LIGHT CAVALRY AND CATAPHRACTS
Light cavalry of the Hellenistic period were generally mercenaries, called Tarentines. Although originally from Taras in south Italy, the name came to mean just a type of light cavalry armed with javelins and a small shield (Head 1982, pp. 115–16). The small shield of Macedonian style from Olympia, mentioned in connection with Cretan archers, could equally have been used by a Tarentine cavalryman. It is a moot point as to whether they wore helmets. We might presume that those who could buy their own helmet would have done so, but that they were not essential. Apart from battles, these soldiers were used chiefly for scouting by all the Hellenistic kingdoms and many Greek states.
A final type of cavalryman, who appears to have been used only by the Seleucid and Bactrian kingdoms, is the cataphract. This was a very heavily armoured cavalryman, who was covered from head to foot with armour, and who rode a horse that was also armoured. They were probably developed by the Parthians, and adopted by the Seleucids and Bactrians, the Greek kingdoms to their west and east in the later third century. Antiochus III had none at the Battle of Raphia in 217, but he did have 6,000 at the Battle of Magnesia in 190. He probably first recruited them following his travels through the eastern provinces in the late third century. There are no clear illustrations of cataphracts from this period, but there are illustrations of their armour on the Pergamum friezes, and an important find of actual armour has been made in Afghanistan (Bernard et al. 1980, passim).
The cuirass for the cataphract could have been of any of the metal types we have already looked at, no doubt fitted with pteruges, but the example from Ai Khanum in Afghanistan, which dates to the second century, is most unusual in that it is made of iron scales. The surviving shoulder piece is made of iron lamellar strips and came down onto the chest with a stud attachment, just like earlier shoulder-piece corslets and the Prodromi cuirass, to be secured with a thong. Various pieces of leather, linen and felt seem to have formed a separate arming jack, and were not directly attached to the cuirass as has been surmised for other metal cuirasses (Bernard et al. 1980, p. 61).
As for helmets, it seems that cataphracts wore a masked helmet which completely encased the head. An example is shown on the Pergamum frieze, and a possible Hellenistic example is in Belgrade’s Archaeological Museum (Russell Robinson 1975, pp. 107, 112). Such helmets were also apparently worn by chariot drivers (Sekunda 1994b, plates 4–5). Whoever wore them, they must have restricted vision dreadfully.
For arm and leg protection, tubular laminated guards were worn. These arm guards are depicted on the Pergamum friezes, and the Ai Khanum find has produced a leg guard made of iron. This guard was for a left leg, with the strips of iron overlapping upwards for greater flexibility like later Roman guards (Russell Robinson 1975, plates 502–4). The topmost part of the thigh was protected by a semi-circular plate, and there was a further plate covering the foot. Earlier arm and leg guards could well have been made of bronze. A small statuette from Syria also exists, which seems to show both arm and leg guards of this style (Sekunda 1994b, figs 32–3).
The Pergamum reliefs also show horse armour in the form of a chamfron (face guard) and plastron (chest guard) and these too are likely to have been for cataphracts, or perhaps for scythed chariots (Sekunda 1994b, fig. 54, plates 4–5). A further piece of armour from Ai Khanum, made up of very thin iron lamellae in a rough square shape, appears to be a horse plastron, although the excavator thought it might be a parameridion or thigh guard (Bernard et al. 1980, p. 61). Given the fact that the rider’s legs were already protected by the tubular leg guards and possibly pteruges, I think a plastron is more likely. It is highly unlikely that cataphracts used a shield as well as all this armour, and most probably they were armed with a spear for frontal assault. The Battle of Magnesia saw the cavalry and cataphracts of Antiochus III’s right wing break through the Roman line and pursue the fugitives to the camp. They were unable to return in time to salvage the collapse of the infantry phalanx, and one reason must surely have been that the cataphract horses would have been exhausted after one charge. The weight of armour, especially if much of it was in iron (not necessarily the case until well after Magnesia), would have protected rider and horse from missiles, and made them a formidable strike force, but must also have exhausted the cataphracts very quickly. The timing of their charge needed to be exact, as they probably could not have been manoeuvred again for any further action.
The chariot had gone out of use among the Greeks when horses had been bred that were big enough and strong enough to be ridden as cavalry. The same had happened in Persia, where chariots no doubt continued in use for ceremonial purposes; but, sometime before 400, the chariot made a comeback as a weapon of war. This new chariot was very different from those of former times. It was a four-horse chariot, whose horses and drivers were heavily armoured. The chariot itself was covered with scythes, and was designed to smash through enemy formations. Scythes projected in front of the chariot from the yoke poles, and also sideways from the yoke, one pointing horizontally, the other downwards. Two more scythes were attached to the axle, again horizontally, and pointing downwards. This latter probably revolved with the wheel to catch both ‘duckers’ and ‘jumpers’ (Livy XXXVII, 41). Persian chariots mentioned by Xenophon in his Anabasis had scythes projecting from under the box of the chariot, but these are not mentioned by Livy describing the Seleucid version and may have been dropped by then. It seems likely that they would have often got caught in the ground if the surface was at all uneven, and that may have been the main reason why Darius III had to level the ground for his chariots before the Battle of Gaugamela. It seems that the horses and drivers of these chariots were armed in much the same way as cataphracts.
Scythed chariots have usually been dismissed as a gimmick that did not work. Livy (XXXVII, 41) describes them at the Battle of Magnesia as ‘farcical’, but they remained in use by Persian and Seleucid armies for over 200 years. Their first known appearance at Cunaxa in 400 failed against disciplined Greek hoplites, who moved aside to let them pass, but in 395 at Dascyleum they scored a victory against hoplites who were panicked by the sight of them. Alexander the Great managed to break up the Persian chariot attack at Gaugamela with light troops, and this became the standard defence. Molon, a Seleucid rebel, used them against Antiochus III in 220, but Antiochus himself never used them against other Greek armies because he thought they could be easily countered. Against Rome at the Battle of Magnesia he thought the element of surprise would count in his favour, but Eumenes, King of Pergamum, was on the Roman side and told them how to deal with the chariots. They were again broken up using light troops. At the Daphne parade in 166, Antiochus IV had 100 six-horse scythed chariots and only 40 four-horse versions, so it is possible he was trying to make them more effective by increasing their size. It is difficult to be sure because this was a parade and not a battle (Sekunda 1994b, p. 26). Chariots continued to be used by the Seleucids until after 150, but are unlikely to have lasted into the first century BC, when the Seleucid Empire had been reduced to a Syrian rump. There is no evidence for scythe-chariot use by other Greek states. Livy (XXXVII, 41) states that Eumenes of Pergamum knew about how they worked in war, but does not suggest that he actually had any himself. There is a slight possibility that they were used by the Bactrian kingdoms in the east, but the terrain there is not really suitable for chariotry.
Elephants were used by the Indian army of Porus, which fought Alexander the Great in 326. Although Alexander was victorious, the elephants had caused heavy casualties among his men. It was rumours of larger elephant armies in India that caused the army’s revolt soon after. Alexander saw the advantages of the elephant, and began to recruit an elephant corps into the Macedonian army. Originally the elephant itself was the weapon, and it was made as imposing as possible. The elephants of Eumenes and Antiochus III had purple trappings, and Antiochus decorated his elephants with gold and silver and awarded them medals for bravery (Scullard 1974, pp. 238–9). If elephants were wounded they had an unfortunate tendency to run amok so, as well as the driver, a soldier or two was mounted astride the elephant’s back, armed with missiles to help protect it.
Later on, elephants were issued with armour, consisting of head pieces like horse chamfrons and leg armour similar to that worn by cataphract troops, although perhaps leather rather than metallic (Sekunda 1994b, plate 7). Livy (XXXVII, 40, 4) mentions headpieces with crests on them. Scale body armour for elephants was also used and features on a damaged statuette of uncertain provenance (Sekunda 1994b, figs 52–3). For offensive purposes the elephants’ tusks could be sheathed in iron (Arrian, Punica IX, 581–3).
As well as breaking up elephant charges with light troops and missile weapons, elephants could be disrupted by weapons placed in front of them. The elephants of Polyperchon (regent in Macedonia after the death of Antipater in 319) were once disrupted with planks lying on the ground, with nails pushed through them from underneath (Scullard 1974, p. 248). Ptolemy improved on this at the Battle of Gaza in 312 by attaching a series of caltrops (sets of spikes) to chains. These could then be quickly moved to where an elephant attack might come.
The best tactic was to make sure you had more and bigger elephants than the enemy. At Magnesia in 190 the Romans had sixteen elephants, but did not bother to use them as Antiochus III had fifty-four. Ptolemy’s elephants at Raphia in 217 were defeated because he had only 75 to Antiochus’s 102, although Ptolemy won the battle in the end. Ptolemy’s elephants were also defeated because they were African bush elephants, which are much smaller than the Indian elephants used by Antiochus. (Connolly 1998, p. 75). After Raphia, Ptolemy captured some Indian elephants, which he used in his army, and in 145 Demetrius II of Syria captured some African elephants from Egypt, which he also used in his army. Generally, however, early Hellenistic kingdoms – including Pyrrhus of Epirus – used the Indian elephant; only the Ptolemies of Egypt, cut off from supply by the Seleucid Empire, were forced to rely on smaller African elephants.
The idea of defensive troops sitting on elephants’ backs was enhanced in the early third century by placing small wooden towers on the animals’ backs. These enabled more men to be carried, and gave those men greater protection. As well as elephant defence, these men now became part of the offensive capability of the elephant. The mahout, or elephant driver, still had to sit outside the tower, astride the elephant’s neck. The earliest representations of towers are both from about 275. A plate from south Italy shows an Indian elephant with a tower containing two soldiers and may represent one of Pyrrhus’s elephants (Connolly 1998, p. 75, fig. 2). A statue of similar date shows an elephant attacking a Celt, and has been dated to the ‘elephant victory’ of the Seleucids against invading Celts in 273. The towers may well have been first used by Pyrrhus at the Battle of Heraclea in 280, but were soon adopted by other Hellenistic kingdoms (Scullard 1974, p. 104).
There are two basic types of tower: the original large tower for the Indian elephant, and a smaller type devised by Ptolemy IV for his African elephants. The Indian-elephant tower, as used by Antiochus III at Raphia in 217 and Magnesia in 190, is as wide as it is high and has three merlons per side in the crenellations. The early plate, sculpture, and an elephant medallion in the Hermitage all show this (Connolly 1998, p. 75, fig. 3). The African-elephant tower has a much smaller base to enable it to sit on the smaller elephant, but is twice as high as it is wide, to make up for the lesser height of the elephant, and it has only two merlons per side (Connolly 1998, p. 75, fig. 1). Livy (XXXVII, 40, 4) states that the elephants at Magnesia had four men in each tower and this is supported by the ‘elephant victory’ statuette, which has two shields attached to each side of the tower. Four armed men in a wooden tower is certainly possible, but was probably the maximum allowed. The statuette of the African elephant shown by Connolly has only one shield each side and suggests a crew of only two.
Ptolemy’s elephant crew at Raphia were armed with sarissas to poke at the opposition, but Antiochus III’s elephant crew probably had two sarissa men and two archers or javelineers. Large amounts of missile weapons could certainly be stored in the tower (Scullard 1974, p. 240). Fear was the elephants’ strongest weapon, but they did have other uses. Perdiccas used his elephants to assault the Camel fort of Ptolemy where they tore up palisades and threw down parapets (Diodorus, XVIII, 34, 2), but they were ineffective against stonework. Horses could not stand the sight or smell of elephants unless they were specially trained. After Demetrius the Besieger had been victorious with his cavalry on the right wing at the Battle of Ipsus in 301, he found himself cut off by Seleucus’s screen of elephants and unable to return to the battle (Diodorus XX, 113–XXI, 2). At the Battle of Magnesia, Antiochus had his horses trained to work with elephants, and each cavalry wing was supported by sixteen of them. To try and prevent light troops from getting close to an elephant and hamstringing it, each elephant was provided with a guard of forty to fifty men, usually archers or slingers (Polybius XVI, 18, 7). Also at Magnesia pairs of elephants and their guards were stationed in between blocks of the pike phalanx, to try and add some flexibility to this formation and to protect the flanks.
After Magnesia, Antiochus III was required to have all his elephants destroyed, but at the Daphne parade in 166 Antiochus IV still had thirty-six elephants equipped for war, as well as a four-elephant chariot and a two-elephant chariot – these last two items surely for parade purposes only (Sekunda 1994a, p. 27). Whether the Romans had not got around to making sure the elephants were destroyed, or whether Antiochus IV had been able to obtain more from Demetrius of Bactria, is uncertain. In 162 Gnaeus Octavius was sent out by Rome and he did destroy the elephants, although it cost him his life at the hands of an outraged elephant lover (Green 1990, p. 437). Further elephants do continue to appear in the sources, although some sources are unreliable. It seems unlikely that either the Seleucid or Ptolemaic Empires used them after c. 140. By then Parthia had blocked off supplies from India, and the African bush elephant was on its way to extinction. Pyrrhus of Epirus had famously used elephants against the Romans in the 270s, but there is no evidence for Epirote use after this time, and none for their use by Macedonia, Pergamum or any of the southern Greek states. It is almost certain that the Greek states of Bactria and India used them throughout their period of existence (down to perhaps AD 10), but their coins show only elephants, elephant heads and elephant scalps on helmets. They do not show elephants with towers or soldiers, which would prove the case.
As has been said earlier, surprise and the fear they caused were the greatest weapons of the elephant. These were the main factors in the elephant victories of the early third century. But when soldiers knew how to deal with them, they were easily managed and became an expensive liability. Once enraged or wounded, they were just as likely to inflict heavy casualties upon their own side as on the enemy. Neither the Romans nor indeed the Parthians ever really bothered with them.
Nearly all of our evidence for Hellenistic artillery – that is, bolt-throwers and stone-throwers – is literary. We have surviving Hellenistic manuals and descriptions in Arrian, Polybius, Livy, etc., of the equipment in action at the various sieges. Parts of some catapults have been found, mostly dating to the Roman period, and these also help with reconstructions. The main archaeological finds are the projectiles. At Rhodes and in other places, round boulders of specific weights, fired from catapults, have been found, and catapult bolts inscribed with Philip II’s name are also known (Connolly 1998, pp. 282–3).
According to Diodorus, the catapult was invented in 399 for Dionysius I of Syracuse (Campbell 2003, p. 3). Unfortunately, we don’t know quite what this machine was. The forerunner of the earliest catapult was the gastraphetes or ‘belly-bow’ devised by Ctesibius, probably towards the end of the fifth century. This was a large, composite bow mounted sideways on a stock, rather like a large crossbow. The arrow or bolt rested on a slider, which moved up and down the stock. The slider was pushed forward until a catch on it was fastened onto the bow string. The end of the slider was then rested on the ground, and the operator pushed with his stomach into the crescent-shaped end of the stock, using his weight to force the slider back, thus drawing the bow. An arrow could then be fitted and the catch released to fire it. It was a very slow and cumbersome effort.
The machines probably presented to Dionysius were similar but mounted in a base, with the slider being drawn back by a winch system. Biton describes four of these machines in his treatise, the first two designed by Zopyrus. The first, still called a gastraphetes, had a 9ft-long bow, and fired two 6ft bolts simultaneously. The second was a smaller version for easy transport to sieges and was called the mountain gastraphetes. The third machine was a stone-thrower, designed by Charon of Magnesia, which could fire a 5lb stone. The last, by Isidorus of Thessalonika, could fire a 40lb stone, using a 15ft bow. These stone-throwers had a sling fitted with a pouch, instead of the normal bowstring. All four machines were mounted on the base by a universal joint, which allowed the machine to be traversed, depressed and elevated with relative ease by one man.
These bow catapults, or ballistas, developed during the first half of the fourth century into the torsion catapult. In this design the bow was replaced by two wooden frames on either side of the stock, each containing twisted bundles of hair or sinew. A wooden arm was inserted into each bundle, and these formed the arms of the bow. After a few shots the elasticity of the bundles slackened, and iron levers were inserted top and bottom in order to retighten them (Marsden 1969, p. 81). It seems that machines of this type, capable of throwing stones, were not developed until the time of Alexander the Great. The last development seems to have been the use of curved arms for extra springiness. The first evidence for this is on the Pergamum frieze, so an introduction date of c. 200 seems likely. The earlier bow catapults continued to be used down until about 240.
The range of bolt-throwing catapults was about 500 yards, with the bolts being 2–5ft long, but they were really accurate only up to about 100 yards. Stone-throwers had a range of about 300 yards – perhaps only 200 yards for the largest – and came in a variety of sizes. The most popular engines seem to have been ten minas (4.4kg), thirty minas (13.1kg) and one talent (26.2kg) machines, these weights being the weight of the projectiles. Stones larger than that have occasionally been found, but they were probably for lifting and dropping by cranes, as Archimedes did at Syracuse in 212 (Polybius VIII, 5). Further developments, mentioned in the treatises we have, never seem to have got off the drawing board. Ctesibius mentioned a catapult with bronze springs, which did not slacken like the sinew or hair in the torsion catapults, but they seem to have proved to be too expensive to manufacture. He also designed a catapult operated by compressed-air-powered springs. This was fine on the drawing board, but could not be accurately manufactured with the techniques available at the time. The final invention in this field was a repeating catapult designed by Demetrius of Alexandria, but this machine was a failure because it was too accurate; all the bolts hit the same target and did not disperse. Its range was also somewhat limited.
Although these machines were designed principally for attack and defence during sieges, they were occasionally deployed on the battlefield. At Mantinea in 207 Machanidas the Spartan stationed catapults, probably bolt-shooters, all along his line in an experiment to counter greater Achaean numbers. Philopoemen and the Achaeans charged the catapults as soon as they saw them and, since they were difficult to move, they were almost immediately destroyed or overrun and played no further part in the battle. In 198 and 191 Philip V and Antiochus III, respectively, used catapults in defensive positions against the Romans at the Aous Gorge and at Thermopylae. In both cases, the Romans found it hard to approach these defences from the front, but they were easily outflanked and captured. These would have been expensive losses, since the price for these machines appears to have been about 500–2,500 drachmas each (Philon 62, 15). For the most part the machines were installed in fortifications for defensive purposes, and could also be used in attack against such fortifications. The best-known Greek attack was that by Demetrius the Besieger against Rhodes in 305–4, at which he used a huge tower filled with catapults and ballistas of all sizes (Connolly 1998, pp. 281–5). The best-known defence was that of Archimedes at Syracuse in 213. Apart from a vast array of bolt-throwers and stone-throwers, Archimedes also had cranes which dropped huge boulders on the attacking Romans, and giant grappling hooks which pulled Roman ships out of the water and then dropped them (Connolly 1998, p. 294).
THE REFORMS OF THE 160S
The Hellenistic warfare we have been describing in this chapter, principally the pike phalanx and the heavy cavalry, was an effective form of warfare that lasted successfully until 168. In that year, the Romans annihilated the army of Perseus, King of Macedon, at the Battle of Pydna. The legionary army proved itself more effective on the day, and this has led Sekunda (1994b, 1995, 2001) to suggest that the remaining Ptolemaic and Seleucid kingdoms in Egypt and Syria remodelled their armies on Roman lines. The evidence for this is actually slim.
A stele from Hermopolis describes new ranks and names for formations, which Sekunda (2001, p. 21) argues is the adoption of the Roman maniple or double century, but the top and bottom of the stele are broken and we do not know what sort of soldiers these are. The semeia which is mentioned could simply be a new word adopted for the syntagma or speira, words previously used to describe a phalanx block of 256 men. The later tacticians like Asclepiodotus and Aeneas use the word semeia, but they are still describing a Hellenistic pike phalanx. The use of the word semeia does suggest, however, the use of standards, so it may be that military standards were introduced into the Ptolemaic army at this time.
Further evidence for this ‘reform’ is provided by the Kasr-el-Harit shield and various stelai from Sidon. The Egyptian shield has already been mentioned by me as a descendant of the Greek thureos, and the soldiers depicted on the stelai are also mercenary thureophoroi, who have no relation to the regular Hellenistic phalanx (Sekunda 2001, pp. 65, 80).
The evidence for the Seleucid Kingdom is almost entirely contained in a sentence of Polybius (XXX, 25, 3), where he is describing the Daphne parade of 166. Here he says that there were 5,000 men equipped in the Roman manner with chain mail. These marched separately from the 20,000 men of the phalanx, and were clearly a different unit. Some commentators have suggested these men were just a bit of a gimmick, like the elephant chariots that also featured in the parade, but Sekunda is surely right when he states that they were a genuine military component. They were not armed as Roman legionaries, however. Apart from the chain mail, there was nothing to suggest that these men used the Roman pilum or shield, or fought in maniples. A unit of 5,000 men could easily be part of the phalanx, but the fact that they were placed at the front of the parade with other obviously mercenary troops – Mysians, Thracians and Galatians – suggests that these men were mercenary thureophoroi, armed with the thureos shield and spears.
A final pointer which seems to confirm that the Seleucid and Ptolemaic kingdoms did not reform along Roman lines was that the Macedonian-style pike phalanx continued to be used, both in the later campaigns by the Seleucids against the Jews and in Mithridates of Pontus’s campaigns against Rome. Mithridates did arm half his army in Roman fashion, but he also seems to have been the last man to employ the Hellenistic pike phalanx.
Further mention should be made of chain mail as a form of body armour, as it was clearly used by some soldiers in the Seleucid Kingdom. Apart from the Daphne parade mentioned above, some or all of the Seleucid phalanx was armoured with chain mail at the Battle of Beth-Zacharia in 162 (Maccabees I, 6.35). Appian (Syrian Wars 30–6) also suggests that cataphracts may have worn chain mail at Magnesia in 190, but he seems to confuse Celtic cavalry with cataphracts, so this idea is perhaps best ignored.) One of the stelai from Sidon mentioned above also has a soldier in chain mail, which may indicate use by the Ptolemaic Kingdom as well (Sekunda 2001, front cover, p. 69).
Chain mail was a Celtic invention of about 300, consisting of rows of interlocking iron rings, each ring passing through two above it and two below it to give a strong but flexible defence. Rows of punched rings usually alternated with rows of butted or riveted rings (Connolly 1998, p. 124), the latter being stronger. Each ring is usually 8–9mm in diameter. The shape of the later mail cuirasses adopted by both Romans and, presumably, Greeks was similar to the shoulder-piece corslet, with two shoulder flaps coming over the shoulders and being fastened down onto the chest. Chain mail corslets appear on the Pergamum frieze, where they probably represent armour captured from the Galatians. The date of c. 170 for this monument shows that this would have been the type of cuirass adopted by the Seleucids in the 160s. Wealthier Roman soldiers wore this form of cuirass, and it seems likely that all legionaries were issued with chain mail in c. 123, after Rome had inherited the wealth of the new province of Asia. It would have been very expensive for Antiochus IV to equip 5,000 soldiers in chain mail for the Daphne parade, which is one reason why Polybius remarks upon it.
Chain mail is an excellent defence, combining the flexibility of leather with the resilience of iron plate, and it lasted as a defence until the Middle Ages. It was even revived as a defensive material for tank crew in the First World War. It would have been quite heavy to wear a knee-length corslet of mail, but the defensive capabilities were excellent. Pointed weapons would be caught in a ring and held, while edged weapons also would not have much penetrative power, especially as the mail would have been worn over a leather jack. As far as the Greeks go, however, it was the last innovation before impotence led to a gradual absorption into the Roman Empire. Macedonia and Greece were annexed in 148, Seleucid Syria in 64 – although it had ceased to be any sort of power since the 120s – and Egypt in 30, although it too had existed since 168 only by the will of Rome.