previous next


TORMENTUM was a general name for artillery among the ancients.

I. DESCRIPTIVE.--The two main classes of tormenta are those which discharged their missiles (1) horizontally (εὐθύτονα), or practically so; (2) at an angle (παλίντονα) with the ground. But these are technical terms. The ordinary names of the engines, both in authors (Diod. 16.74) and in inscriptions (C. I. A. 2.250; 471, 46; 733 B), were ὀξυβελεῖς and liqobo/loi. The former shot arrows, hence their name ὀξυβελεῖς, sc. καταπάλται (not καταπέλται: cf. C. I. G. 2360, 25), and were called catapultae by the Romans: the latter discharged stones (πετροβόλοι, λιθοβόλοι), and were called in Latin ballistae. The arrow discharged from catapults is sometimes called catapulta (Plaut. Pers. 1.1, 28; Nonius, p. 552); and the stone discharged from ballistae called ballista (Plaut. Trin. 3.2, 42). The terms catapultae and ballistae were probably obtained from the Sicilians; and they are used by historians as the two chief classes of engines (Tac. Ann. 12.56; Gel. 7.3). It must, however, be remembered that catapulta was also used as a generic term embracing both classes (Caes. Civ. 2.9, 3; App. Bell. Mithr. 34; Sidon. Apoll. Carm. 22, 123).

The structure of both the engines is in the main principles the same: both, as the name tormentum points out, deriving their force from torsion, not from the elasticity of a bent body like a bow. We shall first attempt to describe a εὐθύτονον or catapulta (in the narrow sense), and afterwards set forth the points of difference between it and a παλίντονον. The two chief ancient authorities who supply materials are Heron's Βελοποιϊκά, and Philon's fourth book Περὶ βελοποιϊκῶν. Both writers flourished about 250 B.C. according to Rüstow and Küchly; about 100 B.C. according to Graux (Rév. de Philol., N. S. 3.92).

1. Catapulta, also apparently called scorpio in Caes. Gal. 7.25, 2; Bell. Afr. 29, 4; Vitr. 10.10, 15; Sall. ap. Non. p. 553: cf. Heron, § 3. This engine consisted of three parts, which we may call the Frame (πλινθίον, capitulum), the Pipe (σύριγξ), and the Support (βάσις). See Fig. 1.

a) The Frame consisted of two strong horizontal beams, a, b, into which four other vertical beams were morticed, c, d (called παραστάται), e, f (called μεσοστάται, medianae). The beams, a, b, were called περίτρητα, paralleli, peritreti, because into the top and bottom of those beams in the centres of the two outside compartments, g, were bored circular holes (τρήματα, foramina), into which were placed the strings (τόνοι, funes) which gave the force which projected the missile. These strings seem to be called vincla in Tac. Hist. 3.23. The diameter of these holes was 1/9 of the length of the arrow, and formed the standard measurement according to which all the other parts of the engine were proportioned. Into these holes were placed nuts, h (σώληνες), sometimes wooden, sometimes of metal, which were circular at the bottom so as to fit into the holes, rectangular in the middle m, and again circular at the top, of the same diameter as the standard. The nuts were rectangular in the middle partly to prevent them slipping down through the holes, partly that when required they could be turned round by a wrench, and so the strings tightened; hence this rectangular part got the name of τριβίς, from its friction against the frame. Across the

Fig. 1. Catapulta or Scorpio. (From Rüstow and Köchly.)

middle of the top of the nut ran an iron bar, n (ἐπιζυγίς, cuneolus ferreus), round which the strings of the engine were stretched. The strings were generally prepared from the sinews of animals, and hence these engines are called νευρότονοι (C. I. A. 733 B), though sometimes we hear of women's hair being used (App. Pun. 93; Hero, § § 26 ff.; Philo, § 12).

The instrument (Fig. 2) used for stretching the strings was called ἐντόνιον, and consisted of a large wooden frame with a windlass (ὀνίσκος, sucula) at each end, two beams d d as strengtheners,

Fig. 2. Ἐντόνιον.

and a centre compartment just the size of the frame of the catapult. Into this centre compartment the frame, duly provided with its nuts, was fixed and firmly wedged, k k k k. The end of the string was fastened to one of the iron bars of the nut at n (Fig. 1), drawn through the other nut and fastened round one of the windlasses (suppose b). This string was then stretched till it became 2 of its original diameter, and fastened by a clamp (περιστομὶς) to the bar opposite to the one to which it had been originally fixed. Then it was loosed from the windlass, drawn round the bar through the opposite nut, and again strained by the windlass c, fastened by another cramp, and so on passing from windlass to windlass till the whole available hollow portion of the nuts was filled with layers (δόμοι) of the string. The number of layers was generally about 10. When that was completed, the end of the string was fastened by a very strong clamp. Vitruvius says [p. 2.854]that each string should be stretched till it gave the same note (10.18 (12), 2; cf. 1.1, 8. Compare Hero, § 28; Philo, § 17).

Through these masses of string from the side turned towards the enemy the thinner ends of two long pieces A, A (Fig. 1) of nonelastic wood, which formed the arms (a)gkw=nes) of the bow, were thrust, so that when the engine was not being worked the thicker ends (πτέρνα) rested on the outer side of the μεσοστάται against an iron-plated knob (ὑπόπτερνις). On the other side of the frame, the arms rested about at their centre or two-thirds of the way from the point γ, against a curve q (κοίλη, curvatura) in the παραστάται, which latter had a bulge on the outer side, so that they should not be wanting in strength. To the ends of the arms γ γ was fixed a very strong string (τοξῖτις), called apparently libramentum in Tac. Hist. 3.23, which was the string by which the arrow was shot (Hero, § 30).

b) Now we come to the Pipe, which projected backwards from the centre compartment of the frame. It consisted of two parts: (1) the pipe proper (σύριγξ in the narrow sense, canaliculus) and (2) what we may call the projector (διώστρα). The pipe proper was a long narrow trough-like construction of wood, open at the end towards the enemy. At its other end it had a windlass for stretching the string, worked by hand-spikes (σκυτάλαι). Running in the pipe, which was dovetailed for about two-thirds of its length, and fitting into this dovetail, was another smaller trough-like construction called the projector (διώστρα), into which the arrow was placed. The trough in the case of this διώστρα was concave, and not angular. At

Fig. 3. Plan and section of the “Pipe.”

the end of the διώστρα was a hook (χείρ, epitoxis), of which a horizontal and a vertical section are given herewith (Fig. 3). It moved on an axle (λ λ) working through a specially inserted frame, θ θ (στήματα). The hook, axle, and frame taken together appear to have been called χελώνιον. The hook consisted of two horizontal prongs with vertical ends; and a hindmost part (ν) very much heavier than the fore part; so much so that, in order to keep the fore part down, the hinder part had to be prized up and supported by a handle, φ (σχαστηρία, manucla), which revolved horizontally on a vertical axis, π (called περόνη). On the hinder part of the projector was a ring, ξ, through which one end of a strong cord (ὅπλον, καταγωγὶς) was fastened, the other end being fixed to a windlass. (Hero, § § 5 ff.; cf. Philo, § § 52 ff.)

Now, when the engine was to be used, the projector was pushed forwards till the hook, prized up by the handle (σχαστηρία), could catch the projecting string (τοξῖτις). Resting against this string, and in the trough of the projector, was placed the arrow. The projector was then along with the string drawn back by means of the καταγωγὶς and the windlasses as far as was required, and the windlass made fast. so that the projector could not move. A. Müller in Baumeister's Denkmäler, p. 547, and Droysen, Gr. Eriegsalt. p. 196, suppose that the pipe had a series of teeth, so that the projector could be fixed at any given point, as we shall see was the case in the γαστραφέτης (see below, § 3). This. is probable enough à priori, but we do not know the evidence for their opinion, and there was not the same necessity for the teeth in the larger engines as there was in the hand-strung γαστραφέτης. When all was now ready for the shot, the handle was pushed violently from under the heavy side of the hook, which must have been very heavy indeed, for it appears. that this side fell down by its own weight, and so released the string, which shot forward the arrow with great velocity. This appears to be what Hero ( § 6) means by “they let the hook loose by tearing away the handle” (ἀπέσχαζον τὴν χεῖρα σπαράξαντες τὴν σχαστηρίαν); other-wise we should suppose that the use of the handle was only to prevent accidental discharge while the string was being drawn back, and that, after it was loosed, as a general rule a blow of a hammer on the hinder part of the hook or something of the kind would have been necessary to release the string.

c) The Base of the catapult, which is described by Rüstow and Köchly, and of which they give their principal illustration (op. cit. fig. 106), consisted of two supports. Such were required only in the case of very heavy engines, and were not much used, among other reasons because the limits of elevation within which they could be discharged were very circumscribed; they were confined to that allowed by the height of a pin which fixed the frame to the foremost support, and this would not allow a change of elevation of more than a few degrees. The ordinary catapult and that principally described by the ancients had only one support, as, in the subjoined illustration, taken from Baumeister, of a catapult built according to the ancient authorities by the Heidelberger Philologen-Versammlung. The base consisted of a beam, q (ὀρθοστάτης, columella), supported on four feet s, by four stays, r. In the top of this beam was a long circular pivot, u, which passed through two horizontal sides of a wooden frame, t, whose vertical sides projected considerably beyond its upper horizontal side. Through the vertical sides above the horizontal sides and parallel with them ran a round iron bar on which the pipe rested; and while the whole upper part of the engine could revolve horizontally on the pivot u, it could be lowered and elevated vertically by revolving on this bar. At. the end of the pipe was a stay, v (ἀναπαυστηρία), which could move up and down on the support w (ἀντερειδίς), which latter was attached by a ring to the main beam, q. The limits of change of elevation of such an engine must have been at all events 8°.

The length of the arrow gave the technical name to the size of the engine: so that catapults [p. 2.855]were classed as three-span (τρισπίθαμος), two-ell (δίπηχος) = four-span, five-span (πεντεσπίθαμος), three-ell (τρίπηχος)=six span; that is, 27 inch, 36, 45, 54. This gave the diameter of the τρήματα in the frame as about 3, 4, 5,

Fig. 4. Catapult. (Baumeister.)

6 inches. Let us call this diameter x; then we can fix the rest of the measurements of the machine, e. g. height of the frame 5.5 x, depth 2 to 1.5 x, breadth 6.5 x, length of pipe 16 x, of ἀγκῶνες 7 x each, thickness of each of the νεῦρα 1/3 to 2/9 x; minimum breadth for working 13 x, height 18 x, depth 20 x. The weight of a τρισπίθαμος was about 85 lbs., and its arrow about 1/2 lb., and it required two or three men to work it; the weight of a τρίπηχος was about 5 1/4 cwt., its arrow over 4 lbs., and it required five men to work it. The three-span catapult of Agesistratus shot 3 1/2 stadia = 2210 feet (Athen. de Mech. p. 8 Wescher), but that was considered something very marvellous. At 1000 feet an arrow from a three-span catapult would be driven 2 inches into a board (Rüstow and Köchly, Kriegsschriftst. i. p. 330, note m): so that, on the whole, we may take the ordinary effective range at about 1200 feet, the actual distance the arrow would reach being somewhat over this (ib. 328, note h). The price of a two-ell catapult they estimate at about 480 drachmas, about £20, reckoning the drachma as a franc.

2. The Ballista (παλίντονον): cf. generally Hero, § 32; Philo, § 6.--The principle of this engine was precisely the same as that of the catapult, the only essential differences being (1) that the hinder part of the pipe rested on the ground to which the pipe itself was inclined at an angle of 45°; (2) that the wooden arms (ἄγκωνες) in the position of rest were not parallel with the ground as in the case of the εὐθύτονα, but inclined at an angle of 30°, hence the term παλίντονον ( “strung at an angle” ). The frame consisted of two smaller frames (ἑνατόνια, ἡμιτόνια), A A and B B, each of which held one of the sets of strings; these frames were bound together by two strong beams (κανόνες), a a and b b: indeed the whole engine was much larger and in all its parts stronger than the catapults. It was used to discharge beams or stones; accordingly it is the weight of the stone which gives the diameter of the τρήματα in this case. Along the pipe, C C, which had no continuous bottom, but had its sides (σκέλη), c d, bound together by pegs (διαπήγματα), extending longwise were narrow bars of wood (πτερύγια), which formed the support for the δίωστρα to run on. Chiefly, as it seems, on account of the

Fig. 5. Ballista. (A. Müller, in Baumeister.)

ladder-like appearance the διαπήγματα presented, the pipe in this engine was called κλιμακίς. The string (τοξῖτις) extending from the ends of the ἀγκῶνες was twisted like a rope, and had at its centre a ring (not represented in the plate) which was caught by the χείρ.

It would be tedious to give in detail all the various measurements of the parts of a ballista suffice to say, that the diameter x of the τρήματα in dactyls (1 dactyl = about 3/4 inch) was estimated by the formula x = 11/10 3√100 w, where w is the weight of the missile in minae (1 mina = about 1 1/2 lb.); that the length of the arms was 6 x each, of the τοξῖτις 12.6 x, of the κλίμαξ 16 x; and that the space required for the engine was at least 20 x in depth, 13 x in breadth, and 17 x in height. The size of the engine varied according as the missile was 10, 15, 20, 30, 50, 60 minae: the latter (= 1 talent) was the heaviest missile that was ordinarily used: engines larger than this, as that of Demetrius (Diod. 20.48) or of Archimedes (Athen. Deipnos. 208 c), which threw three talents, were quite exceptions and of little practical use. The average range was probably about 400 yards or a little more, but a large 60 minae ballista appears to have been barely able to throw 220 yards (Droysen, op. cit. p. 204). The price of a 10 minae ballista Rüstow and Köchly reckon at 4000 drachmas = £160. That the ballistae cannot have been much used in the field may be proved from their weight; so that they always appear in considerably less numbers than the [p. 2.856]catapults. At New Carthage we read that Scipio had 120 large catapults and 23 large ballistae (Liv. 26.47, 5)--numbers which probably in themselves are very much exaggerated: at Jerusalem the Jews had 300 catapults and 40 ballistae (B. J. 5.9, 2). Examples of the working of ballistae are given in Bell. Hisp. 13, 8, and Joseph. B. J. 3.7, 23; it appears that very considerable precision of aim could be acquired by the scorpiones (Bell. Afr. 29, 4; Caes. Gal. 7.25, 2).

Philo ( § § 17 ff.) mentions a great many points in which these tormenta were difficult to work and liable to break down. The frame was often broken in stretching the strings,--itself no easy task, taking considerably over an hour, and requiring the ἐντόνιον, which was not always at hand: the bars round which the strings were fastened used to cut the strings: the tension of the strings used to get loosened and could only be conveniently tightened by screwing the nuts round horizontally with a wrench--a very temporary help, as the elasticity of the strings soon got exhausted thereby; and so on. Philo invented a means of tightening the strings by a frame which could be narrowed by means of wedges; but it does not appear to have been much used. Ctesibius (Phil. § 14) replaced the strings by metal wires (χαλκέντονον); and also we are told that as one of his improvements he used compressed air (ἀερότονον), but there is no clear account of the exact nature of this latter device. The description of the so-called ballista in Amm. Marc. xxiii 4, 1, if it can be explained at all and is not pure “bombastische Confusion,” as Rüstow and Köchly (Kriegsschriftsteller, 1.414) call it, is certainly not of such a nature as to lead to any essential alteration of the description given above from such capable writers as Hero and Philo.

3. The γαστραφέτης or “stomach-bow” (cf. Hero, § § 3 ff.; Bito, p. 61 Wescher) derived

Fig. 6. γαστραφέτης. (Rüstow and Köchly.)

its name because it had to be pressed against the stomach and the ground or a wall, when it was being strung. The accompanying cut gives an idea of it. It was not strictly a tormentum, as its force was got from the elasticity of a bow: it was in fact a cross-bow, with a διώστρα virtually like that of the catapult. The novel feature of it was that the sides of the σύριγξ had a series of teeth, into which two little prongs (κατακλεῖδες) on each side fitted, so as to hold the δίωστρα at just the point required, and to do so with as little loss of time as possible. It was probably the same as the arcuballista of Vegetius (2.15, 4.22). Droysen (l.c.) says it was called σκόρπιος. The so-called βαλίστρα mentioned by Procopius (Bell. Goth. 1.21) was a bow, or most probably from the description a species of γαστραφέτης: and similar in principle, but on a very large scale and worked by windlasses, was the ballista fulminalis of the treatise De rebus bellicis, 8, 10, attached to the Notitia (cf. Marquardt, op. cit. 524, note 2; and Rüstow and Köchly, Kriegsschriftsteller, p. 410). The “four-wheeled ballista” (ib.; cf. p. 418) is said to have shot its arrows “not by strings but by rigid bars” (non funibus sed radis). This “riddle,” as Rüstow and Köchly call it, still awaits solution.

4. The onager (cf. Amm. Marc. xxiii, 4, 4, who calls it scorpio).--This appears to have been a Roman construction, and we only hear of it in post-Constantinian times. It may be described as a horizontal one-armed ballista, which shot small stones. The name is said to have been

Fig. 7. Onager. (Marquardt.)

derived from the fact that the wild ass in its flight dashed back stones with its hoofs on its pursuers. The strings which supplied the force were stretched horizontally, and the arm (ἀγκὼν) inserted vertically into them. When the engine was used, this arm, by a string attached to a point near the top, was pulled down by a windlass till it was horizontal, and then secured by a hook, the missiles being hung in a bag at the extremity. Then the hook was struck away with a hammer and the missiles discharged. The arm struck against a bag full of some soft substance attached to the front part of the machine, reaching about 2/3 of the way up the arm. This would have been a rather hard instrument to aim with, if it were not that it threw a number of stones.

II. HISTORICAL.--Pliny (Plin. Nat. 7.56) attributes the invention of catapults to the Syrophoenicians; but there is no corroboration of this statement. The passage in 2 Chronicles 26.14, 15, where it is said that Uzziah prepared “slings to cast stones,” probably dates from not earlier than the fourth century. In the Hellenic world tormenta first appear in the great preparations made by Dionysius against Carthage in 399 B.C. (Diod. 14.42, 43), and in the next year they were used in the siege of Motye (ib. 50). It was from Sicily that they came into Greece proper (Plut. Apophtheg. 219=2.191). The first mention of them there is in a list of articles contained in the Chalcothêcê in Athens (C. I. A. 2.61, 37), of date between 356 and 348. In 340 we read that the Perinthians borrowed artillery from the Byzantines (Diod. 16.74), and the siege of Byzantium in the same year by Philip of Macedon is the first [p. 2.857]occasion we hear of the use of artillery in Greece in any extensive form. Athenaeus, the writer on artillery (p. 10, Wescher), notices the reign of Dionysius in Sicily and the siege of Byzantium as marking epochs in the use of siege-engines, Polyeidus of Thessaly being one of the most celebrated engineers (cf. Grote, 11.262). On this occasion we hear only of καταπάλται ὀξυβελεῖς (Diod. l.c.); the first mention of λιθοβόλοι appears to be at the siege of Halicarnassus by Alexander in 334 (Arrian, 1.22, 2).

During the period of the Diadochi artillery reached its highest perfection among the ancients. The engines are repeatedly mentioned (Diod. 18.12, 51, 70, 71;--C. I. A. 2.807 b, 129 ff.; 808 d, 53 ff.;--Plb. 4.56, 3; 5.88, 7; 99, 7); and artillery practice (καταπαλταφεσία) became a regular part of the military training of the ephebi (C. I. G. 2360, 25). The Romans did not make any decided improvement or invention in military engines till late in the Empire. Caesar was quite inferior to the Massiliotes in artillery (Bell. Civ. 2.2, 5), and after the battle of Pharsalia had to get engines from Greece and Asia to besiege Alexandria (Bell. Alex. 1, 1). It was in siege-work, both attack and defence, but particularly defence, that these engines were employed (Liv. 26.6, 4; 27.15, 5;--Plb. 8.7, 6). They were altogether too heavy and cumbersome to be used very extensively in the field; if they were used in the field, it was only for the attack or defence of some strong position (Caes. Gal. 2.8, 4; 8.14, 5; B.C. 3.56, 1; B. Afr. 31, 6), or protecting some movement such as crossing a river (cf. Arrian, 1.7, 8; 4.4, 4). During the Roman Empire each legion (Tac. Hist. 3.23; D. C. 65.4), and perhaps each praetorian cohort (Tac. Ann. 12.56), had its own engines; and in the time of Vegetius (l.c.) each century of the legion had a carroballista, a large engine drawn by mules and requiring eleven men to work it, and each cohort an onager.

Into all the minutiae of the construction of these engines it would be impossible here to enter. For them readers must be referred to Rüstow and Köchly, Geschichte des griechischen Kriegswesens, 1852, pp. 378-405; to their edition of the Griechische Kriegsschrifsteller, 1853, vol. i., containing Hero's and Philo's Βελοποιϊκά, pp. 187-346, and Vitruvius, x. chaps. 15-18 (10-12), with a valuable translation and notes; to Wescher's Poliorcétique des Grecs, 1867, for Athenaeus and Bito (pp. 1-68); to A. Müller's article on Festungskrieg und Belagerungswesen in Baumeister's Denkmäler, 1.525 ff.; and to Droysen's Die griechischen Kriegsalterthümer, chap. ix. pp. 187-204.


hide Display Preferences
Greek Display:
Arabic Display:
View by Default:
Browse Bar: