Security Trade-offs in the Longbow vs. Crossbow Decision

Interesting research: Douglas W. Allen and Peter T. Leeson, “Institutionally Constrained Technology Adoption: Resolving the Longbow Puzzle,” Journal of Law and Economics, v. 58, Aug 2015.

Abstract: For over a century the longbow reigned as undisputed king of medieval European missile weapons. Yet only England used the longbow as a mainstay in its military arsenal; France and Scotland clung to the technologically inferior crossbow. This longbow puzzle has perplexed historians for decades. We resolve it by developing a theory of institutionally constrained technology adoption. Unlike the crossbow, the longbow was cheap and easy to make and required rulers who adopted the weapon to train large numbers of citizens in its use. These features enabled usurping nobles whose rulers adopted the longbow to potentially organize effective rebellions against them. Rulers choosing between missile technologies thus confronted a trade-off with respect to internal and external security. England alone in late medieval Europe was sufficiently politically stable to allow its rulers the first-best technology option. In France and Scotland political instability prevailed, constraining rulers in these nations to the crossbow.

It’s nice to see my security interests intersect with my D&D interests.

Posted on January 22, 2016 at 6:44 AM56 Comments


Thoth January 22, 2016 7:09 AM

“France and Scotland clung to the technologically inferior crossbow”

I remembered that crossbows are more technologically advanced and it was the new weapon of European and a sought after one at that time.

“longbow was cheap and easy to make and required rulers who adopted the weapon to train large numbers of citizens in its use”

This is likely to be untrue. Crossbows are easier to train for the peasants since all you need is get them a couple of days training on how to wind the winch and point and shoot. The longbow requires much more dedicated training and efforts for marksmanship whereas for a crossbow, you are expected to just aim and shoot with suppressive fire in mind.

You can think of the longbow as the sniper rifle and the crossbow as the machine gun of that era.

Crossbows are typically deployed in formation for suppressive fire (although the longbow could also achieve that) but the rate of fire, ease of equipping the peasants, short training needed, cheap peasant labour are more advantageous for massive deployment. The quality of a crossbow may not need to be of good quality since you need to mass produce but a crossbow as a fine weapon might need to be carefully crafted to fit it’s user.

M@ January 22, 2016 7:17 AM

Longbows were not the sniper rifles: They were used en masse to cover large areas with “from above” missiles, which is why you needed to train a lot of people on their use, to effectively defend an area or structure. Crossbows were near-useless for long-range combat, and required you to actually pick a target.

All that said, I still thing the authors are very off that France, etc. were “stuck” with “inferior” technology because of some arbitrary stability metric. If I had to guess, it probably had more to do with British fondness of tradition, and less about any quantifiable reason.

Olaf January 22, 2016 8:10 AM

I thought the long held wisdom of this was that it took years of practice to be competent, and strong enough, to effectively weild a long bow.

Longbow men were to a degree deformed by the stress and training of using the bow. It needed a ruler that would force men out of the fields to train.

The crossbow could be used by anyone stong enough to hold and cock it which would have been most men of that era. and while they did need so much training they could be out doing other things.

A bowman could fire more arrows than a crossbowman.

Also I think (I may be corrected) that a crossbow quarrel is easier to make and less fragile than a longbow arrow.

Juhani January 22, 2016 8:13 AM

I remember reading why Tallinn decided to go with crossbow.

The question was price for ammunition.

City had to stockpile a lot of ammunition for defence. Crossbow was at the time very new and expensive weapon, but price of ammunition changed it to be a cheaper system. If I remember correctly the document had exact price calculations, I do not remember if I saw also the original and on which Tallinn museum.

So yes, it’s also about offence or defence, but the reason is price.

Rich January 22, 2016 8:23 AM

I’m not going to read the evaluation of different countries’ political stability. But I am curious: England suffered many internal conflicts in the late medieval period, from the revolts of the late 12th century eventually leading to Magna Carta, the Barons’ Wars in the early 13th and the Welsh revolts (Welsh longbowmen were legendary) later than century, the baronial revolt against Edward II in the early 14th, the peasants’ revolt of 1381, Henry IV’s coup at the end of that century, then the Wars of the Roses in the 15th.

I mean, I know Italy wasn’t a bed of roses in that period. In fact, Italy wasn’t a country at all…

paul January 22, 2016 8:27 AM

It sounds also like a classic case of investing in capital vs investing in labor. If the crossbow is expensive but easier to use, then you can treat your common soldiers as interchangeable and disposable, whereas longbow archers are going to be more like the skilled artisans whose organizations changed the social order over the long term. (Of course, that argument has also long been propagandized as the “british yeoman” ideal.)

The two weapons also call for an entirely different set of tactics because of their respective ranges. A bunch of longbow archers can stand off and decimate an army using crossbows while suffering minimal danger themselves. Crossbows are more effective than longbows at shorter ranges.

Chris January 22, 2016 8:33 AM

In a long war you lose many of your experienced fighters. Longbows require much longer training times than crossbows. Setting up a new army is faster with crossbows than with longbows. The English were able to use the longbow because their longbowmen were not killed off in long wars, since as the article quotes “England alone in late medieval Europe was sufficiently politically stable.”

Wm January 22, 2016 9:00 AM

“These features enabled usurping nobles whose rulers adopted the longbow to potentially organize effective rebellions against them.”

The exact reason our forefathers created the second amendment, to stem out of control government. And the exact reason why socialist democrats fear U.S. citizens owning guns and want to ban or greatly limit them.

Rich January 22, 2016 9:02 AM

Hmm. England fought lots of wars in the period during which the crossbow declined in popularity ( so I would be interested to see any correlation between casualty rates and the attractiveness of the weapon. (The late middle ages also saw the Black Death – not exactly great for maintaining your skills base…) Many of those wars were internal to the British Isles, which makes me dubious about the “stability” claim.

Crecy (1346) seemed to be a critical advertisement for the longbow. You can imagine how tough it would be to discount it (and the training it needed) in England after that victory. Hence the 1363 requirement for all men to practice longbow regularly. But the Assize of Arms ( had already done that in 1252, of course – hardly a time of massive peace and stability in England.

This dissertation isn’t brilliantly written, but shows the assumptions in the OP seem a bit off… Essentially, they conclude each weapon has trade-offs, which is why it was perfectly viable to have forces including either or both.

Rich January 22, 2016 9:07 AM

(Holy heck. Did someone from the gun lobby actually just use the GOVERNMENT-MANDATED training and ownership of a weapon by its 13th century SERFS – who could be press-ganging into military service to die for the interests of the 0.1% – to support a citizenry armed AGAINST its government in the 21st century? You know, discounting everything else, that’s just plain bad history and logic, right there.)

Winter January 22, 2016 9:32 AM

I remember that is was argued that cross-bows were better if you defended a stronghold, say, a castle. Longbows were more useful in the field, infantry wise, as you could fire faster rounds.

There was also the question whether a longbow arrow could penetrate body armor. Crossbows could, but I am not sure about longbow arrow.

Clive Robinson January 22, 2016 10:09 AM

The long bow has advantages in range, if and only if you used the right wood in the right way. Thus they were not as cheap a weapon as many think, the wood also requires the correct seasoning and heat treatment. But importantly they could be made with a hand axe, knife and a small fire likewise the arrows, unlike the crossbow and it’s bolts.

The longbow also had a significant advantage in the rate of fire. Good longbow men could get an arrow off every ten seconds, crossbows took upto a minute to reload and bring back to the aim. Importantly a crossbow required a break in stance which ment that aim acuracy quickly diminished as did agility.

Thus when it came to rate of fire the longbow outclassed the crossbow by a very long way.

Then there was the effectivness of the ammunition. The arrow of a longbow was considerably longer could carry a heavier tip and was considerably more stable in flight and had less wind resistance. Thus they could deliver rather more of the energy into the target than the crossbow bolt at any sensible range.

Further the longbow could be used repeatedly not just staticaly but on the run and on horse back, something the crossbow could not do. The crossbow was in effect a one shot weapon if not being used staticaly.

But there were other problems with crossbows, they were not reliable in the release mechanism, and if the bowstring broke which they did rather more frequently than the longbow due to the tensions involved it was well nigh impossible to restring on the battlefield. Another problem was the length of the release stroke, it was about a yard in a longbow but only six inches to a foot in crossbows, which means that when it came to energy transfer the longbow gave a better performance especially with heavier arrows. The crossbow due to the short stroke tended to damage the bolt befor it had started moving effectively thus bolts were split and had very poor aerodynamics which ment a further reduction in fire power.

But there are other factors the longbow arrow could be fired so it provided indirect fire that descended onto the enemy as well as directly fire at them, which gave problems with the use of defensive shields use by the enemy in that they needed two sets. Longbows could be fired over shield walls easily not so crossbows and crossbows due to a number of factors were effectivly usless at anything other than direct fire.

But if you look at battle field weapons one of the prefered weapons of mounted knights was the mace or war hammer, lances were single use and swords not much use in either a press or melee. The reason is they are easy to swing down on the heads of infantry that were at best weakly protected. If you examine the tip of an arrow and the spikes on a mace or war hammer you will see they are very similar. Thus a curtain of descending arrows was more leathal to soldiers than several hundred mounted knights. Crossbows due to direct fire would most likely miss the head and hit the much more heavily armourd body. Interestingly layers of leather and wool in tightly stitched layers in the form of a jerkin were very much more of a defence against crossbow bolts than just beaten plate armour or chain mail.

The advantage of the crossbow as far as the nobility were concerned was that it took rather more than an axe and knife to make one, thus they could control the supply of arms held by those of low status, which ment politicaly the nobles could repress those of low status harder without fear of kickback.

Clive Robinson January 22, 2016 10:20 AM

@ Bruce,

If you actually want to get to know more about longbows a world renowned expert is the actor Robert Hardy, who wrote the definitive work on the subject,

Amongst many other things he is a trusty of the Mary Rose Trust and I met him several years ago through that, and he was very gracious in talking to me at a social function about longbows and their history.

As you might or might not know the recovery of the Mary Rose from Portsmouth sound gave many barrels of Tudor longbows that provided much unknown information on their capabilities.

r January 22, 2016 10:39 AM

There’s definitely a greater level of control and freedom available to a long bow, are you only comparing the European crossbow though? I saw something about the Chinese metal receiver type being much stronger and faster to load/aim… I think that one was fired from ones back? The casted receiver also allowed incremental design improvements, were the European ones all carved wood?

Peter January 22, 2016 10:39 AM

Bruce Schneier:
It’s nice to see my security interests intersect with my D&D interests.

What’s D&D? Dungeons & Dragons?

Patrick von der Hagen January 22, 2016 11:30 AM

Reading the comments leaves me with the impression that there are two ways to shoot a longbow: aim it at a target, like you would with a crossbow (sniper-mode), or having dozens (hundreds?) of longbows fire in the air to have the arrows dropped in an area (more like shotgun mode?).
I suppose aiming a longbow at a target is likely to require more training than using a crossbow, but how hard can it be if you simply get the order “fire in the sky so that the arrow drops in the field over there”?

So, which way was more common in medieval Europe? Anybody has any insights?

gollum January 22, 2016 11:41 AM

For a nice engineering exam of bows and crossbows, have a look at cap.6 of “The Science of Structures and Materials” by James E. Gordon.

Some thing to consider: longbows must be used standing, crossbows don’t, traditional yew bows suffers hot climates, naval use of crossbows…

Naval use was important for Italian “repubbliche marinare” (Genoa was renowned for crossbowman).

I am quite skeptical about the longer range of long bow. An article, from Scientific American (Fowley, Palmer, Soedel) gives these figures:
longbow 200 m., crossbow 250 m., ballista over 300m.

Big troubles for crossbow was to adjust the string (tendon) tension when humidity did relax it, so big troubles for rainy days battles, like at Crecy… maybe a big reason to prefer bows or crossbows was climate: hot and dry? crossbow, variable, wet and cold? bow…

IIRC crossbow were limited in England by Magna Charta, maybe because better suited for sneaky usage?

Fred P January 22, 2016 12:58 PM

@Rich – “Late medieval France was far from the stable and dominant world power it
would become in the early modern period. Between 1100 and 1450 France’s bor-
ders expanded and receded like the tide, with independent states pocketing the
continuously changing territory of the country like tidal pools in the sand.” pg 17,

Dr. I. Needtob Athe January 22, 2016 1:24 PM

These posts remind me of the old tale that a frog dropped into hot water will jump out, but a frog dropped into cold water that’s slowly brought to a boil will stay and die. (It was brought up in the movie Dante’s Peak, which is about people who lived at the base of a volcano and didn’t want to leave when it started acting up.)

I hope there weren’t too many frogs boiled alive to test that, because it doesn’t really matter what a frog would do, nor does the history of archery really matter here. I think the subject raised here was “Institutionally Constrained Technology Adoption”, and I’m wondering if it’s a real phenomenon. If so, how much of a problem is it?

JB January 22, 2016 1:41 PM

Path dependency.

“What’s the best way to train a longbowman?”

“Start with his grandfather.”

If you don’t have longbowmen, it takes decades to develop a good supply. You can have a new crossbowman in a week.

If you do have longbowmen, your supply of potential crossbowmen is reduced, and the remaining untrained people are better used by giving them pikes.

AJWM January 22, 2016 1:55 PM


That article seems to severely underestimate longbow ranges. Wikipedia says this:

The range of the medieval weapon [longbow] is not accurately known, with estimates from 165 to 228 m (180 to 249 yds). Modern longbows have a useful range up to 180 m (200 yd). A 667 N (150 lbf) Mary Rose replica longbow was able to shoot a 53.6 g (1.9 oz) arrow 328 m (360 yd) and a 95.9 g (3.3 oz) a distance of 249.9 m (272 yd). A flight arrow of a professional archer of Edward III’s time would reach 400 yds. It is also well known that no practice range was allowed to be less than 220 yds by order of Henry VIII.

Early crossbows had a point-blank range of about 70 yards, but indirect fire (45 degrees up, unaimed) could have been two or three times that. By the 15th Century, steel crossbows had ranges over 300 yards, but by then firearms were starting to replace them (hand cannons circa 14th century, arquebuses in the 15th).

A Mora January 22, 2016 2:08 PM

Based on the audience participation Bruce should look into setting up a “Schneier on History” blog so we can debate the finer points of the long running Mace Vs Maul debate of 1503…

For my part, I will cite the Battle of Agincourt as proof that Longbows were effective against armor. I also point out that the arms v armour (Shant use the American English for a discussion of the pre-colonial era) technology race continued from the short bow and spear, through the longbow, heavier and stronger crossbows, early matchlocks, wheel locks and flint locks. Eventually a suit of armor that would provide full and effective protection from these threats became so heavy and expensive that body armor went into decline. Armies began trading coverage for mobility, and being able to equip and train more troops for the Crown, Sovereign or shekel.

English Longbows are also really hard to make, and require slow growing woods that were a limiting resource. (Check “The art of making primitive bows and arrows” by Waldorf). Crossbows are easier to sight, can be held at ready without exhausting the archer, and can have much higher draw weights then any conventional bow. They also are slow compared to a competent archer. Reloading a crossbow required a variety of methods like like using your foot in a stirrup at the end of the bow, a winch, or a hinged device called a “crow’s foot”. Longbowmen can walk and fire a shot every few seconds, something that only light draw weight crossbows can do.

There are some interesting differences in ammo. Arrows need to be matched to their bow to work well. An arrow for a bow of a different length or draw wight will not shoot straight. Arrows are also longer and more expensive to make. Bolts are are shorter, and cheaper, so in army level volume they may tipped the scales in favor of the crossbow.

Both can be used for direct or indirect fire. That is more a matter of how far your enemy is. Direct fire is generally much more effective if your opponent is close enough, though the archer will be aiming in either case. Contemporary “Roving archers” will still hit a man sized target using indirect fire. Indirect fire is also what a line of archers would use against a formation of troops, but with the advent of decent shields (Back in the B.C. era) it was not unsurvivable.

Thomas January 22, 2016 2:54 PM

It’s nice to see my security interests intersect with my D&D interests.

(Jester fails savings throw against surprise, jaw hits the floor)

… political instability prevailed, constraining rulers in these nations to the [inferior technology]

I’m picturing some historian some hundreds of years from now trying to figure out why we persisted with AV/Firewalls/Cloud/Windows for so long coming to the conclusion that it was a $TLA plot to keep the serfs manageable.

wumpus January 22, 2016 3:27 PM

@KOLO: Aaand it turned into a nerd discussion about weapons, Americans favorite Hobby, hahaha!
Turned? When the article specifically stated it was chosen to match D&D interests?

@JB “Start with his grandfather.”

While I’m inclined to believe this, it doesn’t seem to match typical warfare requirements of the time. Basically archers had two types of targets:

Shooting a pike square (less useful for shooting mounted knights (at least the horses), but I’m sure they tried that as well).
Shooting an individual target.

As far as I know, the vast majority of individual [targeted] shots taken by English(/Welsh*) archers were going to be at the King’s deer**. It would certainly help to have sharpshooters in your army, but it wasn’t what you needed in a typical archer.

Shooting an arrow every 10 seconds took practice. I’m reasonably sure I could hit something the size of a pike square after a week of Boy Scout camp (of which archery merit badge was a small part) [and no, I never practiced for speed. The youtubers I’ve seen doing ‘fast shooting’ don’t appear to aim at all]. The other question is just how much time Yeomen were expected to practice archery. Golfers love to claim that a Scottish king banned (or maybe just “cried down”) golf for interfering with archery practice, but I have to wonder just how much the lost/broken arrows cost. Such a cost must have been non-trivial for peasant yeomen (assuming they practiced a lot on anything other than the King’s deer. My guess is that at least of of those was not true).

  • let’s acknowledge the heroes of Agincourt regardless of Shakespeare’s words.
    ** Bows were used on other game, but you don’t really need a longbow (and the draw weight needed for military use) for rabbits.

r January 22, 2016 4:27 PM


Or conversely, a halberd. Probably expensive but it’s one the anti-horsemen weapons I think the Scotts developed.

Buck January 22, 2016 5:45 PM


Lol! That took no time at all. 😛
Funny how much more comments there are about ancient warfare than the new shiny stuff, isn’t it?

The good (even if particularly pungent) doctor tries to right the course, but alas, no dice…

While off the top of my head, I can’t recall any specific examples of “Institutionally Constrained Technology Adoption” – surely it’s at least somewhat related to the Brain Drain effect (of which there are multiple contemporary examples). Computation machines and nuclear weaponry are interesting to think about in this regard. Not really a case of the institution opposing adoption of technology itself though… More about other policies of the rulers that discouraged opportunities for initial research and development. Then there’s the first mover advantage that probably applies just as much to warfare as it does to business. Hmmm…

tyr January 22, 2016 7:57 PM

You might want to keep the salt around when listening
to moderns debate the differences between weapons they
have never used much.
Crossbows can be used from horseback, you can use a
longbow from a horse with extreme difficulty and not
well. You can also shoot a longbow from a kneeling
position, but you lose the adavantage of your sight
picture (no altitude referent). Plunging fire works
better because gravity helps your arrows along. Most
crossbow bans were because the bolts could penetrate
armor quite nicely if used as an assassination tool.
An assassin carrying a longbow sticks out like a
sore thumb.

A 90 lb draw longbow has the strike power of a 30-06
at the shorter ranges and can do better penetration
in tests using wood.


First Diplomacy, now D&D one would suspect you of
having a Munchkin deck or two hidden away as well.

Joe K January 23, 2016 12:30 AM

@Dr. I. Needtob Athe

I think the subject raised here was “Institutionally Constrained
Technology Adoption”, and I’m wondering if it’s a real phenomenon. If
so, how much of a problem is it?

Some institutional constraints that come to mind are the Microsoft mantra of “Embrace, Extend, Extinguish”, state-sponsorship of the business model of (so-called) traditional academic publishers, and more generally all things “Intellectual Property”.


Come on, it isn’t that nerdy. There have been no heated arguments about ranged weapons in NetHack.

Joe K January 23, 2016 12:56 AM

@Dr. I. Needtob Athe

These posts remind me of the old tale that a frog dropped into hot
water will jump out, but a frog dropped into cold water that’s slowly
brought to a boil will stay and die.

Maybe I misunderstand your point, but I don’t see anything slow-moving about the war on cryptography.

(It was brought up in the movie Dante’s Peak, which is about people
who lived at the base of a volcano and didn’t want to leave when it
started acting up.)

Gah, disaster movies. Blech!

Even if pulling up stakes and moving someplace else were an effective solution, we can’t all live in Iceland. (Which, of course, is literally volcanically active.)

Marshall January 23, 2016 12:38 PM

What Clive said. Norbert Elias is an extended interpretation of the political situation during the period of state formation.

Note that common yew is an excellent bow wood, and bows of shorter range are useful for taking deer and smaller game. So bow-making skills would be widespread among the crofters. Perhaps bowmen could even have been expected to supply their own equipment.

supersaurus January 23, 2016 5:23 PM


A 90 lb draw longbow has the strike power of a 30-06
at the shorter ranges and can do better penetration
in tests using wood.

I don’t know what “strike power” is, but comparing energy a standard hornady 180gr 30-06 round has over 3000 ft-lb energy at the muzzle and a 400gr arrow at 400fps (regardless of what it is shot from) is about 142 ft-lb, not even close. an unscientific google query says a 30-06 will penetrate 1/4″ steel (various alloys) or more; I couldn’t find anybody dumb enough to waste arrows shooting at steel plate.

Joe K January 23, 2016 10:25 PM

One conceivable explanation that occurs to me, for use of longbow in
Britain vs use of crossbow on the European continent, is that Britain
was not entirely subjugated by the Romans. So many local cultural
practices that were erased on the continent by Romanization survived
in much of Britain.

One obvious example is language. The French speak French, not Gaulish,
whereas Welsh survives to this day.

My cursory (and admittedly rather uninformed) impression of most
discussion regarding the choice of longbow vs crossbow is that such
discussion attempts to answer the question of why various military
authorities judged one or the other weapon to be more effective in
some objective sense.

This assumes that self-appointed authorities had a relatively free
hand to choose between the two, and were free to exercise rational
choice in the matter.

But existing cultural practices have a say, as well, don’t they?

I am inclined to suspect this factor is being underestimated.

Gerard van Vooren January 24, 2016 2:25 AM


Please use the metric system for math. We don’t live in the 18th century anymore, at least outside of the US we don’t.

tyr January 24, 2016 7:37 PM


It is energy delivered into the target. It is opposed to
blowthrough in which the excess energy carries the
projectile past the target. That size arrow will get you
a faceful of wood splinters. Longbow arrows are heavier.

I think it was Howard Hill who used to kill elephants
with a longbow. You can do it with a 30-06 but you need
to be an excellent shot and know the animals anatomy.
Most people prefer a bigger gun for elephant.

Calculations are nice but personal experience is a lot
better guide to how things work. American modern use
of the bow only came into vogue after Isshii was
discovered, even though the Indians were still doing
it to a limited extent. There used to be a couple of
arrow types called battleshafts and forgedwoods that
were compressed by machine to get the density to
withstand the stresses of the newer high powered bows
and the heavier longbows.

The local SCA types tried using padded arrows and low
powered (25 pound draw bow limited to half draw) but
gave up when armored victims were hurt badly and had
ugly bruises from a hit on the armor. It had seemed
like a good idea when they thought it up, but it was
a nasty experience.

JG4 January 24, 2016 8:43 PM


Your skepticism regarding youtube is a virtue. I saw the rebuttal video come up in the queue and flagged it to watch later. I’ll concede that there is some hype in Lars’s video, but it still is an impressive demonstration of skill. It also clearly shows at least one advantage of regular bows over crossbows. I suspect that in times past, when people practiced regularly, similar skill levels were reached by more people more often. Another ancient weapon that has performance levels not appreciated by moderns is the sling. It can deliver projectiles at nearly 200 miles per hour with accuracies better than a few inches at 30+ yards.

Chris S January 25, 2016 12:02 PM

You want “Institutionally Constrained Technology Adoption”?

It’s widely accepted that GCHQ (in the land of longbows…) developed first the concept, and then published details of a possible implementation, for both RSA and DH, well ahead of the the discovery in the academic community.

Rather than leverage the potential of these discoveries, they were both classified, and – judging from this end of history – ignored.

trsm.mckay January 25, 2016 7:33 PM

I used to present history seminars annually at a gaming convention (as part of a SCA presentation), and probably the most popular talk was Crossbows vs. Longbows. The Hardy book (mentioned @Clive) was one of the best sources, and Clive’s posts jive with my understanding. One of the points about crossbow being harder to make, is that it makes them easier to control them physically (e.g. lock them up in an armory and only give them out to the local militia in times of need as determined by the person who controls the armory). Here are a number of other points I would like to mention or expand on:

One of my favorite contrasts was that bows were illegal in most European countries (partly because of militarily concerns, but mostly because they were a poacher’s weapon and they wanted to keep people from being proficient). In England they have a different trade-off for bows despite the poaching concerns; but as crossbows became more common they were made illegal for the lower class because of poachers.

Also note that while it is true that the best of the long bows depended upon rare and specialized wood (ideally yew with a mix of barkwood and heartwoood), lesser weapons could easily be made with materials more ready at hand. Hardy has a section detailing early bows c1100 (probably made of Elm and Hickory) that probably had draw of over 120 pounds.

Clive’s otherwise pretty through set of posts did not include an interesting point about compensating for weather (though @gollum called it “string tension”). I think is better explained as the inability to overdraw a crossbow, because of its fixed anchor point. If the bow stave is less effective (earlier crossbow prods were mostly made of composite horn with primitive glue, not good in rainy weather). So unlike a bow, the fixed anchor point means you can’t overdraw to compensate for the reduced power.

Speaking of drawing, which have implications for range and penetrating power, you need to be a little careful because crossbows come in many different sizes. The quicker crossbows (using stirrups and/or levers) are pretty lightweight, more useful for hunting than war against armored opposition. The heavy crossbows (like the German 1/2 and full bow) were manned by teams of people up in the 400-500 pounds of draw range. Also note that horn-composite prods are about 70% efficient, while the iron prods used in heavier weapons were only 60% efficient (meaning they required more poundage just to break even).

From the SCA (a medieval interest group that does hands-on anthropology), I have some pretty good comparisons between longbows and light-crossbows. At typical draw lengths (and ignoring mass), it takes about 70-pounds of pull for a crossbow to have the same initial energy as a 30 pound bow (with typical drawlengths). I am quite familiar with these values because they are also the upper limits as dictated by the SCA for their combat recreations (using cushioned arrows and real armor).

Not sure what group @Try mentioned (giving up on 25 pound bow as even hits against armor caused bruising), but there is a 20+ year history of combat using bows up to 30 pounds in the SCA (California region). The cushioning on the arrow has gotten better, but the arrow itself has gotten heavier and hence slower (due to the padding, anti-bounce back devices, etc.). Hitting a part of the body without armor or padding can result in a bruise, but these types of shots are uncommon, and armor is quite effective against the normal shots.

The SCA has an organized shoot that includes speed rounds; lasting 30 or 60 seconds, aimed at targets 20/30/40 or ~120 yards (last is less common, hard to find a range that allows this). The best archers I have seen can fire 10-12 arrows in 30 seconds hitting a 5″ group at 40 yards. Leaving aside the guys who have Olympic potential, even a moderate somebody (like myself when practicing once a month or so) can fire 6-8 arrows in 30 seconds hitting a “7 group at 20 yards. By contrast the quickest crossbow man I saw fired 6 shots in 30 seconds, and 4 per 30 seconds is more typical; albeit usually with better accuracy. But in closing consider that this is only for the lightest and quickest of crossbows (70 pound draw with a stirrup). Actual war crossbows designed to penetrate armor will get 1 or at most 2 shots off per minute.

There are no good measures of how long it takes become a sufficiently good archer, but Hardy compares modern experience for Olympic archery training which takes about a year (incidentally, archery has one of the shortest training period to become an expert of all Olympic sports). And that does not include the strength regimen of being able to draw a bow of more than 125-175 pounds (the calculated strength of the bows recovered from the Mary Rose, a Tudor era ship). By contrast a crossbow men could be trained in about 6 weeks (and musketeers took 5 weeks, or was it the other way around). In any case, the true astounding factors of the British longbow (paraphrase from Hardy) is both that the elites trusted, and that their society was productive enough, to produce a relatively large amount of men healthy and strong enough to become proficient with these specialized weapons.

Simon January 26, 2016 7:31 AM

I don’t know of any archer on a good Olympic team that had shot for less than 6 years. The top end of the sport is dominated by these who have practised since childhood.

Just about all long bow / warbow opinions formed before the raising of the Mary Rose were wrong, including all those armour piercing tests that shot mild steel tipped arrows from 40lb bows.

I expect most healthy 20-30 year old males could train hard enough over a year or two to pull a modern 50lb longbow for a 150 round shoot, spread over a day. A 175lb warbow would be a different beast altogether; my 40 year old shoulders would never do it – the training required would do too much damage.

Chris January 26, 2016 2:32 PM

Institutionally Constrained Technology Adoption has been seen in the battlefield historically. Bio/gas weapons are forbidden by institutional law in the form of international treaties, as is the use of nuclear weapon powered space ships. I believe the current cryptowars could be described in this fashion as well, since it is an attempt to restrict technological adoption by law.

supersaurus January 26, 2016 7:08 PM

@Gerard van Vooren

“…metric system…”

3000 ft-lb / 142 ft-lb -> dimensionless quantity. the measurement system doesn’t change that, nor does the century.

supersaurus January 26, 2016 7:34 PM

“…Most people prefer a bigger gun for elephant…”

most people also prefer not to use any size bow for an elephant.

in re energy delivered to the target arrows and bullets have the same problem: if they pass through the target some of the energy is not delivered. I don’t know about arrows, but bullets can be chosen to suit the target, i.e. armor piercing rounds are not the same as you might use to kill big game. have a look at ballistic gel results if our host doesn’t strip the link, or google for “ballistics gel test” and select “images”.

Gerard van Vooren January 27, 2016 12:32 AM

@ Supersaurus,

The problem with the non-metric system is that, like Roman numerals, it doesn’t scale well. That is the only thing I have to say about it.

Wayne P February 15, 2016 5:50 PM

I ran across what appeared to be a thoughtful treatise on the bow written by some English noble person back in 1895, or so. Seems he set out to test some of the theories that folks had regarding longbows – range, penetration, and so on.

As folks have already stated, a lot depends on the weight of the shaft, draw of the bow, and so on. Bottom line is that . . . It all depends.

For those of you who want a quick read (one chapter, at least), check out:

Douglas Emerson January 8, 2017 2:37 AM

Sorry to be to this discussion! Excellent points here around why armies would adopt longbows vs crossbows (or vice versa).

One differentiating factor I did not see raised here is that a longbowman could readily unstrung and restring their bow (even with its prodigious draw weight – trained longbowmen were made physically remarkable through their training.). This allowed them to keep their bowstrings dry in inclement weather – which became a factor at Crécy. When a rainstorm broke over the battlefield, the Genoese and French crossbowmen were not able to unstrung their crossbows, and the wet strings stretched referring them largely useless for the battle, while the English bows were unaffected.

James June 24, 2019 1:54 PM

I respect everybody’s opinion. All have his own opinion. My opinion is that crossbows are more fun and more efficient. Thank you.

Luis August 13, 2020 2:20 AM

This type of posts remind me of the old tale that a frog dropped into hot water will jump out, but a frog dropped into cold water that’s slowly brought to a boil will stay and die. (It was brought up in the movie Dante’s Peak, which is about people who lived at the base of a volcano and didn’t want to leave when it started acting up.)

I hope there weren’t too many frogs boiled alive to test that, because it doesn’t really matter what a frog would do, nor does the history of archery really matter here. I think the subject raised here was “Institutionally Constrained Technology Adoption”, and I’m wondering if it’s a real phenomenon. If so, how much of a problem is it?

Leave a comment


Allowed HTML <a href="URL"> • <em> <cite> <i> • <strong> <b> • <sub> <sup> • <ul> <ol> <li> • <blockquote> <pre> Markdown Extra syntax via

Sidebar photo of Bruce Schneier by Joe MacInnis.