In ancient times, cities often had strong walls around them, and warfare against these cities had always involved the basic tasks of breaking the walls, going over or under the walls, or starving the defenders into surrender. In the Middle Ages, Europe’s decentralized political structure put a new twist on the siege by planting heavily fortified castles all over the landscape. Constantinople’s thick city walls were similar to the fortresses of Roman, Greek, and more ancient times. Northern Europe, on the other hand, had several hundred small fortresses that were designed to hold off disproportionately larger attackers. In order to capture a region, an invader would need to besiege more than one fortress.
After the period of the First Crusade, knights returned with much grander ideas of defensive fortification. They had seen Byzantine fortress designs and had participated in attacks on Antioch, Acre, Jerusalem, and Tyre. Crusaders had built their own fortresses to hold the new territory, and they had used local engineering and labor to build much larger stone fortresses than Europe had at the time. When they came home, many rebuilt their family castles to incorporate the new defensive features. Castles became harder to capture by direct assault.
Sieges, attacks that stretched out over a long period of time, were the only way of capturing a castle unless it was taken by surprise. Sieges were expensive for both sides. The attackers had to sustain an army in hostile territory for a number of months, while the defenders had to make their food and water last. Both sides worked hard to attack or defend the walls. Walls could be broken down or surmounted by going over or under the walls. Siege machinery falls into three basic types. Catapults threw projectiles over the castle walls, either into the castle or from the castle toward the attackers. Rams battered the walls to make them fall down. Siege towers lifted attackers to the top of the wall so that they could enter.
Because of the high stakes and expense, sieges were not governed by the polite rules of chivalry. No trick was too dirty, gross, or savage. Treachery was one of the best ways of breaking a siege, if an insider could be bribed to open the gates or tell of a secret weak point. Poison or bacterial contamination of food or water was a popular way to break a siege.
Climbing, Ramming, and Digging
The simplest siege weapon was the ladder. The attackers wanted to get into the fortress, and one way was to go over the walls. Siege ladders had been used against city and fortress walls since ancient times. Basic facts that governed the construction of siege ladders began with length: if a ladder was too short, it would not allow the attacker to go over the top, but if it was too long, its top would stick up where defenders could shove it away. The ladder had to lean enough to be stable, but it had to be vertical enough to be strong. The ideal siege ladder came to just below the top of the wall, and its foot was placed at a distance from the wall equal to about half its length. Since the walls of a town or castle were of varying heights, and were surrounded by varying terrain, the attackers had to build custom siege ladders for each position.
A refinement on the simple ladder was a ladder with a bridge. The bridge was a sturdy plank hinged at the top of the ladder, raised by ropes. The ladder had to be somewhat freestanding, like a platform, since it could not lean against the wall. Some engineers designed folding ladders that could be made in advance and carried with the army or ladders that could be assembled from short sections. Some sieges also used ladders made of rope or leather, with hooks at the top. These ladders were for quiet night attacks, when the ladders could suddenly appear hooked on top of the walls by long poles without the defenders having seen any ladders.
Defenders tried to repel attackers on ladders by using the force of gravity. Standing at a higher level, they could drop harmful substances on the climbers. Most often, they threw large rocks to knock the attackers off the ladders or force them to cover their heads. Sometimes they threw or poured boiling water, oil, or any other hot substances they had on hand, such as tar. They could also throw quicklime, a highly caustic, alkaline material that burned on contact. In sandy places, they could heat sand to red-hot and fling it down. In some cases, they could fling nets onto the attackers when they reached the top and trap them.
To protect against all these defenses, attackers used heavy shields. Since classical times, there had been siege shields made tall, curved back, or with a small roof, and at times on wheels. Many shields were large enough for more than one man. Medieval sieges used all forms of wooden shields, covered with leather. In the 15th century, the tall siege shield was called a pavis. It often had a spike to drive into the ground and a pole to hold it up.
Of course, the first defense against siege ladders had been put in place before the siege began, when the fortress was designed. Most fortresses used a ditch or moat that came as close as possible to the outer walls. Attackers had to fill in the ditch with sacks or barrels of rocks and earth. In some cases, they resorted to using catapults to land rocks and dirt in the moat. Unless the ground was reasonably level approaching the wall, their use of siege machines would be limited.
If the attackers continued to try to go over the walls, but needed more than ladders, the next logical step was to make portable sheds. Sheds could be made fi re resistant with water and fresh skins. Sheds could also disguise or protect structural attacks, such as digging or battering rams.
The purpose of a ram is simple. It is a strong tree trunk that hits a wall, gate, or door repeatedly until the object is smashed. The design of a battering ram had three aims: to strengthen the ram itself, to increase its force, and to protect its operators from counterattack.
The end of the battering ram was strengthened by a metal tip. Sometimes this was actually in the shape of a ram’s head, invoking the ram’s butting strength and using the ram’s extended snout as the focal point of the battering force. More often, it was a blacksmith’s iron binding so that the wood did not shatter as easily from the force put on it. The ram’s bulk was suspended by ropes that could swing it, so the operators of a battering ram did not need much human power to strike with it. Longer ropes, of course, gave it more swinging power. The frame that suspended the ram was usually roofed so that its operators were protected from arrows or stones. Finally, the roof was often covered with damp animal skins as fire prevention.
Defenders dropped projectiles and hot liquids on the operators of battering rams. They could also try to disrupt the action of the ram, if the ram’s housing was too well defended to be vulnerable to rocks or fire. When the ram struck the wall, they could try to hook it and pull it up, either deflecting its blow or flipping its shed over.
Battering ram technology had been well explored during classical times, and, although rams were still used, castle designers built walls to withstand them. The thickest parts of the walls were at battering level, and gates, the main target of rams, were protected by gatehouses and moats. Attackers had to find new ways to use rams during the Middle Ages. Small rams could be mounted on ladders and lifted up to smash parapets. Attackers could build an earth ramp to a higher point in a wall, where it was likely to be thinner.
Attackers could also try to drill holes in the walls. Borers, too, had to work in the shelter of sheds and shields. It was not easy to drill holes in stone walls, so borers were more commonly used against brick. They were not a large feature of Northern European siege warfare, since most French and English castles were made of limestone and granite. A strong brick wall could be sufficiently weakened by holes that a ram could bring it down. Holes could have wood pushed in and set on fire, and the heat further weakened the walls.
By the 14th century, castle walls were built to be too high and thick for ladders and rams to be effective. If ladders and rams could not boost attackers over or batter walls down, a more elaborate machine could be built. A siege tower was a heavy, cumbersome machine, not designed for a lightning attack or for secrecy. It was part of an all-out assault on a weakened castle. The tower was a tall wooden structure on wheels; it was sometimes called a castle or a cat. It had protective walls and a roof and was fi reproofed if possible. Inside, it had wooden floors as stories where attackers could stand. A ladder led from bottom to top, so each layer of attackers could climb the ladder in turn. A top floor allowed archers to give further defensive cover to the attackers. The siege tower also had a bridge to cross to the top of the wall. This bridge could be a drawbridge, operated by a windlass in the bottom story.
Certain engineering issues governed the construction of siege towers. They had to be tall enough to reach the walls and stable enough not to tip when loaded with climbing soldiers. They also had to be portable, usually on wheels. Medieval siege towers were as tall as 75 feet high, but they were often shorter. Designs in medieval illustrations appear to favor a small fortress on a rolling platform, reached by one or more ladders. The attackers expected a tough fight before they could cross a bridge, and they designed it to have walls or even a roof. Other siege towers were more like rolling platform ladders with bridges. The builders had to think about fire, since the most common way to defend against a siege tower was to set it on fire. In the Byzantine region, siege towers became obsolete when it became clear the defenders would hurl Greek fi re at them. Northern Europe was able to use siege tower tactics longer, since it was easier to defend wood against ordinary fire. The tower could be roofed with fresh turf or newly skinned wet hides.
Siege towers were heavy and could easily tip over. It was difficult to move them into position from the safe distance where they had been built. The ground had to be level, and many teams of oxen were needed to move them. They also needed to be moved close to the walls, which normally meant that pushing, not pulling, force was required. One way to move a very heavy siege platform was to sink one or more posts into the ground by the castle walls and loop heavy pulleys and ropes around them. The platform was then attached to the ropes, and it could be moved forward by oxen walking away from the battle. The siege tower inched closer to the fortress walls, but the muscle power moving it only moved farther out of range. The tower could come right up to the pulleys, if the defenders had not disrupted them. Towers could also be moved with levers, but, in any case, they moved very slowly because of their great weight.
Undermining a wall could be the most successful attack, and there were fewer ways for the defenders to work against it. Ideally, the defenders would not know that sappers were digging a tunnel under the walls. The first somewhat successful underground attack in medieval times was carried out by the Vikings when they besieged Paris in 885. After the Norman conquest of England, mining was part of many sieges. The siege of Rochester Castle in 1215, when King John of England was putting down a rebellion, was one of the few times when mining was a key factor in the castle’s surrender. Miners dug under two outer walls so that the defenders were trapped in the keep. Chateau-Gaillard, built by King Richard I of England, was designed to be impregnable, but miners collapsed its walls twice. Mining was a large part of Crusader warfare, on both sides.
The best location to begin a sapping operation was in a place where the defenders could not observe what was going on without leaving the fortress. Sappers sometimes needed to start some distance away, on the other side of a hill. The attackers could put up a wooden palisade so that the defenders could not see what they were doing on the other side. If the diggers had to start in a place where the defenders could observe them, they needed a strong shed to protect them. The shed was sometimes nicknamed a “tortoise” or a “sow.”
An attacking army drafted industrial miners to dig their siege tunnels. Since stone was tunneled from deep underground, even from under the city of Paris, miners knew how to dig any length of tunnel required, through any materials. Beginning in a safe place, they dug underground and moved in a carefully planned direction toward the walls. Sometimes, two tunnels were dug as parallel galleries. As the miners tunneled, they shored up the walls of the mine with strong timbers. Mining was an operation that required a large number of laborers, which made it difficult to carry out deep in hostile territory.
When a tunnel successfully reached a point under the defensive wall, the miners nearly always started a fire. The intense heat caused the ground to expand, which cracked the walls and collapsed the tunnel. Added to the wood carried into the tunnel, oil and fat made the fire burn hotter; one mine fi re, in the siege of Rochester, used 40 pigs as sources of fat. The wood props securing the tunnels also burned, allowing the tunnels to collapse faster.
Once gunpowder was in use, it was even easier to produce a hot blast. It was harder to get away safely, since the combustion happened so quickly and the blast collapsed the tunnel. The best way was to approach the wall with snake-like curves and then use the curved passages to set a long fuse, out of sight and reach of the blast. As the fi re crept along the fuse, the miners could escape out the end of the tunnel. Since gunpowder came into use at the end of the Middle Ages, it did not become a major force in siege mining until the Renaissance period.
Most walls collapsed when the ground supporting them caved in. There were few ways to build walls that were not vulnerable to sapping. One way was to reinforce the walls with stone columns laid like pegs through holes in the building stones. Places with ruined Roman or Greek columns could use them this way, but most places did not have ruined columns. The fortress design could also use very deep digging to place a moat or a wall in vulnerable places.
Defenders tried to detect tunnel digging when they could not see it. A bowl of water, set over an area being mined, quivered with the vibrations of the tools. If they could tell where the miners were approaching the wall, the defenders could dig down to meet and surprise them with combat. They could sink a hole nearby and try to set the attacking tunnel on fire, or they could flood it if they had a moat or river inside the walls. The attackers tried to make their tunneling less predictable by making decoy tunnels or by making the tunnels take unexpected paths. Tunnels could branch out, or they could zigzag or curve.
Working trebuchet at Château des Baux
Machines that threw projectiles were known by many names in their time, although today we refer to them all as catapults. There are a few simple forces that can provide ballistic power without explosives or motors. Levers and gravity can be harnessed to provide flinging power. The power of both tension and torsion derive from a material being bent so that it will spring or unwind back to its original state.
Tension engines worked by bending wood; it would spring back to shape when the tension was released, thus flinging a projectile with the force of its movement. Crossbows and longbows work on this principle, and some larger forms of crossbows could act as siege weapons, throwing larger projectiles. These great crossbows were built on a frame and used a windlass at the back of the frame to wind the bolt on its string far back. When the windlass was released, the wooden bow’s tension thrust its heavy bolt forward with speed and great force. But wood’s ability to bend and snap back is limited by its tendency to crack. Wooden bows could not throw anything larger than a bolt and could not take aim at walls, but only at people.
Torsion is the force exerted by a rope that has been twisted tightly and tries to untwist. It is the principle of a child’s toy boat or airplane that uses a rubber band wound up tight to drive paddles or propellers as it unwinds. Torsion had been used to drive throwing machines since ancient times. The Romans had a throwing machine called an “onager,” a wild donkey. It used a very thick band of rope, highly resistant to being twisted, as the torsion spring. When a lever was inserted into the torsion spring and cranked back so that the rope was forced to twist, on release it sprang into the air. The lever had a sling on the end with a heavy stone. As it sprang into the air, it struck a bar that stopped its movement, and the stone flew out of the sling. The onager’s simple torsion spring provided great velocity and force.
Medieval uses of the torsion spring are not as clear. There is evidence that torsion machines of this kind were known in the time of Charlemagne. Artists’ illustrations show a machine similar to the Roman onager, but instead of a sling at the end of the lever, there is a spoon-shaped cup for the rock to be placed in. It was probably called a mangonel. Turkish medieval sources picture a device similar to the Roman one, called a manjaniq, used by Muslim armies.
In the 14th century, there were large crossbows that did not use bent wood, but rather had two separate arms with torsion springs. Different types were known variously as ballistae and espringals (and in other languages, springarda or springolf). They were more often used by a fortress’s defenders, since they shot bolts at individuals, rather than rocks at walls. The espringal was built into a wooden frame, mounted on a tower. On each side, the frame had a torsion spring made of very thick horsehair rope that was resistant to twisting. Levers inserted into each spring were pulled back by ropes attached to the firing mechanism. The espringal’s firing system was like a crossbow, with a long groove for a bolt. The operator cranked the bolt back, pulling on the levers and the torsion springs. Released, the torsion springs untwisted and the levers shot the bolt forward, through the groove and out toward the target. The bolts were long and heavy. They could be expected to pierce wooden shields, steel armor, and sometimes more than one body.
The third type of throwing machine used levers and gravity. Since ancient times, people had known that if a lever is put over a fulcrum, like a seesaw, and the lengths are not equal, it takes a much heavier weight on the short end to balance a lighter weight on the long end. If the short end is suddenly weighted, the long end will fly into the air very fast. Unlike tension and torsion, which depend on the strength of bent wood or twisted rope, lever-based machines can throw very heavy objects with relative ease. As long as the lever’s arm and the stand with the fulcrum hinge are strong enough, there is no load limit.
The perrier used only the lever to fling large stones. The perrier depended on a sudden downward pull by men or horses. Its frame lifted the lever’s short arm above the men’s heads, with a rope dangling down, and the long end rested on the ground with a sling. They could load a heavy rock into the sling. When the payload was in place, men with ropes pulled the short end down, as hard as they could, and the long arm with its rope swung upward suddenly, flinging the projectile into the air. In order to achieve significant force, the pull had to be both sudden and hard. Many ropes attached to a bar allowed many men or horses to pull. Sudden pull could be achieved by having the throwing arm restrained by a latch as the men began pulling, so the latch could suddenly be released. The perrier may have been in use by the 11th century.
The trebuchet used a lever with a very heavy counterweight on its short end. The long arm, with a sling on the end, was winched to the ground, forcing the boxy counterweight to lift into the air. Men loaded a large stone into the sling as the long end was held down firmly. When the long arm was released, the counterweight fell to the ground, suddenly lifting the long throwing arm and releasing its sling-propelled payload into the air. Because the machine’s power depended on gravity to pull the counterweight down, not on men or horses to tug it hard, the trebuchet was the strongest of the throwing machines.
Trebuchets could be built larger and stronger to throw ever-larger payloads. Instead of a windlass, the winching could be accomplished by one or two wheels, the way the tallest cranes raised loads. Several men stood inside the wheel and walked on its steps, using their weight and a pulley system to magnify the force. The counterweight, perhaps by now a large wooden bucket filled with many large stones, slowly lifted into the air. The throwing arm was lashed down, the men got out of the wheels, and the counterweight could be released.
Rocks are the best-known catapult payload, and they were the most commonly used. A machine could deliver a series of rocks to the same spot on a wall if the rocks were the same weight and the machine had not been moved. This pounded the wall over and over, increasingly weakening it. Iron shot was even better than stone shot, but it was more expensive.
As a siege went on, trebuchets were loaded with new payloads that were intended to frighten or harm the people inside. The trebuchet was now aimed to fling over the wall, not at it. Most often, armies threw dead animals or even dead human body parts. Severed heads were a common payload. Corpses spread disease, a deadlier attack than any rock. An assault with corpses was also a psychological terror weapon, especially if the heads or other body parts belonged to the targets. Trebuchets could also fling manure.
A shrapnel effect came from “beehives”-clay pots packed with rocks. They burst open on contact, and the rocks flew into the town to smash windows and injure people. Armies also threw incendiary mixes, such as hot tar and quicklime. Incendiary mixes were often called naphtha; there are a few existing recipes. Quicklime was the key ingredient, because water causes combustion on contact. Other ingredients were flammable substances: pine pitch, tar, oil, animal fat, and dung.
Greek fire was the most famous incendiary compound of the time. The name “Greek fire” caught on because it was invented in Constantinople, after they had lost territory to the invading Muslims. Byzantine soldiers used catapults to fling pots of Greek fire at besieging Muslim armies. It caught fire on contact, and even water did not put it out. They could pump it at attacking ships and burn up whole fleets; in this way, they saved Constantinople in the seventh century when other cities were conquered by the invading Arabs. The secret composition of Greek fire was carefully guarded for a long time, but eventually first Muslims and then Christian Europeans learned how to make it. It became a component of trebuchet attacks during sieges. However, there is no surviving account of what was in Greek fire. Many scholars speculate that it must have contained quicklime or saltpeter, and others believe it had to use petroleum as a main ingredient. The use of petroleum in some form seems very likely, since Greek fire was described as a liquid that burned even on top of water.
After the introduction of gunpowder, cannons became the main siege breaking weapon. The largest cannons, called bombards, required large trains of horses and oxen to move their parts and heaps of stone shot. They had to be moved off their wagons with large cranes, and they were fired either from heavy wooden frames or from trenches dug into the ground. The idea was not to fire the stones or iron balls into the fortress, but to fire them straight at the defensive walls. A bombard that was placed lower to the ground could aim right at the ground level. It was most effective when it was close to the wall, its operators defended with wooden walls.
Further Reading Bennett, Matthew. Fighting Techniques of the Medieval World: Equipment, Combat Skills, and Tactics. New York: Thomas Dunne Books, 2005. Carey, Brian Todd. Warfare in the Medieval World. Barnesly, UK: Sword and Pen, 2006. Donnelly, Mark P., and Daniel Diehl. Siege: Castles at War. Dallas: Taylor Publishing, 1998 Keen, Maurice. Medieval Warfare: A History. Oxford: Oxford University Press, 1999. Nossov, Konstantin. Ancient and Medieval Siege Weapons. Guilford, CT: Lyons Press, 2005. Partington, J. R. A History of Greek Fire and Gunpowder. Baltimore: Johns Hopkins University Press, 1999. Payne-Gallwey, Ralph. The Book of the Crossbow: With an Additional Section on Catapults and Other Siege Engines. Mineola, NY: Dover Publications, 2009. Rihll, Tracey. The Catapult: A History. Yardley, PA: Westholme Publishing, 2010. Wiggins, Kenneth. Siege Mines and Underground Warfare. Princes Risborough, UK: Shire Publications, 2003.