No CrossRef data available.
Article contents
On Artillery Towers and Catapult Sizes1
Published online by Cambridge University Press: 27 September 2013
Abstract
Some recent statements in the scholarly literature exaggerate the length of ancient catapults, so a critique is offered, and an explanation for the source of the error is suggested. A brief account is given of how to calculate the dimensions of the ‘classic’ types of ancient catapult according to the formulae for their construction in Philon and Vitruvius. The implications for fitting artillery to fortifications are highlighted. It is emphasised that the type of device for which first generation artillery towers were designed were tension-powered catapults, not torsion-powered models. They were bow catapults (tension), not spring catapults (torsion), and therefore, since the formulae apply only to spring catapults, none of the ancient formulae apply to the machines for which first generation artillery towers were built.
- Type
- Articles
- Information
- Copyright
- Copyright © The Council, British School at Athens 2006
References
2 e.g. by himself in ‘Towards a typology of Greek artillery towers: the first and second generations (c. 375–275 BC)’ in van de Maele, S. and Fossey, J. M. (eds), Fortificationes Antiquae (Amsterdam, 1992), 147–169Google Scholar, and Winter, F. E. ‘The use of artillery in fourth-century and Hellenistic towers’, Echos du monde classique/Classical Views 16 (1997): 247–92Google Scholar.
3 The principal ancient sources for the classifications are Heron Ktesibios'Belopoiika 74.5–75.9 and 104.4–106.5 (Wescher), Philon, Belopoiika 51.8–55.11 (Thévenot), and Vitruvius, De architectura 10. 10. 1–10. 11. 7. English translations are available in Marsden 1971.
4 The formulae and key dimensions are given in the same sections of the treatises cited in n. 3.
5 The reference to ‘p. 34 and fig. 34’ in Marsden 1969 is wrong; it should be ‘p. 34 and fig. 18 on p. 36’. Marsden does not present this figure as a formula.
6 It is not relevant here that stone-throwers could be adapted also to shoot sharps.
7 A half-talent = 3,000 drachmai. Applying the formula for palintones (1.1 ∛x) and rounding figures to the nearest quarter for simplicity of presentation (but not rounding them in the calculations), the cube root of 3,000 is c.14½, multiplied by 1.1 = 15¾, which is therefore the recommended width in dactyls of the spring diameter D for a half-talent palintone. This is about 300 mm in modern measures, taking 1 dactyl as 19.3 mm with Marsden. Multiplying by 19 to find the length of the stock gives c.5,810 mm or 19 feet in modern measures (so Marsden 1969: 145). Let us now consider a larger machine to throw stones twice as heavy as those above. One talent = 6000 drachmai; the cube root of 6,000 = 18¼; multiplied by 1.1 = 20, so that is the spring diameter in dactyls of a one-talent palintone, which is 385 mm in modern measures. Multiplying this by 19 to obtain the length of the stock gives c.7,315 mm or 24 feet long.
8 = 288 dactyls.
9 = 96 dactyls.
10 Hellenistic Fortifications from the Aegean to the Euphrates with revisions and a new chapter by N. P. Milner (Oxford, 1997), 10. I calculate it to be 5.8 m, approximately 19 feet.
11 Ober gives 8.4 m as the ‘stock length’ for such a catapult (table 4, p. 601), and this, together with the last note, highlights another point. We are not all using the same conversion scales for dactyls to feet or metres, and most scholars do not give the conversion ratios that they are using. I use 1 dactyl = 19.3 mm as Marsden.
12 Ober is referring to draw length though he does not use the term; the mechanics are explained by Cotterell, B. and Kamminga, J., The Mechanics of Pre-Industrial Technology (Cambridge, 1990): 180–92Google Scholar, and Denny, M., ‘Bow and catapult internal dynamics’, European Journal of Physics, 24 (2003), 367–78CrossRefGoogle Scholar, though the latter does not understand how a torsion catapult is spanned or ‘drawn’.