The terms used to refer to the individual throwing/hurling machines throughout available primary sources are used indiscriminately. It is not certain whether that is due to a loss of translation when the technology was passed from civilization to civilization, the relative literacy rates of the period, carelessness on the part of designers, or simply unfamiliarity on the part of historical chroniclers. Those individuals who made an attempt to chronicle the history of these periods oftentimes used interchangeable names when referring to siege weaponry. For example, the Onager could also be referred to as a Petraria or sometimes Springald. However, what is known is that the ancient world had developed the necessary technological and mathematical expertise required in order to successfully construct, and accurately operate, these war machines. Over time the designations for individual machines has been somewhat codified resulting in the classifications seen below. Within each machine class there is the possibility of broad variations in design, however the process by which the siege machine delivers its payload is the common denominator (gravity, torsion, etc).
The Ballista was the earliest form of siege weapon seen on the battlefield. First fielded sometime in the 4th century BC, their construction and material lead us to believe that the weapon was designed for anti-personnel or light structure targets. It was made primarily of wood and sinew, metals being extremely rare/valuable restricted their application to very few, but very important, pieces. The general design looks much like a large bow, however the arms themselves are separate rather than being one piece. The individual arms were wrapped extremely tightly with sinew and attached to pivots. Then a cord was strung between the two and cranked back in order to load the mechanism. The sinew based power generation of the weapon is known as torsion spring bundle technology and its discovery allowed for the development of the machine. The ammunition was typically a steel tipped bolt that looked much like a short, thick arrow, or rounded stones; either way they were targeting either personnel or wooden structures. The first mention of a ballista in literature was by Dionysus of Syracuse in 400 BC, and has proven to be a scarce that early in history. However, the conquests of Philip II of Macedon and his son Alexander (The Great) spread the technology and perpetuated its application on the battlefield. With the consumption of the Greek States by the Roman Empire in 146 BC, they became a tried and true component of legions for the majority of the period. Smaller versions than the siege class, known as scorpions, were developed and deployed throughout the Second Punic War while others were mounted on chariots and dubbed Carroballista. Generally they were between 9 and 15 feet long with a maximum range of 500 meters but effective range extending to just over half that distance. They were transported from battle to battle disassembled and were reconstructed at the site due to the immense strain they were under while in combat configuration. With its earliest sightings cemented firmly in the Hellenistic and Roman Periods, it continued to serve on the battlefield through the Middle Ages and is reported to have been used in the First Crusade at the Battle of Nicaea in 1097. The Islamic defenders placed a number of them on the city rooftops allowing them a clear field of fire at the besieging Christian forces. 
The Onager is the first machine that appears like the traditional notion of a catapult. However, its means of generating force were actually the same as the Ballista, just applied in a slightly different manner. Its construction was almost entirely of wood with very few metal parts, but unlike the ballista was mounted with wheels in order to make it mobile. Just like the torsion spring bundle energy generation on the ballista, the Onager’s system was the same except rotated. The sinew was wrapped about an axle set in the base frame, and much like a T-bar, the firing arm extended from the center of that axle. The first design can be attributed to the Roman Empire around the 1st century. It was applied across a broad range of theaters as a relatively common siege weapon with the legions. Where the ballista was intended for personnel, the Onager was its balance, used to destroy buildings or fortifications. Its payload delivery was a system that used the power generated by the torsion spring bundle and the pendulemic effect of a sling to launch its projectile. Using a sling as the payload delivery system was singularly important because it allowed for variation in the vertical nature of the projectile’s trajectory as well as an adjustable measure for distance and targeting. The maximum range of the weapon was fairly similar to the ballista, yet its effective range extended to that limit based on the vertical nature of the projectile. Being a bow based weapon, the Ballista and its variants fired horizontally while the Onager fired vertically on account of the sling launch. The design was propagated throughout the major powers of the Roman Period and became the basis for variation and the development of the Mangonel. Both the Onager and Mangonel were applied throughout the period; however, geographic location seems to have played a part in preference. By the time the 12th century came about, siege machines were being organized into artillery batteries. The Siege of Rouen in 1174 showcased siege weapons organized in batteries operating day and night for the duration of the siege by arranging the soldiers in rotating 8 hour shifts.
The Mangonel is the second torsion spring bundle catapult used, and its development mirrored its very close cousin, the Onager. It was first written about in the 12th century by an Arabian scholar named Mardi bin Ali al-Tarsusi who penned a number of military treatises for Saladin throughout the early crusade era. He reportedly described the Mangonel, its construction, possible improvements, and further classified the individual siege weapons. The weapon system itself had been in existence for a much longer period of time; however al-Tarsusi’s writings are the earliest form of descriptions and chronicle pertaining to the machine. The Mangonel used a torsion system exactly the same as the Onager (sinew about a T-bar) but the difference is in their payload delivery systems. The spoon adaptation allowed for varied applications across the battlefield due to its ability to control a number of different ammunition types. The spoon made it either anti-personnel or anti-structure because it could fire a mass of smaller stones or single enormous boulders. There are reports that the length of the firing arm could be adjusted as well thus significantly altering the trajectory and targeting of the war machine. The peak use of this weapon system occurred throughout the crusade and medieval periods from about 1076 through the mid to late 1400s. It seems that throughout al-Tarsusi’s area of expertise, Arabian conflicts post 1200; the spoon delivery was the most popular and trusted. Interestingly, the decision on which system to use seems to have had a geographical stimulus; the Arabian civilizations (Turk, Saracen, etc) relied heavily upon spoons while the Western powers (England, France, etc) had a particular proclivity towards slings. A rather prominent display of the Mangonel’s power occurred during the Siege of Malaga in 1487. Ferdinand and Isabella of Spain used the machine to hurl the remains of an assassin into the city during the Reconquista against the Emirate of Granada.
The Trebuchet is the stereotypical version of a catapult envisioned by most of the public today, it is also one of the most powerful weapons in pre-firearm military history. Throughout the Medieval Age the largest siege operations usually were accompanied by the construction of trebuchets. There is evidence to suggest they first came to be around the 12th century; however their existence cannot be verifiably proven until the following century. Due to their enormous size and the corresponding forces exerted, Trebuchet’s were actually built on-site according to the needs of the bombardment. Instead of torsion spring bundles, the trebuchet uses the force of gravity coupled with centripetal motion to generate the power on the payload. Constructed of huge timbers, corded sinew, metal banding, and enormous counterweight, the typical machine could stand as high as 30 feet loaded and extend to 45+ feet while firing. A firing arm is attached to a pivot point between two A-frames and locked into place. A counterweight weighing as much as 300 times the weight of the payload is attached to the short end of the firing arm, while the long end includes a sling for the payload itself. Simply releasing the firing arm swings it about the pivot and sends the payload soaring towards the target. With a maximum range depending on size and counterweight, the largest could reach out to targets as far as 800 meters away firing stones weighing nearly 300 pounds. Ammunition was not restricted to boulders, various types included smaller stones, fire pots, bodies, body parts, etc. Being the largest siege weapon available, their typical battlefield role revolved about the demolition of opposing fortifications and structures, however prisoners of war and bodies were fired to spread terror and disease. The Trebuchet is the only siege weapon described here with its invention remaining solely in the Middle Ages. Every other siege weapon mentioned has some aspect of its creation grounded in the Hellenistic or Roman periods. The most famous application of the Trebuchet is their use throughout Richard the Lionheart’s Third Crusade. During the Siege of Acre from 1189 to 1191, the machines were used extensively in a series of sieges and counter-sieges. Richard relied on them so heavily that he even had a contingent shipped from England to the Holy Land for the duration of his campaigns there.
Bradbury, Jim. (1992) The Medieval Siege. Woodbridge: Boydell Press.
Brundage, James. (1962) The Crusades: A Documentary History, Milwaukee: Marquette U Press.
DeVries, Kelly. (1992), Military Medieval Technology, Peterborough: Broadview Press.
Duncan Campbell and Richard Hook (2003), Greek and Roman Artillery 399BC - AD363, Oxford: Osprey Publishing.
Gravett, Christopher .(1990) Medieval Siege Warfare. Oxford: Osprey Publishing.
Nicholson, Helen. (2004), Medieval Warfare: Theory and Practice of War in Europe, 300–1500, Basingstoke: Palgrave Macmillan.
Reston, James Jr. (2001), Warriors of God: Richard the Lionheart and Saladin in The Third Crusade, New York: Random House.
Tyerman, Christopher. (2008), God’s War: A New History of the Crusades, Cambridge: Belknap Press.
Warry, John. (1995), Warfare in the Classical World, London: Salamander Books.
 Duncan Campbell and Richard Hook. (2003), Greek and Roman Artillery 399BC – AD36., Oxford: Osprey Publishing, pp14.
 John Warry. (1995), Warfare in the Classical Wordl, London: Salamander Books, pp178.
 Jim Bradbury. (1992), The Medieval Siege. Woodbridge: Boydell Press, pp250.
 Bradbury, pp251.
 Campbell, pp17.
 Bradbury, pp257.
 Bradbury, pp255.
 Bradbury, pp253.
 Christopher Gravett. (1991), Medieval Siege Warfare. Oxford: Osprey Publishing, pp53.
 Helen Nicholson. (2004), Medieval Warfare: Theory and Practice of War in Europe, 300 – 1500. Basingstoke: Palgrave Macmillan, pp168.
 Bradbury, pp260.