Collections: Ancient ‘Tanks’? Chariots, Scythed Chariots and Carroballistae

Building on last week’s post on tanks and a few of the comments there, this week I wanted to talk about the ancient (and medieval) weapon-systems often analogized to tanks and the degree to which they had a role similar to tanks.

I have lost count of how many times I have seen in this or that documentary made for the public an interviewee explaining to the camera how “______ was the tank of the ancient world!” or some similar explanation, for a variety of different kinds of ancient combat vehicles: chariots, scythed chariots or the Roman carroballista, for instance. Now obviously none of these ancient weapon-systems are going to fill our definition of a tank from last time, but we can talk about how they were used and the degree to which that use does or does not overlap with tanks.

In particular what folks tend to mean when they say this or that thing was the ‘tank’ of the ancient or medieval world, they mean it was a heavy and heavily armored platform capable of tremendous offensive power, able to ‘smash’ infantry out of defensive positions or engage in maneuver-and-fire assaults. Mostly, as we’ll see, this is not how these ancient war vehicles were used.

As always, if you like what you are reading here, please share it; if you really like it, you can support me on Patreon. And if you want updates whenever a new post appears, you can click below for email updates or follow me on twitter (@BretDevereaux) for updates as to new posts as well as my occasional ancient history, foreign policy or military history musings.

Chariots

We’ll start chronologically with the oldest of these, the two-wheeled war chariot, which was first developed on the Eurasian Steppe and percolated into Bronze Aged Mesopotamia, Anatolia and Egypt beginning in the 18th century BCE (though it takes a while to finally get all the way through the region) and China in the 13th century BCE.

And it isn’t hard to see the ‘tank’ analogy: in their hey-day before the introduction of cavalry, chariots were the most expensive, cutting edge military technology, requiring lots of specialized artisans to produce and maintain and specialized soldiers to employ (typically a trained but commoner chariot driver and an aristocratic warrior to whom the chariot actually belonged).

And in film and the broader popular culture, chariots tend to be depicted with some ‘tank’-like attributes: heavy shock weapons that can ‘smash’ infantry out of position. If my warfare survey, I use a clip from Exodus: Gods and Kings (2014) as an example of this popular vision of the chariot: the chariotry first devastate the enemy with a massive arrow volley and then crash through their infantry-line effortlessly using the raw weight of the chariots. Very tank-like! This is also exactly how ‘chariot’ units function in most Total War games: they have massive ‘charge bonuses’ and high ‘mass’ but low base attack and defensive stats, making them devastating in impact but vulnerable if they get stuck or stick around.

The only problem is that this vision of chariot warfare has almost nothing to do with actual ancient chariots.

Let’s start with weight. In a little ‘making of’ featurette, the folks behind Exodus: Gods and Kings (2014) go on about how proud they are of their chariot battle and how hard it was to make. Of note they comment on how it was hard to build the chariots and keep them light, with most of their chariot-cabs weighing around 250-350kg. It makes one wonder how the Egyptians could have used such a heavy weapon-system, but of course the answer is they didn’t. We have a handful of preserved Egyptian war chariots and they tend to mass around 35kg, about a tenth of what the movie props massed. Chariot wheels were spoked to save weight (over time the number of spokes increases so as to reduce the pressure on any individual spoke). Far from being armored, the chariot cab was built for speed, which with two horses means building light. But you certainly don’t want to take a high-speed impact in such an expensive but lightly built cab!

Well, surely you could at least drive them up and hit someone with your weapons? Well, no. Not well, anyway. Again, the problem is that chariots are built for maneuvering at speed, which means they need a wheel-base wide enough to not tip over when they turn. Measurements here vary, but they tend to be pretty wide; the Lchashen chariot has a wheel-base around 7ft. Chinese chariots tended to be even a bit wider, with the axle length generally running around 3m. In short with most Bronze Age cutting weapons (axes, swords), it would be difficult if not impossible to actually reach out to offend a target outside of the wheel-track; a spear could reach, but the angle is all wrong since your own horses are in the way of a thrust forward.

(For more on chariots, Lee, Waging War (2016) has a solid introductory chapter for the lay-reader. See also E.L. Shaughnessy, “Historical Perspectives on the Introduction of the Chariot into China” JHAS 48.1 (1988) and P.R.S. Moorey, “The Emergence of the Light, Horse-Drawn Chariot in the Near-East c. 2000-10500 B.C.” World Archaeology 18.2 (1986))

Instead the primary weapon of the chariot-riding aristocrat were missile weapons: the composite bow or in some cases javelins. This was, in essence, a skirmish platform. Now it is worth noting that chariots had limited ability to absorb the shocks of moving across ground (some later chariots use reed-work floors to cushion this somewhat) so accuracy on the move is going to be limited – you have no ability to predict the vertical motion of the chariot cab the way a horse archer can time his shots with the footsteps of his horse. But you can harass enemy infantry while on the move or rapidly reposition to fire accurately while stationary.

That said, I don’t think the comparison of even the formidable composite bow with a tank’s main armament is particularly useful. As discussed previously, part of the tank’s advantage is bringing a lot of direct fire to the front where it can support infantry, but for all of its expense, for all of the space it takes up, a chariot has exactly the fire potential as a single foot archer (but much more potential to aggrandize elites, which may explain much of the chariot’s popularity).

Finally, chariots are scary: they are big and loud and fast and could be charged for the morale impact. If the enemy ran away you could ride them down firing arrows and throwing javelins; if they didn’t run away, you could swerve aside and ride off! But you absolutely did not want to smash-impact them because remember that your unarmored, vulnerable and panicky horses are in the front of your vehicle.

So a light, fast, barely-armored weapon-system meant mostly for aristocratic display and skirmishing with some morale-shock potential – the chariot is less the ‘tank’ of the ancient world and honestly more like a technical, an improvised military vehicle produced by attaching a heavy weapon, typically a large machine-gun but sometimes even heavier weapons, to the bed of a civilian truck. This comparison seems particularly in the sorts of conflicts where – due to insufficient purpose-built military vehicles – technicals acquired important prestige signalling function among local warlords in addition to tactical importance as ‘battlewagons.’ But I suppose ‘the technical of the ancient world’ doesn’t quite have the same ring to it.

Alternately one might compare the role of chariots to the way that some expensive prestige combat systems, particularly fighter jets, are used by some autocrats as a tool of aristocratic signalling as much as an actual component of combat readiness. The chariot might have been categorically outperformed by the later development of true cavalry, but as a platform for aristocratic display and legitimacy building, it was practically unbeatable: a literally elevated platform for kings and their elite retainers to perform military excellence for their armies in highly visible ways. It is striking that, even once chariots lose their battlefield role, they are retained in some places as command platforms for kings and officers and even among the Romans, who as far as we can tell never made much battlefield use of chariots, the chariot was still the prestige vehicle of the triumphing general in his grand victory parade back in Rome (in part because even after their battlefield use dwindled to zero, chariots remained in both Roman and Greek thought as the vehicles of the gods).

Scythed Chariots

Scythed chariots come on the scene rather later and the distinction is important to make. The war chariot as a battlefield weapon practically vanishes in the Near East beginning in the early Iron Age (there’s a transitional period, but by the end of the 7th century BC, it is functionally complete), because cavalry (in particular, horse archers) could do everything chariots could do, but better and with half the horses and half the humans. Chariots stuck around as prestige vehicle for kings, but no longer saw extensive use on the battlefield.

Scythed chariots were developed later, although there isn’t much scholarly consensus on exactly when; possibly under the Achaemenids, possibly in the Neo-Assyrian period (which would be contemporary with the decline of traditional war chariots). Scythed chariots essentially take one element of the older war chariot – the terror of their charge – and dial it up to eleven, putting large blades both on the wheels and also beneath the chariot’s cab; unlike the older war chariots these seem frequently to have been four-horse chariots. Unlike war chariots, these were always specialty limited-use systems: you only brought a handful, compared to much larger numbers of cavalry. As Jeffery Rop notes (“Reconsidering the Origin of the Scythed Chariot” Historia 62:2 (2013)), scythed chariots required ideal terrain – open and flat – to function and as a result seem to have been, at best, infrequently used; we have just a handful of instances of the Achaemenids using them total over at least a century.

Generally, scythed chariots attacked in front of the advance of the main body of the army and the idea here seems to have been that they would break up enemy formations and disorder the enemy thus allowing the main attack to succeed (e.g. Xen. Anab. 1.8.10). Alexander Nefiodkin has proposed that this was explicitly a response to Greek heavy infantry, but I tend to think he has this exactly backwards: this trick almost never works against dense, cohesive heavy infantry – where it does work (e.g. at Cunaxa, Anab. 1.8 and also in 396 against a foraging party, Xen. Hel. 4.1) it is against either lighter troops or troops scattered in the open (and sometimes not even then, consider the failure of Seleucid scythed chariots against skirmishers at Magnesia in 191).

Indeed it is difficult to come away from the sources with a very high opinion of scythed chariots: they tend to be notable because of their notable failures to deal with heavy infantry. At Cunaxa, Xenophon contends that the onset of the chariots caused no losses among the hoplite formation (Xen. Anab. 1.8.20), with the Greeks simply opening ranks to allow the chariots past; Alexander uses the same tactic as Gaugamela (Arr. Anab. 3.13.5-6), but there also integrated light infantry with javelins, we are told, also disabled many of the chariots before they reached the lines. That use of light infantry also occurred at Magnesia where Pergamon light infantry allied with the Romans similarly routed Seleucid scythed chariots (Liv. 37.41.10-12).

As I noted when discussing war elephants, the Romans seem to have classified scythed chariots with war elephants and other Hellenistic ‘gimmick’ weapons unsuited to serious warfare and it is hard to disagree. Livy’s quip that the scythed chariots at Magnesia were an inanis ludibrium, “a useless mockery” of war compared to the ‘real thing’ that began moments later (Liv. 37.41.12) serves as a neat summation of the Roman opinion (especially when paired to his dismissal of war elephants as elephantomachae nomen tantum sine usu fuerunt, “the war elephants were just a name without any use,” Liv. 44.41.4). In practice, scythed chariots seem to have underperformed precisely when it counted: when the enemy was well-organized and highly motivated.

In the sense of a heavy, armored vehicle designed to shock infantry out of position, it seems like, in concept the scythed chariot is something like an ancient tank. But unlike tanks, which ended up as a staple weapon-system is nearly every military in the world, scythed chariots strike me as being more akin to various wunderwaffen – ambitious if often over-complicated ‘wonder weapons’ whose costs or limitations were not equaled by their utility.

The Carroballista

The Roman carroballista, which first appears in the second century AD, is also sometimes described as being like an ancient ‘tank.’ Saying anything about the carroballista is complicated by the scarceness in the sources; it is mentioned in a grand total of two texts, both of them late: Vegetius’ Epitoma Rei Militaris 2.25 and the anonymous De Rebis Bellicis.¹ Both sources are troublesome; Vegetius is notionally writing about the ‘old’ Roman legions from the late fourth century, but Vegetius is at best careless, drawing elements from a variety of periods and mixing them together. The De Rebis Bellicis is also quite late (possibly early fifth century) and anonymous. Were it not for the fact that carroballistae are also clearly depicted on the Columns of Trajan and Marcus Aurelius, scholars might well doubt this was ever a real military device.

Nevertheless they are so depicted, multiple times and so it seems that the carroballista was a very real thing and used fairly regularly; Vegetius, for what it is worth, claims each legion would have had 55 of them. The figure is not, on its face, implausible.

From what we can see, the structure of the carroballista was generally a two-wheeled cart pulled by a pair of mules or horses with a two-armed torsion arrow-throwing catapult (the ballista) mounted on the cart.² This was a crew-operated weapon as Vegetius notes, with each carroballista having a contubernium – a unit of eight soldiers – assigned to it, though on the Columns of Trajan and Marcus the weapon itself seems to be worked by a two-man team. And unlike the chariot, which could only develop the same fire potential as a foot archer, the carroballista could hit a lot harder; Vegetius again notes that neither the heavy armor of cavalry nor the shields of infantry could stand up to it, which is perfectly believable for a Roman torsion catapult. These things hit hard.

So we have a mobile, powerful direct fire platform. Is it a tank? Well, no.

First off, from what we see on the Columns of Trajan and Marcus, the carroballista is entirely unarmored; not only the cart but also the animals lack any kind of armor. Moreover, the animals on the columns are mostly mules, not horses. The choice of mules suggests an emphasis on endurance and pulling-strength over raw speed; we have no idea how heavy a carroballista was, but the choice of mules would seem to suggest fairly heavy indeed. Moreover, a close look at the carroballistae on the columns reveals that at least one of the crew members is always shown on foot next to the cart, either assisting in firing or helping to push it; only one soldier is ever shown riding in the cart. If, as Vegetius suggests, the whole system had a contubernium assigned to it, there’s no way they could all have moved on the cart or its draft-animals; there isn’t enough space. So the mobility of the carroballista is necessarily limited to the foot-speed of its operators.

Which brings us to operation. Loading a ballista required the working of a winch (with a ratchet to prevent premature release) at the rear of the weapon to pull the bow-string back; this stored energy because the string was attached to a pair of levers which were in turn inserted through a pair of torsion-springs (typically made out of sinew or hair) kept under tension. Pulling the string back (further) twisted the torsion-springs so that once the string was released they would snap back into their previous position launching the projectile (with some engines designed for throwing arrows and others for throwing stones).³ Needless to say that requires a fair amount of muscle-power to be put to the winch back, which in at least one of the panels of Trajan’s Column seems to be what the soldier on foot is doing (one wonders if in practice the weapons were designed to allow two soldiers, one to teach side, to winch together; other Roman torsion catapults were designed like this); the soldier apparently in the cart aims the weapon. The important implication here is that this is a system which cannot fire while moving (or more correctly, cannot reload while moving). That impression is reinforced by how Vegetius describes their use, with the carroballistae having two roles: defending camps and firing from behind the heavy infantry line.⁴

So what we want to imagine here is not a hyper-mobile ‘tank’ dashing about the battlefield firing on the move, but rather a mobile field artillery piece, able to relocate (at something like a jogging pace) during a battle to set up at one protected firing position to the next and offering fire support over the infantry line which protects it. In this sense, the carroballista is less like a tank and more like 17th-20th century horse artillery. While nearly all artillery of that period was drawn by horses, horse artillery took this a step further by employing a much larger team of horses for each gun, allowing the entire crew to be mounted. Of course they couldn’t fire on the move either; the idea was that they could rapidly reach a new firing position to unlimber⁵ and fire, similar to how modern self-propelled artillery can shoot-and-scoot.

Conclusion

It is perhaps not surprising that none of these ancient systems is actually a very good analogue for tanks: tanks were, after all, created to solve a battlefield problem – overwhelming firepower that could effectively prohibit infantry advance over open ground – that didn’t exist before the modern period. Contrary to what one might sometimes see in movies and video-games, even the best archers could not produce enough downrange fire to actually stop a determined infantry charge (the French at Agincourt reached the English after all, even if they didn’t then win), but of course machine guns could.

Now at the beginning I said I would stick to ancient vehicles, but of course there is one other kind of pre-modern combat arm which gets compared to tanks: cavalry. And there is something to that: many country’s armored formations are the direct heirs of earlier cavalry formations and at various points tank or other armored fighting vehicles have done some of the jobs, especially scouting, that would have in earlier eras been assigned to cavalry (at some points light tanks were used for reconnaissance and these days that is often one of the roles of specially outfitted IFVs like the M3 Bradley Cavalry Fighting Vehicle). We have, of course, discussed cavalry dynamics here at some length, though focused purely on the pre-gunpowder era.

Discussing the way that the lineage of various cavalry forces shaped the emergence of armored forces is beyond the scope of this post, but for the curious I should note that Nicholas Moran has a series of videos detailing the interwar development of tank doctrine in various militaries which often gets into the debates that were had about the future battlefield use of the horse and the place of tanks and other armored fighting vehicles in a truly modern, mechanized cavalry.

That said, I think that comparisons describing this or that ancient or medieval fighting system as the ‘_________ of the ancient/medieval world’ tend to obscure more than the elucidate and often serve as cover for a failure to really interrogate how those pre-modern combat arms actually worked in practice. It is, unfortunately, the rare bit of public-facing history that efforts to do this effectively without recourse to inaccurate modern analogies or absurd hyperbole.

In any case, pre-modern war vehicles existed, but they were not tanks or even very much like the public’s conception of a tank.

Next time: something that doesn’t have much or anything to do with tanks!

1. I do not have a copy of the latter at hand to check the exact citation. My notes on the carroballista were, alas, written by a younger version of me who was less careful!
2. It must be noted there is some necessary speculation in all of this because, again, the sources are very limited.
3. Ancient torsion engine terminology here is tricky. In modern parlance, we tend to call arrow-throwing devices ballista and rock-throwing devices catapults, but in ancient usage these terms doesn’t precisely follow. In the Late Republic and the Early Empire, it is the larger stone-throwing engines which get term ballista, while smaller arrow-throwing engines are a catapulta or more commonly a scorpio. For reasons that aren’t entirely clear (but may be related to some technical differences in the design of the engines and the shape of the torsion-arms), by the late Empire, the terms largely reverse positions, with catapultae being larger rock-throwing engines (note also the onager by this point), while ballistae are the smaller, arrow-throwers. For more on this technical question and on ancient artillery in general, the standard reference work remains E.W. Marsden, Greek and Roman Artillery, 2 vols (1971), though it can be quite difficult to get a hold of.
4. non solum autem castra defendunt, verum etiam in campo post aciem gravis armaturae ponuntur, “they not only defend the camp, but indeed in the field are placed behind the line of heavy infantry.”
5. Gloss: Set up the guns.