Keynote Speakers

Agathe Couderc, “Swiftness, Accuracy, and Secrecy: the Entente Cipher Bureaus’ cooperation during World War I”

During the First World War, signals intelligence (SIGINT) emerged as a significant tool for informed decision-making. It relied on three main fields: interception of messages, direction-finding, and cryptology, the latter being the speciality of the Cypher Bureaus. Such units were to break the codes and ciphers of the enemy and assumed responsibility for the communication security.

In the early 1910s, France and the United Kingdom dealt with SIGINT in a defensive way: they produced common cryptographic systems to ensure the secrecy of their communications in the impending war. From the outset of the Great War, their Cypher Bureaus worked closely together in attacking the enemy encrypted messages, to ascertain the intentions of the Central Powers. Throughout the war, these Bureaus increased their workforce, their units, and their efficiency.

This keynote aims at encapsulating the main objectives of such a cooperation. The Entente code-makers and cryptanalysts were engaged in a continuous race with the Central Powers, concerning both aggressive and defensive cryptographic issues. Precision was paramount when trying to reconstruct a code book or identify a cypher, in order to provide an accurate insight into the enemy’s situation. Furthermore, discretion on their results was crucial, and they had to trust that their allies would safeguard their secrets. This five-year collaboration played a considerable part in the final victory of the Entente and its allies and illuminates the intricacies of intelligence collaboration within a coalition force.

Agathe Couderc is a Lecturer in Intelligence at the Conservatoire national des arts et métiers (Cnam, Paris). She holds a PhD in Contemporary History of International Relations and Europe (2022, Sorbonne University), and is a historian of contemporary intelligence, especially in the First World War. Her research focuses on the emergence of intelligence services, the introduction of specific secrecy practices, on SIGINT (signals intelligence), and on collaboration in the field of intelligence.

Daniel Larsen, “Codes, Ciphers, and Cultures of Secrecy: The Dawn of US Diplomatic Secrecy during the First World War”

As important as codes and codebreaking are in their own right, the history of cryptology can also provide us with a profoundly useful window into wider cultures of secrecy within government. American diplomatic codes were profoundly vulnerable before and during the First World War – the British had easily solved the US codebooks by late 1915 – but this cryptographic vulnerability offers us a much more complex story than simple arrogance or incompetence, as one might initially suppose. Rather, American diplomats operated in a wider cultural context where secrecy was seen as necessary only for matters of “national defence”, a concept that decidedly did not include diplomacy. Before 1914, Americans actually practiced an extraordinary diplomatic transparency, with the State Department routinely published its own communications for decades. The exigencies of the First World War brought about a new culture of secrecy in American diplomacy, which is well reflected in a shift in State Department attitudes towards their codes and ciphers, and this set the stage for the secrecy practices associated with “national security” today.

Daniel Larsen is Lecturer of Intelligence and War Studies at the University of Glasgow. He previously held a range of fixed-term posts within the University of Cambridge, and his first book, Plotting for Peace: American Peacemakers, British Codebreakers, and Britain at War, 1914-1917 was published by Cambridge University Press. He is a historian of American and British foreign policy and intelligence in the first half of the 20th century, with a particular interest in its political, economic, and legal dimensions, and he has published a range of articles including in Intelligence and National Security, the International History Review, Diplomatic History, and the Harvard Law National Security Journal.

Nagwa M.Metwally, “Cryptographic Writing in Ancient Egypt: A Historical and Functional Study”

The ancient Egyptians developed a writing system aimed at obscuring meaning. The roots of this enigmatic writing can be traced to the early stages of the Egyptian language. Its use continued until the end of the Egyptian linguistic tradition and is attested in the last known hieroglyphic text dated to 394 A.D. Cryptographic writing was widely employed during the New Kingdom and later became a characteristic feature of temple inscriptions in the Graeco-Roman Period. This mysterious style of writing is referred to as “cryptography” or “enigmatic writing”. In Arabic, it is known as al-kitābah al-muʿammah (الكتابة المعماة). The use of cryptography was not confined to temples; it also appeared in tombs and on various objects. The primary purpose of cryptographic writing was to emphasize the significance of specific texts. Certain meanings were rendered in unusual or complex forms within otherwise plain texts to highlight their sacred or symbolic value. Over time, cryptography served additional purposes, including the concealment of magical formulas and the restriction of specialized knowledge related to narcotics and toxic substances. Moreover, individuals employed cryptographic writing to secretly record personal wishes, and in some cases, to write their names in a concealed manner in the belief that this would grant immortality and protection from harm. By the 3rd–4th centuries A.D., this writing system had also been adopted by monks in their correspondence. In addition to the above-mentioned functions, cryptographic writing served a wide range of purposes.

Egyptian cryptography introduced hundreds of new signs, numerous orthographic variations, and unconventional renderings of personal names, place names, and pronouns. These elements operated according to specific cryptographic principles, including direct pictorial representation, the rebus principle, acrophony, the principle of consonants, the interchange of signs, changes in the order and placement of signs, and phonetic alteration. This keynote conference examines selected examples of cryptographic words and writing elements, demonstrating how their multiple phonetic values were transmitted across different periods of Egyptian writing, culminating in the Graeco-Roman era. Egyptian cryptography thus constitutes a significant historical source, offering valuable insights into the cultural, religious, intellectual, and political dimensions of ancient Egyptian civilization, and underscoring its importance for the field of Egyptology.

Nagwa Metwally is director general at the General Administration of Scientific Publishing Suprem Council of Antiquities (Minister of Tourism and Antiquities, Egypt). She holds a PhD in Egyptian Archaeology (2016) from Cairo University. Her fields of expertise include Ancient Egyptian language and scripts, cryptographic and enigmatic writing and epigraphy. She published Cryptographic Writing in Ancient Egyptian Civilization (Alexandria, 2010) and A Dictionary of Cryptographic Writing in Ancient Egyptian Civilization from its Origins until the Beginning of the Fourth Century BC (Cairo, 2023).

Peter Stokes, “Unsolved problems in HTR: Insights from eScriptorium”

There is little question of the very significant advances that machine learning and other digital and computational methods have brought to manuscript studies, of which HTR (handwritten text recognition) is a clear example. Tools are now readily available that can successfully treat many different cases of historical writing, supported by rapidly increasing quantities of training data for different writing systems including Latin, Greek, Hebrew, Arabic, Cyrillic, Syriac, Georgian, Chinese, Japanese, and many others. However, this is by no means to claim that the problem is “solved”, as there are still significant challenges that remain. In some cases the problem is one of training data, and therefore the solution seems clear: we simply need to produce more data. However, this “simple” solution hides a multitude of difficulties. How can we produce homogeneous standard-compliant data for all different types of documents? Is it possible or even desirable to have a single standard of transcription that can meet all the different reasons why one might transcribe a document? Other problems are more technical, such as how to encode all the variety of signs that one might want to transcribe in all the world’s history of writing, how to manage different directionalities and non-linearity in writing. Other challenges go to the very foundation of what writing is, such as how to decide the boundary between the graphetic as the graphemic in transcription. In this keynote, I will therefore present these difficulties as they have arisen in the eScriptorium projects, considering how we have addressed them and what problems still remain.

Peter Stokes is directeur d’études at the École Pratique des Hautes Études – Université PSL (Paris, France), specialising in digital and computational approaches to palaeography. Formerly at the University of Cambridge and then King’s College London, his past projects include DigiPal, Exon Domesday, and Models of Authority, while his current roles include Co-Director of the eScriptorium platform for Automatic Text Recognition, Co-Responsible for Cluster 4 of Biblissima+, and Secretary General of the Comité internationale de paléographie latine.