberately try to conceal their names in the early Middle Ages, and how did they do it?
Name that monk:
Ciphered names often appear in colophons: the notes sometimes left by scribes at the end of the manuscripts they copied. Colophons usually record why a manuscript was made, for whom or by whom, where and when, or simply express relief that the long, hard work of writing was finished at last.
The scribe of this manuscript, copied in the late tenth or early eleventh century in Luxeuil, France, left behind a colophon. It appears at first to be gibberish:
Hbfc Stfphbnxs scrkpskt p[er] prfcfptb brchkinb[er]tk mbgkctrk
But this is actually a sentence encoded using a simple substitution cipher, in which vowels are replaced with the consonant that immediately follows them: a with b, e with f and so on. The decrypted sentence reads:
Haec Stephanus scripsit per precepta Archimberti magistri.
This was written by Stephanus at the command of Master Archimbertus.
The main text above the colophon is the Life of St Deicolus (or Dicuil), containing a history of the Benedictine abbey Deicolus founded in Lure, France, until the year 990. It names other abbots of Lure, including Baltram and Werdolphus (Werdulf). Perhaps all these names were an inspiration for Stephanus to record his own, and that of his master?
A similar colophon, using the same code, appears in this early eleventh-century prayerbook. Deciphered, it reads:
Frater humillimus et monachus Aelsinus me scripsit, sit illi longa salus. Amen. Ælfwinus monachus aeque decanus me possidet.
The most humble brother and monk Aelsinus wrote me, may he have boundless health… Ælfwine, monk and also deacon, owns me.
We know that in 1031 or 1032, Ælfwine became abbot of the New Minster in Winchester, in which Aelsinus (the Latin form of the name Ælfsige) was also a monk. Aelsinus therefore wrote this manuscript before this date. Was his use of cipher for the names he mentioned in this colophon an expression of monastic humility?
Interestingly, during the twelfth century someone added feminine endings to many of the prayers, suggesting that the prayerbook came to be owned by a female community — perhaps Nunnaminster, a royal monastery founded by Queen Ealhswith, the wife of Alfred the Great.
Other ciphers were much less formal than these colophons. One example was left by a scribe called Ekkehart the Fourth (c. 980–1056), who lived in the monastery of St Gall in Switzerland (and was the fourth of a series of scribes called Ekkehart). At the back of the manuscript, on a page used for doodles and pen-trials, Ekkehart left a code he called ‘clophruna’, from the Old High German verb klopfon ‘to knock, to tap, to rap’, and the word for ‘rune’. But this ‘knock-rune’ code has nothing in common with runes: it is based on the Latin alphabet. Each letter is numbered according to its place in the alphabet, and these numbers are then indicated by dots: a = 1, b = 2 and so on. 
In the manuscript, the series of dots marked 5-10-10-5-8-1-17-19 can be decoded as E-K-K-E-H-A-R-T (taking into account that i and j weren’t distinguished in the medieval Latin alphabet). This code enabled monks to exchange messages when they were keeping their compulsory hours of silence alone in their individual cells — tapping the messages out on the walls, letter by letter.
 R. Derolez, Runica Manuscripta: The English Tradition (Bruges, 1954), pp. 134–35.