Fireside Friday, May 9, 2025 (On Lighter Bows) – A Collection of Unmitigated Pedantry

Fireside this week! First off, it seems like last week’s post on the Hollywood myth of archery volley fire really got out there, so if you are a new reader just joining us, welcome! If you are in to discussions of historical tactics with an eye towards correcting common myths in games and film, you might consider also reading the three(ish)-part series on “Total Generalship” (I, II, IIIa, IIIb, IIIc), which discussed how pre-modern generals could – and couldn’t – actually command an army in battle, or our series on pre-railroad logistics (I, II, III), discussing how ancient and medieval armies managed marching and eating on campaign.

For this week’s musing, I wanted to expand a little bit on an element of last week’s archery post that I gestured at but didn’t really get into, which is the question of cultures with bow traditions, but with meaningfully weaker bows. You will recall that a lot of the math on arrow penetration, range and lethality we were working with came from this older post of mine, which took as its starting point the shooting characteristics of the English longbow and the Steppe composite bow, historically the strongest war bows used. In practice, my sense is that in much of Eurasia, by the Middle Ages at least, most cultures with strong archery traditions were using war bows in that range.

But what about the exceptions?

Here there are two that spring to mind. The first is that it’s not clear that ancient composite recurve bows had the same very high draw weights: Sean Manning has collected the – admittedly sometimes sparse – evidence in a pair of his posts, which are also going to be in this week’s reading recommendations below. The suggestion we might glean from the evidence – and it is only a suggestion, given how thin it is – is that war bows seem to have developed heavier draws over time, with late Bronze and very early iron age bows being reconstructed with draw-weights in the 40-65lbs range, while by the fifth century or so the draw weights push more confidently into the 60-70lbs range. Second, I have seen it generally supposed, that the war bows in use in the Americas were also not as powerful as English longbows or Steppe composite bows, although I can’t put a number to how much less powerful.

We might wonder first why that might be and then what that might mean for our understanding of archery tactics. For the why we are necessarily speculating, but I think the most immediately plausible answer is the tactical and especially armor environment. As Manning notes himself, ancient military treatises that talk about archery often stress rate of ‘fire’ as being as important as raw power and of course the speed of shooting over an extended period is going to be fairly closely connected to the strength required to pull the bow. In short, there are tradeoffs for powerful war bows: ‘heavy’ war bows (in terms of draw-weight) require ‘heavy’ (literally) arrows that can withstand the force of acceleration and more strength and endurance to shoot. So if fighting is happening in an environment where you do not need that power, there would be advantages to ‘lighter’ (in draw-weight) bows.

And here it is worth noting that in early Iron Age (c. 1200-500 BC in the Eastern Mediterranean is the range I mean with ‘early’), armor of any description seems like it might have been a fair bit rarer than it would be in subsequent centuries. The evidence for the armor of regular warriors in the Bronze Age – only chieftains and kings would have worn something like the Dendra Panoply – is extremely limited but also suggests, to me at least, that most soldiers had at most a textile defense, if anything. Egyptian infantry in the Bronze Age, for instance, are – so far as I have seen – almost always depicted as wearing knee-length loincloths, but no helmets and bare chests (and carrying shields). The picture in the pre-contact Americas is a bit more complex – the Americas were really big – but armor was limited to organic materials (as there was no tool metal use prior to contact), which is to say the best available armor might have been in the same basic performance class as a gambeson or buff leather.

And that makes a big difference: you will recall, working very roughly from Alan Williams testing in The Knight and the Blast Furnace (2003), lethal penetration of an unarmored human requires relatively little energy, on the order of 10-30J (generally to the lower end of that range). Adding a thick textile defense might raise the impact energy required up to something on the order of 40-60J (very roughly). An mail with padding or a solid breastplate in iron or bronze might push that figure up considerably and so on. You can see my rough estimates in the chart below:

So in a high or late medieval environment, where a lot of combatants on the battlefield – not just in Europe, but in the Middle East, South Asia or East Asia – are going to be wearing some mix of textile, hardened leather and metal armor, it makes a lot of sense that bows push their impact energies up as high as they can by maximizing draw-weight. In essence, they are ‘keeping pace’ with accelerating armor technology.

By contrast, as I’ve argued elsewhere, the Romans deploying a nearly-all-metal-armored heavy infantry battle line in the second century BC was an innovation. Ancient armies were almost never so heavily or uniformly armored. Even as late as the Hellenistic period, we see reportedly unarmored soldiers (with shields, of course) – lighter Greek thureophoroi, Gallic, Celtiberian and Iberian warriors, etc. – used as ‘line infantry’ holding the line of battle (rather than fighting as skirmishers). In that context, it’s not hard to see why a relatively moderate draw-weight, allowing fore more rapid or prolonged shooting, would be advantageous, especially given that functionally no amount of draw-weight would allow an archer to defeat the real key protection these combatants had against arrows: their shields.

But does that combat environment, with lower draw-weight bows combined with much lighter armor, change our expectations for archer lethality? I don’t think it really does, actually. From a pure model-perspective, we’re still dealing with infantry that generally use pretty big shields, so body coverage is still very high, meaning even most ‘on target’ arrows won’t inflict wounds. But also, we know that Greek hoplite phalanxes, Macedonian sarisa-phalanxes, Roman legions, Gallic battle lines and so on confidently marched into contact against arrow-heavy forces all the time: the exception were horse archers who could, of course, withdraw faster than a shield-bearing heavy infantry formation could advance.

Meanwhile, in much of Native North America, what we see through the limited pre-contact and early post-contact evidence is a fairly normal ‘first system‘ interaction, where ‘battles’ consist of often prolonged exchanges of missiles (arrows, javelins, sling shot) that produce very low casualties. These sorts of low-lethality battles, as Azar Gat notes in War in Human Civilization show up in a lot of cultures. Some of those cultures develop armor specifically for those contexts, but many do not and casualties remain low because the exchanges happen at long distance, often from behind shields, in loose order. The long range makes arrow ‘dodging,’ feasible, while the shields and loose order mean that a very high percentage of missiles are simply not going to hit anyone. Instead, the extremely high casualty rates of first system warfare come from raids and ambushes, where weapons (both contact and missile weapons) are being used at very close range against enemies who have little hope of defending themselves. Of course that is also a military context where a ‘heavy’ draw bow doesn’t offer much advantage: it isn’t going to be appreciably more lethal against an opponent hiding behind a shield or a tree (nor more lethal against an opponent being ambushed at close range), but it will be much more exhausting to use and shoot slower.

In short, then, I find the basic supposition – supported, as far as I know, by the evidence – that draw-weights on bows were lighter in Eurasian antiquity and in the New World quite plausible. It fits with the broader threat environment those bows were operating in that whatever small advantage of lethality might be offered was outweighed by the loss of shooting speed in a low-armor environment. At the same time, that interaction doesn’t seem to have really changed the arrow lethality situation, because shields and dispersion were still very effective ways to limit the lethality of archery.

On to Recommendations:

Of course since I’ve mentioned it above, let’s start with Sean Meanning’s two-part series on bow draw-weights, discussing the archaeological and reconstruction evidence first from the Medieval period (in the first part) and then from the ancient world (in the second part). Sean’s blog, Book and Sword also just generally has a lot of very good ancient warfare material in the back catalog, so give that a look if that is your jam. Sean is an academic expert on ancient warfare (particularly for the Achaemenids, a field that is grossly understudied) with a PhD from the Universität Innsbruck; he will not steer you wrong.

Meanwhile, if you like watching historians comment on the historical context and mistakes of battle scenes in movies, Insider has put together a massive, three hour long supercut of their videos interviewing historians over just that, featuring appearances from friends-of-the-blog Michael Taylor and Roel Konijnendijk, as well as Michael Fulton, Evan Wilson and William Short.

And for more on what is happening in the study of the ancient world, also check out the May 1 issue of Pasts Imperfect, with a focus on Seth Bernard’s recent work estimating the economic role of enslaved laborers in Pompeii, as well as noting a fascinating comparative study of inequality between the Roman Empire and Han Dynasty by Guido Alfani, Michele Boll and Walter Scheidel. This latest Pasts Imperfect also includes a gift/author link (which I’m not sure I can reproduce) to the ‘big splash’ ancient history article of the month, “Punic people were genetically diverse with almost no Levantine ancestors.” That’s not precisely a new hypothesis (as PI notes, Josephine Quinn argued for it back in 2017), but it is exciting to see it appear so clearly in the genetic evidence. We’ll be coming back to Carthage a bit later this year on the blog, but this sort of evidence certainly forces us to rethink the nature of Carthaginian imperialism in the western Mediterranean.

Finally, I suppose I ought to note that I wrote again for Foreign Policy, this time on the role that pro-trade policies – despite elite disdain of merchants – played in Roman success. Alas, this one is presumably behind the paywall, but for those with access, it may interest.

Finally, for this week’s book recommendation, I want to lead with something out of last week’s essay on archer and suggest Michael Livingston, Agincourt: Battle of the Scarred King (2023). Livingston’s core focus here is a revised account of the Battle of Agincourt, in particular shifting the location of the battle, in order to better fit what the sources are actually telling us as opposed to some of the mythology that has sprung up around the famous battle. But that ‘intervention’ is in turn embedded in a book that serves as an excellent introduction to the battle even for readers entirely unfamiliar with it. Indeed, the reader who opens the book expecting swiftly to be seeing the hail of arrows at Agincourt will need to be patient, but their patience is rewarded: Livingston begins with a brief history of the Hundred Year’s War to 1415 which sets the battle in its context and then a careful narrative of the Agincourt campaign in 1415 itself, which sets the stage for the battle.

The stage set, Livingston moves to his main historical intervention (what we call it when a historian is trying to change our understanding of something) which is focused on the location of the battle (and then a bunch of follow-on implications of that). What is really handy for the regular reader though is Livingston begins by giving a capable, serious account of the preexisting consensus, what her refers to as the Vulgata-version of Agincourt, before moving to his explanation for why an alternate site for the battle provides a superior understanding of the sources. Likewise, in his discussions of the battle, he makes clear where he deviates from the Vulgata and I think this is useful for giving the reader a sense of how we make history and why it can shift and change as we come to understand our sources just a bit better.

All of that done, the book can finally close with a vivid account of the battle itself, as Livingston works through what I’ve termed here a ‘physics of the battlefield’ approach, trying to harmonize the often difficult sources with what we can understand about the physical place and the way weapons, armor and of course humans would function in that space. This Livingston mixes with vivid descriptions of the horror and savagery of a battle that was, in the end, decided not by arrows but in the mass press of close combat between armored men struggling in muddy fields.

Livingston’s approach here is rigorous, but the book reads very easily: there’s plenty of moments of engaging color and evocative description, while the core historical arguments are delivered clearly in easy-to-follow plain language, making this book both a fascinating stage in the scholarship of the battle and a good introduction for a reader looking to come to grips for the first time with one of the most famous battles in European history.