Roman Engineers Didn’t Read Their Vitruvius

Anyone who has looked at fortifications built by the Roman army of the early empire knows that they were stupid about towers. These forts are often generously provided with towers, but those towers don’t stick far enough out from the walls to provide flanking fire against anyone trying to climb them. They provide extra height for fighting and observing, and protection from the weather on cold wet nights, but they don’t let people shoot and throw things at anyone trying to get over the walls between the towers (or sitting at the base of the wall trying to dig into it and pry things out ). The basic idea of how to use projecting towers had been known since the Middle Kingdom in Egypt, and although the Greeks were slow learners by the time of Alexander the Great some of them had understood these principles and even written textbooks. Roman forts became more sophisticated in the fourth century CE as Roman urban society was struggling. I just realized that Vitruvius explained how to use towers tool in his first book on architecture!

Vitruvius, de architectura, 15.2: From the exterior face of the wall towers must be projected, from which an approaching enemy may be annoyed by weapons, from the embrasures of those towers, right and left.

Vitruvius, de architectura, 1.5.4: The distance between each tower should not exceed an arrow’s flight; so that if, at any point between them, an attack be made, the besiegers may be repulsed by the scorpions and other missile engines stationed on the towers right and left of the point in question.

Vitruvius also mentions the one possible disadvantage of projecting towers:

Vitruvius, de architectura, 1.5.5: The towers should be made either round or polygonal. A square is a bad form, on account of its being easily fractured at the quoins by the battering-ram; whereas the circular tower has this advantage, that, when battered, the pieces of masonry whereof it is composed being cuneiform, they cannot be driven in towards their centre without displacing the whole mass.

Since the Picts and the Suebi did not have battering rams in wheeled tortoises, I don’t think this was an issue for architects in the first century CE. More than 90% of existing fortress towers from the ancient world are tetragonal, because it was slower and required more skilled stonemasons to make any other shape (citation on page 141 of Whitehead’s edition and commentary to Philo on Sieges).

Why didn’t engineers in the first two centuries CE employ these principles? The only explanation I have is that those Roman forts were built by Roman soldiers on the model of a camp with wooden or turf walls. If such a camp was reinforced with towers with a framework of wooden posts, you would want to keep the posts inside the fortifications where nasty people could not climb them, chop then down, or lift them up out of their post-holes. As the Roman frontier began to settle in place in the Claudian Army Reforms, soldiers began to build the same type of fort in stone. These forts just had to face barbarians without siege engines, so for hundreds of years they were good enough. But they were offensively foolishly designed.

Help keep my words from being ignored like Vitruvius’ by sharing, commenting, or donating!

Further Reading

Lawrence H. Keeley, Marisa Fontana, Russell Quick, “Baffles and Bastions: The Universal Features of Fortifications,” Journal of Archaeological Research, vol. 15 (2007) pp. 55–95 DOI 10.1007/s10814-006-9009-0

Y. Garlan, Recherches de poliorcétique greque (Paris 1974)

David Whitehead, Philo Mechanicus: On Sieges. Translated with Introduction and Commentary. Historia. Einzelschriften, 243. Stuttgart: Franz Steiner Verlag, 2016.

(scheduled 3 September 2023)