A QUANTITATIVE RECONNAISSANCE
The lack of precision in the generalizations about medieval warfare is very understandable: to list all the battles and all the sieges recorded in all the chronicles and other records of the Middle Ages would be a lifetime’s labour, and even then it would only give us a ratio of the recorded events, which would be biased in favour of battles, since they are more dramatic and larger in scale than most sieges, and so more likely to be written about, especially in the laconic texts of the early Middle Ages. But when it comes to the sieges (and battles) of the Hundred Years’ War, it is much more practical to follow the advice of Jacques le Goff that “one must count what can be counted in the source material”.
In the summer of 2018, in a summer ‘History Research Lab’ at West Point, we set out to do just that. We followed the methodology pioneered by Ronald W. Braasch III in his 2018 article on skirmishes in the Hundred Years’ War. We intentionally restricted the source material to the chronicles of Jean Froissart and his continuators Enguerrand de Monstrelet and Mathieu d’Escouchy, because they collectively cover the whole period of the war in a deliberately consistent way and include practically every combat (battle, skirmish, or siege action) that came to the attention of the authors. We knew we wanted to collect data that would enable us to give quantitative answers to many more questions than just “how many sieges were there”: for example, we also wanted to study the durations of sieges and the rates of success, and to track changes in those and other areas over time. Limiting our dataset to sieges and other actions included in this contiguous group of chronicles minimized the risk that changes in the reality of warfare would become confused with changes in the richness or the bias of the sources. But these chronicles still gave us plenty of data to work with: in the first edition of the Thomas Johnes translations that we used, they fill 25 volumes. Although we relied on the chronicles for most
