Tarantine cavalry

Luke Ueda-Sarson

A print version of this article was published in Slingshot, 236 (September 2004), pages 21-25

This page created 25 October 2002, and last modified 14 March 2005 (to put in the above note about appearing in Slingshot).

The Hellenistic tactical manuals of Asklepiodotos, Aelian and Arrian all make mention of a type of called "Tarantines", and such cavalry figure prominently in Polybios' acount of the fighting between the Achaian league and the Spartans at the 3rd battle of Mantineia. The name clearly derives from Tarantum, the large and powerful Greek city on the west coast of Italy formerly named Taras (modern Taranto), whose conflicts with the native Italian peoples were to prove an alluring attaction to military adventurers from the Greek peninsula in the late Classical and early Hellenistic eras. By Polybios' day however, such "Tarantines" did not come from Tarantum; rather any cavalry that fought in the manner of Tarantine cavalry were described as such.1

The first literary account of such cavalry that has come down to us appears to be the body of horse mentioned by Diodoros serving Antigonos One-eye in his campaigns against Eumenes. As these are mentioned (19.29.2) as having "come up with him from the sea" (ie. with Antigonos from the Mediterranean) when they are first introduced to the reader, they are quite likely to have come from Tarantum itself. They are described as being "men picked for their skill in ambushing, and very well disposed to himself", seemingly more than just the usual run-of-the-mill mercenaries. At Paraitakeni the "Tarantine" horse is said to have been 2200 strong, but it seems clear that this figure also includes the 2000 Median longchophoroi also serving in Antigonos' army and who are missing in the Paraitakeni troop list, as elsewhere (Diodoros 19.39.2), they are described as being 200 strong and accompanied by Medes numbering 2000. This small band of Tarantines was instrumental in Antigonos' successful conclusion of the campaign, as they and the Medians were detailed to captured Eumenes' army's baggage at the battle of Gabene; a task in which they succeeded admirably.2 This resulted in Eumenes' Macedonian soldiers turning him over captive to Antigonos in exchange for the return of their baggage. The utility of these soldiers appears to have led to them being incorporated as a permanent part of the army: Antigonos' son Demetrios used Tarantines in Greece (Polyainos 3.7.1) and at the battle of Gaza in 312 BC Demetrios had 100 Tarantines organised in three squadrons as an extreme left wing flank force (Diodoros 19.82.2). Without wishing to overemphasise the exactness of the number, it may be noted 100 men is the equivalent of three turmae, the standard Italian cavalry troop

Paraitakeni was unlikely however to have been the first time that a Macedonian force had been acquainted with Tarantine cavalry. While Alexander the Great was busy conquering Asia, his uncle, Alexander of Epiros, was campaigning in Italy at the request of the Tarantines (Justin 12.2). Alexander of Epirus owed his throne to Philip of Macedon (Justin 9.6.7). Alexander the Great himself spent some time in Epirus with his uncle (and brother-in-law) while in exile before becoming king, and the fate of Molossian Epirus was from then on intertwined with that of Macedonia. It appears just as likely that it was as a result of these contacts that Macedonian military practices spread to Epirus than than their being introduced by Pyrrhus over 30 years later

It is possible that some of the Tarantine cavalrymen who served with Alexander of Epirus ventured east after his death in action against the Bruttians (ca. 330 BC) - Alexander of Epirus is said (Curtius, 8.1.37) to have complained that while he was fighting real men in Italy, his nephew had been fighting nothing but women in the east! The prospects of such easy service, not to mention the quantities of loot being generated, proved a far greater attraction than the great distance proved a disincentive to tens of thousands of other Greek mercenaries, and Tarantines are no less likely to have been found campaigning in India as Arkadians or Rhodians. By the time Alexander the Great was campaigning near Skythia, his army had come to include units of "hippakontistai" - ie. mounted javelinmen. These are usually assumed to have been Asiatics - Medes and the like, but one cannot rule out the possibility that some were Tarantines, since after Gaugamela, the sources are largely silent as to which regions most of his troops came from. However, any Tarantines he employed are just as likely to have been included in the Greek mercenary cavalry listed separately from the hippakontistai in actions such as that against the Skythians by the Jaxartes.3

4 Tarantine coins

Numerous coins from Tarantum exist showing armed horse riders whom we may take as cavalrymen, so that the appearence of Tarantine cavalry does not have to be reconstructed from literary texts alone. Pictured above are four typical examples. At the top left is an example dated to approximately 330 BC, the period of Alexander the Great. The rider is depicted nude (the horse similarly lacks any blanket or padding) but he clearly carries three large javelins or small spears, with one held in the weapon hand pointing downwards and the other two with the shield; the shield itself is round and to judge from the prominent rim, appears to be the standard Greek aspis as carried by hoplites. Because of the obvious nudity of the figure, it is uncertain if we can rule out a helmet being part of this man's normal battlefield kit - such artistic conventions are presently poorly understood.

Moving clockwise around, the second example is dated to some time in the early 3rd century BC, and depicts a cavalryman from the other side. The face of the shield is this visible, and carries the dolphin blazon often shown on such coins (a hippocampus is another motif not infrequently seen); two javelins are carried with the shield. The large shield hides the ride's face, torso and weapon hand; the helmet appears to be of the Boiotian style, and bears a flowing crest. The third coin shows a very similar rider, but the shield blazon is an 8-pointed star; this coin is thought to date from the 270s BC, during the time of Pyrrhus' occupation of the city.

The last coin dates from after the time of Pyrrhus. The helmet appears to some sort of Attic variant, again bearing a flowing crest, and it is possible that the rider wears a cuirasse, although a muscular nude torso is at least as possible an interpretation. Only one javelin can be seen, although others held in the hidden shield hand can not be ruled out. In general terms the appearence of the cavalrymen is quite consistent throughout the series - a rider armed with javelins rather than a long spear, and carrying the large round rimmed shield usually associated with Greek hoplites. Helmets appear to be usual - if not essential; and other body protection is possible. Strabo (6.3.4) mentions that when at its most powerful, ca. 450-350 BC, Tarantum could field 3000 horsemen and 1000 hipparchoi, literally "horse rulers" and thus used in the sense of "cavalry commanders". As 1000 cavalry generals is clearly nonsense, Duncan Head in his Armies of the Macedonian and Punic Wars interprets these as heavy cavalry, although perhaps 'stud owners' might be an equally valid explanation.

Tarantine cavalry were not the only cavalry from the area using shields in the 4th century BC, as they are depicted on a number of Campanian red-figure vases of the period. Indeed shields seem to have been a common feature of Italian cavalry warfare, for the Greeks no less than the natives whom they no doubt imitated. This is in contrast to the Greek mainland, where the usage of shields by cavalry - at least when fighting mounted - was far from normal until well into the 3rd century BC. However, contrary to Duncan's speculations, I consider it unlikely Tarantines were reponsible for the introduction of cavalry shields to Greece at that time, because the typical Greek cavalry shield of the 3rd century was a flat spined shield rather than the heavily rimmed and bowl-shaped 'Argive' aspis. Therefore such spined shields were probably adopted as a result of contact with the invading Galatians in exactly the same manner as the thureos was, since both the thureos and the round spined shield were used by the Galatians. I have yet to see a coin from Taranatum showing a spined shield.

4 more Tarantine coins

On the basis of coins dating from 380 to 345 BC showing horsemen apparently carrying a small round shield, such as the one shown above to the top left, Duncan concluded that Tarantine shields increased in size during the 4th century (a view also espoused more recently by Sekunda in his book on the Seleucid army, see note 5 below). However, this is more apparent than real. Coins showing apparently small shields were produced throughout the period. The example above at the top right dates from 344 to 340 BC, while that to the lower right dates from Pyrrhus' occupation, ca. 275 BC. As even the small shields feature the prominent rim characteristic of the standard Greek Argive aspis, the natural conclusion is that the shield is shown smaller than life size, so that the nude torso of the rider may be seen (what function the heavy rim on a such a small shield would serve is unclear). This has exact parallels in the depiction of Greek footmen carrying aspides who are all too frequently shown with shields that are notably undersized. The coins after all do not depict soldiers in the strict military sense, ie. as cavalrymen, but rather as athletes engaging in mounted athletic activities. Other Tarantine coins for instance show the riders entirely devoid of weapons, and being crowned with victory garlands. One of the main purposes of these athletic events was the glorification of the body, and hence the interest in depicting nude riders rather than realistic shields. A compromise between realistic portrayal of the shield and the nude torso is shown in the coin at the bottom left, dating from 270 to 235 BC. The shield is depicted somewhat larger than those typical of the 'small shield' examples, though still decidely smaller than lifesize, and it is held back horizontally rather than vertically behind the rider, enabling the rider's body to be depicted while still showing a reasonably sized shield. This also allows us to see the deep curvature characteristic of the Argive aspis. It is perhaps noteworthy that all the coins featuring small shields are showing the riders in the act of dismounting ("apobates"); if the shields were enlarged and held at a realistic angle, this would be difficult to depict as the bent thighs would be hidden, as would the fact both legs are on the same side of the horse.

Combining as they did several javelins with a large heavy shield, what then was the tactical role of such Tarantine troops? The provision of javelins and apparent lack of armour clearly points to a skirmishing style of warfare, and this is reinforced by their brigading at Paraitakeni with the Medeian longchophoroi, described as being 'experts in wheeling and retreating'. However, the provision of a shield both heavy and large is not normally a feature of skirmishing cavalry, so it may be that such men were also expected to be able to engage in close combat. It may be that Tarantines were particulary adept at fighting dismounting, given the emphasis on mounting and dismounting in the Tarantine equestrian events, and an aspides would be particularly useful in this capacity; literary support for this proposition is lacking however.

The three surviving Hellenistic manuals differ slightly in their descriptions and classification of cavalry - including Tarantine cavalry.4 Asklepiodotos' manual states that there are in general terms three types of cavalry: spear-bearing cavalry fighting at close quarters only; bow-carrying cavalry that fight at long range only; and an intermediate category - skirmishers ("akrobolistai") that can be either bow or javelin armed. The intermediate category that throws javelins is further divided into two classes; one, the "elaphroi" (light), throws javelins and then closes in for close combat, while the other, the Tarantines, throw javelins without engaging in close combat.

Aelian divides the cavalry into two classes - katafractoi (fully armoured men on fully armoured horses) and those that are not. These latter are divided into the spear-bearing cavalry who fight hand-to-hand, and the akrobolistai, who fight with missiles. The akrobolistai are themselves divided into horse archers and those that use javelins, "hippakontistai", and these last are also called Tarantines. Of the Tarantines, some are 'properly' called Tarantines, and throw their weapons from a distance; the others, the elaphroi, throw one or two javelins before fighting hand-to-hand. Arrian's description is essentially the same as Aelian's but somewhat more detailed. In particular, he says that the 'proper' Tarantines do their shooting while either far away, or riding in circles; while the elaphroi, having shot at the enemy, either then engage them with spear or sword, or are kept in reserve.

Thus the three are very similar in their treatment of Tarantines - they are a type of light cavalry that throws javelins, and those 'properly' termed Tarantines solely do this; however, other cavalry also termed Tarantines fight hand-to-hand after having shot at the enemy. As it seems improbable that any single army could have would use both of these two different troop types bearing the same name at the same time, what we have here is evidence that either the role of such cavalry changed over the centuries, or that different Hellenistic armies used the same name for different troop types, or some combination of the two explanations. As the manuals incorporate material ranging from the time of Pyrrhus of Epirus (who wrote extensively on military matters), through the Achaian Greek general and historian Polybios (who also wrote a tactical manual, now lost) and finally to the philosopher Poseidonius who belongs to the start of the 1st century BC, and who is probably responsible for the arrangement and selection of material in the manuals as they have come down to us, all of these options are possible.5 Which then is the most likely?

To help answer this, we need to turn to literary accounts describing Tarantine cavalry, which alas are all too few. Diodoros does not tell us whether Antigonos and Demetrios' late 4th century Tarantines merely threw javelins or also engaged in hand-to-hand combat. His description of the mixed force including Tarantines that Antigonos dispatched to engage Eumenes' elephant convoy might imply they fought hand-to-hand, since they fell upon the 400 cavalry escorting the convoy "with all their weight" (19.39.5), but this is hardly the only conclusion possible, and clearly they later engaged the elephants only with missiles. Since the Tarantines were not the only ones present, even if some of the attacking force did charge into melee, it would not be conclusive proof the Tarantines did so.

According to Polybios' account of the 3rd battle of Mantineia (11.12.7) over a hundred years later, the Tarantines serving Philopoimon the Achaian were ordered to engage the Tarantines serving Machanidas the Spartan who in turn ordered his to counterattack ( here 'synapheinai'; the word seems to have no particular connotation of charging into contact). Polybios goes on to relate (11.13.1-2) that "at first the Tarantines alone were engaged, fighting gallently, but as the light-armed infantry (euzonoi) gradually came up to the support of those who were hard pressed, in quite a short time the mercenaries on both sides were mixed up. They were fighting all over the field, in a confused crowd and man to man". It is not clear if these mercenaries fighting at close quarters include the Tarantines or not; the Tarantines may well have been mercenaries, but 'mercenaries' in a Greek context in Polybios, without further qualification, usually implies men on foot. Moreover, Polybios divides Philopoimon's cavalry at the start of the battle into Achaians and mercenaries, with the latter being formed up in close order (11.11.7). Either the Tarantines are Achaians, and not mercenaries, or they are mercenaries who formed up in close order. Furthermore, it need not be the case that all the mercenaries were engaged in the melee even if the Tarantines are meant to included amongst those who qualify as mercenaries. The surviving evidence of Polybios is inconclusive with regards to tactical usage.

Tarantines are also mentioned by Livy (37.40) in the battle of Magnesia on the Seleucid left wing but alas they are not recorded as contributing to the battle, they seem to have been caught up in the rout of the Seleucid chariots, nor are their numbers recorded. Since the account of Livy is clearly derived from Polybios we can be reasonably sure that the word Tarantine appeared in Polybios' original account, and Polybios himself says Zeno reports them being present in the Seleucid army ten years earlier at Panion in 200 BC (Polybios 16.18). But in the most detailed description we have of the Seleucid army; that of Polybios of the parade at Daphne in ca. 164 BC, he does not call any of the cavalry he describes there "Tarantines". He does however mention 3000 civic ("politikoi") cavalry.6 As these did not wear helmets (they were wreathed or crowned for the festive occasion, some with gold, some with silver), it is tempting to conclude that charging into hand-to-hand combat was not their primary combat role. Whether they can be called skirmishing cavalry is another question: a force of 3000 purely skirmishing cavalry would seem to be a very large number, and it is very doubtful if such a large force of civic (militia?) horsemen could have attained and kept the skills necessary to become effective skirmishers on horseback.

It is notable that Poseidonius, the likely most recent source for the tactical manuals, came from Seleucid Apamea, just such a place where such civic cavalry would be recruited from. I would like to speculate that it was these civic cavalry that led to the definition of 'proper' Tarantines in the manuals. Sekunda has hypothesised that this is simply Poseidonius' muddled understanding of a troop type that no longer existed, but I fail to see how this could account for two different definitions - unless there were two such troop types in the past for him to muddle: this seems even less likely than one past and one contemporary troop type to be confused about.

It seems that the civic cavalry at Daphne were a recent innovation, at least as far as being called up in numbers as part of the field army (another innovation being the Roman-style foot described at the parade), since none of the cavalry troop types mentioned at Magnesia would seem to fit - except possibly the Tarantines already noted above, whose numbers were probably relatively small. While the Roman-style infantry appear to be a concession to developments in the west, the model for these cavalry is not so readily apparent.

There is a tendency for scholars to concentrate on the Seleucids' dealings with their western neighbours, not least because the evidence is so much more plentiful, and in particular, for the reign of Antiochos Epihanes, the organisor of the Daphne parade, his dealings with the increasingly militant Jewish population of 'hollow Syria'. However, for the Seleucids of the time, these were actually relatively minor concerns. The true threat came not from Rome, or the Jews, but from the east - in the form of the Parthians. The Parthians, from relatively obscure beginnings by the south-east corner of the Caspian sea, had risen to become a substantial threat since Antiochos the Great had campaigned in the area 40-odd years before.

The sources are scetchy,7 but they seemingly took advantage of the weakening of the Seleucid state by the huge war-indemnity the Romans imposed after Magnesia. By the time of the Daphne parade, they had expanded well beyond their ancestral homelands and were pressing hard against the hugely important satrapies of Media and Babylonia. Under a strong monarch such as Antiochos Epiphanes, they were seemingly held at bay, but once he died and the Seleucids were divided by civil war, the Parthians could not be contained. All of Media - source of the best part of the Seleucid cavalry arm, was lost by 147 BC, and Babylonia, the eastern core of the empire, was conquered in 141 BC. Indeed, to judge from the ratio of catafracts in the Daphne parade compared to Magnesia, the Parthians had already by then made considerable inroads into the ability of the Seleucids to tap their traditional Median recruiting grounds. Such losses emperilled the very existance of the state, and as soon as the kingdom was (temporarally) reunited, Antigonos Sidetes gathered together every force at his disposal in attempt to win back the lost territories (Justin 38.10.2 - an exaggerated total of 80000 soldiers and 300000 non-combatants!). It was the failure of this attempt - the entire force was lost - that ruined the Seleucid kingdom, reducing it to a mere rump of its former self.

Thus any innovations in military equipment during the first half of the 2nd century BC, particularly in the cavalry, are most likely to have in response to the threat posed by the Parthians. The Seleucid heavy cavalry had already adopted the catafract equipment that the Parthians used;8 as for the even more typical Parthian arm, light horse archers, two courses were possible - to imitate it or to counter it. The evidence points to both being attempted. In terms of imitation, in the latter period of the Seleucid kingdom's existence, horse archers made up an increasing percentage of the army. At Magnesia they appear to have made up 1200 of the 12000 Seleucid horse; in 147 BC the by-then 3000 strong Seleucid cavalry still apparently included some 1000 horse archers (1 Maccabees 10.79-89). Seleucid armies of the period are frequently referred to as including Bactrians, Dahae and Parthians, even when serving against the Parthians.

Horse archers however, perhaps more than any other troop type, have a tendency to be "born" in that horse archery skills are not acquired over-night. Thus creating a force of such troops from scratch is not something to be attempted in the middle of a crisis - such as when the Parthians are knocking at your gates. If Antiochos Epihanes or any other Seleucid innovator of the time wanted to counter Parthian horse archery tactics with cavalry of his own (and infantry were certainly not the ideal way of going about the task) an alternative would be required. Catafracts are expensive, and in any case, unable to catch light horse archers. Purely skirmishing javelin-armed troops would seem to be a waste of time against horse archers who could outshoot them both in terms of range and volume of missiles, and whose one slight disadvantage, armour peneration ability, would be irrelevant against them. What would be required was a troop that could attempt to chase them from the field if possible, and if not, at least attempt to shoot them as best possible. Missiles were thus required, and in the absence of horse archery skills, javelins, long a Greek mainstay, would have to do. This meant that a shield could still be employed for protection, but it needen't have been a heavy aspides - against arrows rather than spears, javelins or swords, more flexible but lighter materials provide equivalent or better protection while improving mobility. Such a troop type would look very similar to the traditional Tarantine however and I would suggest that this is why the old name was retained.

I would posit therefore that such civic milita cavalry were armed with javelins, were unhelmeted, and carried a light but large shield suitable for warding off enemy missiles. The shield would probably not be the long almond-shaped thureos typically carried by Galatian or Thracian cavalry in Seleucid service, since according to the manuals, cavalry carrying the thureos were spear-bearing cavalry ("doratophoroi"), not akrobolistai. A large round shield (but not a heavy Argive aspides), such as the flat spined shield used by many other Hellenistic cavlrymen, would fit the bill nicely, and several Seleucid terracottas of troops that apparently answer this description exist.

Thus 'proper' Tarantines, true skirmishers, albeit perhaps able to fulfill other duties, gave way to a later 'false' Tarantine class of cavalry, at least in Seleucid service.


1. See Duncan Head's discussion of Tarantines in his Armies of the Macedonian and Macedonian Punic Wars, WRG (1982). Return

2. Given the amount of loot this baggage raid secured, it would indeed have needed some troops who were 'very well disposed' to Antigonos to ensure that it was not simply purloined by its captors. Incidentally, the passage in Diodoros where the stray figure of 2200 Tarantines is quoted has been used to argue that these men could not have come from Tarantum, and thus the term Tarantine alreayy meant somebody fighting in the manner of Tarantines even at this early date. This conclusion loses its force when one realises only 200 of these men were actually Tarantines. Return

3. Eg. Arrian 4.4.6 and 4.4.7. Return

4. Asklepiodotos' manual is available as part of the Loeb series, with a translation dating back to World War One by Oldfather. Readers should ignore his frequent comments about the manual being divorecd from real world practice; this merely shows Oldfather's ignorance of the sources regarding the real world (excuseable given the scolarship of the period however). Aelian's manual is translated into English by A.M. Devine in Ancient World, 19 (1989), 31. The only English translation available of Arrian's manual is unfortunately a truly appalling effort by J.G.DeVoto (Ares, 1993), who for instance routinely translates longche as 'pike' leading to skirmishing cavalry throwing their pikes, while close-combat cavalry impale their opponents with 'javelins' (for kontos). However, as the Greek text taken from the Teubner edition is included, such details can be checked by the reader. Return

5. A short note on the Polybian influence on the Hellenistic manual tradition is expounded by A.M.Devine in "Polybius' Lost Tactica: The Ultimate Source for the Tactical Manuals of Asclepiodotus, Aelian, and Arrian?", The Ancient History Bulletin 9.1 (1995) 40-44; the influence of Poseidonius is discussed by N.Sekunda in Seleucid and Ptolemaic Reformed Armies 168-145 BC. Volume 1: The Seleucid Army, Montvert, 1994. The two theses are complementary rather than mutually exclusive. It seems to me that Poseidonius took Polybios' manual, rearranged it, added certain Seleucid-based information, such as that regarding chariots and elephants (and that this was responsible for many of the alternate formation names being introduced in other sections). Poseidonius however, being an armchair philosopher rather than a practicing general muddled things slightly on the way. Return

6. Polybios, 30.25.6. N.Sekunda deals with this parade at length in his above-mentioned book. Return

7. Justin's wretched epitome of the lost history by Pompeius Trogus is the starting point for any discussion of the declining years of the Seleucid empire, supplemented by Josephus' account of the ancient history of the Jews. Return

8. Exactly whom the Seleucids adopted catafract equipment from is debateable. By the time of Magnesia (190/89 BC), it seems that most of the Seleucid cavalry were catafracts, but at Panion in 200 BC, to judge from Polybios' critique of Zeno's account (Polybios 16.18), only the left wing was (contrary to Sekunda who claims Polybios doesn't mention catafracts at the batttle at all), and catafracts are not mentioned by Polybios in the fighting in Bactria in 206 BC. Thus is appears reasonably certain that they were introduced as a result of Antiochos' eastern campaign. However, he there fought three peoples that could have used such equipment. Firstly the Parthians - noted users of catafract gear, but as Duncan Head has pointed out, there is no direct evidence that they themselves were actually using complete catafract gear at this time, since all the available evidence is considerably later. Secondly, the Graeco-Bactrians also used catafract equipment - but again the evidence for them again only comes after this date; see V.P.Nikonorov, "The Armies of Bactria 700 BC - 450 AD", Montvert, 1997. Thirdly, the Skythians, who are the only people currently known who can be demonstrated to have used exceptionally well-armoured cavalry before the Seleucids did so. That they were already using heavily armoured horses in Alexander the Great's day is plainly shown by Arrian's account of Gaugamela, and while it may be legitimately doubted that these were true catafacts, the Skythian fragment dated to approximately 300 BC illustrated (no. 29) in Sekunda's book showing a horse which appears to be completely armoured, and carrying a fully armoured man wielding a heavy lance fitted with a butt-spike would appear, at least in form, to portray nothing less than a 'standard' catafract. Return


I would like to especially thank Duncan Head for his comments on an earlier draft of this article.


Return to my homepage index