Among the forgotten Bronze Age siege weapons is an entire system of tools that cracked city walls for two thousand years before torsion artillery existed. The Old Hittite narrative known as the Siege of Ursu records a king’s fury when defenders destroy the battering ram during an assault, a throwaway complaint that assumes the reader already knows what a ram is, how a crew operates it, and why losing it to fire is catastrophic. That casual assumption in a text from roughly the eighteenth century BCE tells us that these were not primitive improvisations. They were established tools operated by trained crews within a coherent tactical system, and the evidence for all of them survives in tablets, reliefs, and cylinder seals waiting to be read carefully.
What counts as a forgotten Bronze Age siege weapon
The Bronze Age, roughly 3300 to 1200 BCE, predates torsion artillery entirely. No catapults. What attackers actually used was a toolkit constrained by two hard realities: equipment had to be built from local timber and leather in the field, and the fortifications it faced were specifically engineered to defeat simple approaches. Middle Bronze Age builders across Syria-Palestine laid stone-faced glacis slopes at 35 to 45 degrees to deflect ram strikes and prevent ladders from finding purchase. They created inset and offset wall faces giving defenders flanking fire. Gate passages bent sharply to break momentum and deny a direct line of attack. Against those obstacles, the answer was simple tools in coordinated sequence, each one creating conditions for the next.
Three types of evidence survive, and each carries its own bias. Royal reliefs, including the Ramesseum scenes of Ramesses II attacking Dapur in Syria around 1274 BCE, compress hours of brutal work into a single panel celebrating the preferred moments of victory. Cylinder seal impressions from Early Bronze Age Syria, particularly those from Tell Beydar analysed by Davide Nadali of the Sapienza University of Rome in his 2009 study published in Historiae, show wheeled frames and siege structures in schematic miniature. Narrative tablets like the Siege of Ursu preserve emotional content, a king’s anger at a broken component, without engineering detail. Reading all three together, cross-referencing what each takes for granted, is how the operational picture forms.
How the battering ram actually worked
The Old Hittite Siege of Ursu already assumes the ram as standard equipment, placing the technology in use by the early second millennium BCE at the latest. The device in its Bronze Age form was not the enormous armoured vehicle of later Assyrian palace reliefs. It was a heavy straight trunk, sometimes reinforced at the striking end with a hardwood or bronze cap, suspended from a rectangular wooden frame on ropes or rawhide straps so it could swing through a short arc. Two to three crew on each side hauled the ropes backward and released on a caller’s signal. One man kept the cadence. Buckets of water sat beside the frame to douse the roof covering when defenders dropped burning bundles.
The ram’s most productive targets were timber gates and wall joints where two building phases met or where a repair had used inferior bonding. Well-made mudbrick resisted a direct face strike for a long time. Gate fittings, however, loosened under repeated shock. The metal hinges and lock-bar sockets expanded under vibration and cracked the surrounding masonry before the wood itself splintered. A crew working in short, rhythmic strokes with a wet rawhide roof above them could work a gate for minutes without every member being struck. Terrain determined which version of the tool was used. Where the ground was clear, a wheeled frame could be positioned and repositioned. Where a steep glacis blocked approach, crews switched to a hand-worked beam under a simple lean-to, and the tool changed shape to fit the ground rather than to match a blueprint.
Defenders had standard countermeasures that attackers planned against before the ram reached the wall. Sacks of wool or folded rawhide hung on ropes from the parapet absorbed and spread the force of each strike. Wedge-braces driven behind gate planks added structural resistance. Burning bundles of pitch-soaked wood dropped onto the ram’s roof were the most serious threat, which is exactly why early sources mention wet hides and why later reliefs show buckets near engines. Defenders who could not destroy the machine through fire tried to hook the striking head with chains or iron loops and drag it out of alignment. Nadali’s 2009 analysis of Early Bronze Age Syrian seal impressions shows wheeled four-wheeled frames with projecting beams positioned beside towered walls, suggesting that at least in northern Syrian contexts, crews were already mounting rams on frames mobile enough to reposition during an assault.
Mobile shields and how they made the last fifty metres survivable
The fifty metres between the edge of arrow range and the wall face was the most dangerous ground in any ancient siege, and mobile shields were the technology that made it survivable. A mantlet in its simplest Bronze Age form was a large rectangular screen taller than a standing man, built from thick planks or layered wicker faced with rawhide. A wooden model from Asyut in Egypt, dated to the Twelfth Dynasty around 1981 to 1802 BCE, shows the type clearly: a broad rectangular shield built at exactly the scale needed for this role. The Dapur reliefs from the Ramesseum confirm the tactical use, showing archers shooting from behind large portable screens while assault teams approach the gate, the two activities presented as simultaneous and coordinated.
In practice, two or three men carried the screen by handles set into its rear face, one person knelt to brace the base against the ground, and an archer stood a half-step behind to fire past its edge. The shield did not eliminate the threat from above. It eliminated the threat from in front, which forced defenders on the parapet to expose themselves over the edge to shoot downward at an angle rather than firing straight across. That exposure cost time and made them targets in return. When the team reached the base of the wall, the mantlet tipped from vertical to angled, acting as a lean-to roof under which pick crews could work at the lowest courses without being hit by material dropped straight down.
Defenders adapted specifically to mantlets. They aimed for the ankle gap at the screen’s base, threw jars of burning resin so fire spread on contact rather than bouncing off, and dropped heavy stones to break the carrying frame rather than trying to penetrate the face. The answer was to build mantlets from thicker material and to angle the screen so dropped stones struck at a deflecting angle rather than straight down. The weight penalty that came with thicker material was accepted as preferable to the alternative of arriving at the wall with a broken screen and no cover for the pick crew behind it.
Covered wagons and proto-towers: what the evidence actually shows
Several Tell Beydar seal impressions, examined by Nadali in his study of Early Bronze Age glyptic art, show multi-storey wheeled frames with men in stacked compartments positioned beside what reads as a fortified structure. Another seal pairs a wheeled tower with a ram striking a gate simultaneously, which suggests that in at least some early third millennium Syrian contexts, the combination of an elevated shooting platform and a ground-level striking beam was already a tactical concept. Seal art is tiny, schematic, and produced under conventions that prioritise symbolic legibility over engineering accuracy, so strong claims about structural detail should be treated with caution. What the pattern of images across multiple pieces justifies is a functional reading: Early Bronze Age Syrian armies were already imagining, and very probably building, structures that combined elevated archer positions with ground-level striking capability.
In practical terms, such a vehicle was almost certainly far simpler than the enormous Assyrian siege towers of the ninth century BCE. A covered gallery on two or four wheels, low enough to avoid presenting a large target, could shelter a ram crew below and archers behind a frontal screen above. Terrain set hard limits. Ditches, berms, and the steep stone-faced glacis that characterised Middle Bronze Age Syrian fortification stopped wheeled vehicles entirely unless a causeway or ramp was prepared first. This is exactly why the ram-on-wheels and the hand-worked ram under a lean-to existed in the same army’s toolkit at the same time. Which one was used depended on what the ground at the gate face actually offered.
Built out of a love for history, kept free from distractions.
Spoken Past is an independent project shaped by curiosity, care, and long hours of research. Reader support helps keep it ad-free, sponsor-free, and open to everyone.
Sapping picks and how they brought walls down from below
Where a ram could not generate a useful angle of attack and a direct assault on a gate was too costly, sappers working at the base of the wall offered a slower but dependable alternative. The primary tool was a bronze pick of the type well-represented among Luristan metalwork of the late third and early second millennia BCE, a heavy curved implement with a pick face designed to drive into stone joints and mudbrick courses and lever them apart. Getting the crew to the base of the wall alive required mantlets moving ahead of them and archery suppression pinning defenders back from the parapet simultaneously. Once at the toe of the wall, sappers removed lower courses behind a temporary prop of stacked timbers or heavy baskets filled with rubble.
When the cavity beneath the wall was large enough to threaten the structural integrity of the courses above, the crew set fire to the timber props. The fire weakened the props, the weight of the wall above overcame their support, and the facing slumped into the cavity. Where the wall incorporated horizontal timber lacing beams, which many Bronze Age mudbrick walls did, the heat also damaged those internal timbers and accelerated the collapse. On a stone-faced glacis, sappers aimed for the joint between the stone facing and the mudbrick core behind it. Opening a gap in that joint caused the stone facing to lose its key and slump forward, leaving the softer core exposed to further attack by pick and fire.
Defenders countered by driving countermining tunnels toward the attacker’s shaft from inside the wall’s thickness, breaking into the sapping gallery and fighting underground in extremely confined conditions. They also flooded the undercut by pouring water into the wall base from inside the city, turning the spoil into paste thick enough to impede pick progress. Attackers answered by improving drainage behind their screens and by rotating fresh crews through the working space more frequently than defenders could sustain underground sorties. The back-and-forth below the wall was invisible to everyone except the men involved, and it decided outcomes just as often as the more visible contest on the parapet above.
Fire: the cheapest and most versatile weapon in the arsenal
Bitumen, pine resin, and plant oils were widely available across the ancient Near East, which meant incendiary weapons were never scarce and required no specialist manufacturing. Attackers used fire in three distinct tactical roles, each targeting a different vulnerability. Against timber gates, sustained fire weakened the bronze or copper strap hinges and lock-bar sockets set into surrounding masonry by expanding the metal and cracking the stone holding the fixings, rather than simply burning through wood, which took far longer. Against sapping cavities, fire applied to the timber props triggered collapse without requiring sappers to remain in the shaft during the final dangerous seconds. Against the parapet walk, clay jars of liquid pitch with burning textile wicks thrown in a high arc scattered burning fuel across sections that defenders then had to abandon or smother before resuming effective fire at the crews working below.
Sand, not water, was the correct defensive response to burning pitch. Water dispersed burning liquid rather than extinguishing it and could spread fire to adjacent areas faster than the original throw had reached. Defenders kept dry sand in pots along the wall walk beside teams of men stationed specifically to smother any incendiary that landed before it spread. Wet hides stretched over gate surfaces slowed ignition. Clay packed into joints around heating fittings insulated the surrounding masonry long enough for a relief force to push ladders away. The balance between these measures meant fire was rarely decisive on its own. Used at the right moment in the right sequence, it turned positions defenders were holding into positions they could not hold long enough for help to arrive.
How all the forgotten Bronze Age siege weapons worked together
The Dapur reliefs from the Ramesseum are the most detailed single visual source for how a Late Bronze Age assault combined its components, and they show what every account of the period implies: the weapons operated simultaneously, not sequentially. Archers suppressed the parapet, forcing defenders to crouch behind merlons rather than aim, while shield-bearers moved the assault team to the gate face. That forward movement created the conditions in which pick crews could reach the base of the wall and ram crews could operate the beam for more than a few seconds without every member being hit. The archery did not need to kill defenders to be effective. It needed only to make the parapet walk a place where standing upright became a fast way to die, which is a lower bar and achievable even against well-built fortifications.
Once the assault team reached the gate, the ram worked the timber while sappers attacked the adjacent wall section simultaneously. Running two efforts at the same time forced the garrison to split its response. Split defenders are slower defenders, and slower defenders cannot maintain effective missile fire, resupply material to the wall walk, and countermine below the surface at the same time. Fire was applied to the gate surface and to sapping cavities as each effort reached the stage where ignition would accelerate the breach rather than create a hazard to manage. When a section gave way, ladders went up on either side of the breach to stretch the garrison further while the main assault poured through the gap. The most revealing detail across all the sources is not any single device but the way the devices interlock: archery makes space for shields, shields make space for picks, picks make a place for fire, and fire opens a door for men.
Why this arsenal stayed forgotten
The main reason these weapons are poorly remembered is that Assyrian palace reliefs of the ninth and eighth centuries BCE are visually spectacular and survive in abundance. The enormous wheeled towers, sophisticated ram carriages, and complex combined-arms assaults shown on the walls at Nineveh trained modern scholars to think of those images as the origin point of siege warfare rather than its mature elaboration. When earlier evidence was examined, it looked modest by comparison and was sometimes dismissed as too sparse for detailed reconstruction. That judgement missed how much information the earlier sources actually contain once you read across formats instead of treating each one in isolation.
The other factor is material survival. Timber, rawhide, wicker, and rope decay completely in the archaeological record of the ancient Near East unless burned into charcoal or kept in exceptional dry conditions. The working components of Bronze Age siege weapons almost never survive to excavation. What survives is the bronze fitting, the pick head, the occasional clay firepot, and the wall that was attacked. Reconstructing the equipment that connected those fragments requires exactly the cross-reading between tablets, seals, and reliefs that the evidence demands, and that takes longer to do carefully than simply looking at a Nineveh palace relief and noting what is visible.
Sources and further reading: Davide Nadali (Sapienza University of Rome), Representations of Battering Rams and Siege Towers in Early Bronze Age Glyptic Art, Historiae 6 (2009): 39-52; Ancient Near Eastern Battering Rams: Questioning Their Penetrative Power and Target Location (Academia.edu, 2016); Aaron A. Burke, Walled Up to Heaven: The Evolution of Middle Bronze Age Fortification Strategies in the Levant (Eisenbrauns, 2008); Old Hittite Siege of Ursu tablet, translated in G. Beckman, Hittite Diplomatic Texts (Scholars Press, 1996); Ramesseum Dapur siege reliefs, Thebes, 13th century BCE; Robert B. Partridge, Fighting Pharaohs: Weapons and Warfare in Ancient Egypt (Peartree Publishing, 2002).
