Collections: War Elephants, Part I: Battle Pachyderms

Today we’re starting the first part of our three part series on War Elephants (by reader request!). In this first part, we’re going to talk about how elephants performed in battle: how did they fight and what was their battlefield purpose? The second two parts will focus on the (as we’ll see) far more important question of how elephants performed in war: part two will focus on the drawbacks which limited (and eventually ended) the use of war elephants in the ancient Mediterranean, while part three will look at the continued use of war elephants in India.

But first – Dear my students (and also my colleague’s students): this is a blog post. Yes, yes, I think I am a cool, knowledgeable fellow too. And I know you are thinking about writing about war elephants for that paper (because someone always is). But this is a blog post. Do not use a blog post – even one so handsome as this one – as a source for your college (or high school) paper. I will make references to books and articles: go to the library, read those, and cite them (also, for that paper, war elephants are actually not a great topic (for reasons we will touch on in a moment) – consider something else). Thank you.

In particular, I’ll be leaning quite heavily on T. Trautmann’s Elephants and Kings (2015), one of the better works on the topic I’ve seen – particularly for the war elephant in India, where I am less well versed myself.

Still my favorite Etruscan elephant plate, from the National Etruscan Museum at the Villa Giulia in Rome. Note (as discussed below) that the tower here is a Hellenistic device and doesn’t make it to India until much later.
If you look closely, you can see the mahout (the driver) holding up a pointed hook; the is the ankusa, the iron hook used to control the elephant in battle. Several of these have been recovered archaeologically, see Trautmann (2015).

The Trunk of the Problem

The interest in war elephants, at least in the ancient Mediterranean, is caught in a bit of a conundrum. On the one hand, war elephants are undeniably cool, and so feature heavily in pop-culture (especially video games). In Total War games, elephants are shatteringly powerful units that demand specialized responses. In Paradox’s recent Imperator, elephant units are extremely powerful army components. Film gets in on the act too: Alexander (2004) presents Alexander’s final battle at Hydaspes (326) as a debacle, nearing defeat, at the hands of Porus’ elephants (the historical battle was a far more clear-cut victory, according to the sources). So elephants are awesome.

I mean – look at that stat-line. That charge bonus. Awesome.
Though it is a touch odd that the Romans have apparently managed to chip exactly one tusk per elephant. No more and no less. That’s Roman precision for you.

On the other hand, the Romans spend about 200 years (from c. 264 to 46 B.C.) mopping the floor with armies supported by war elephants – Carthaginian, Seleucid, even Roman ones during the civil wars (Thapsus, 46 B.C.). And before someone asks about Hannibal, remember that while the army Hannibal won with in Italy had almost no war elephants (nearly all of them having been lost in the Alps), the army he lost with at Zama had 80 of them. Romans looking back from the later Imperial period seemed to classify war elephants with scythed chariots and other failed Hellenistic ‘gimmick’ weapons (e.g. Q. Curtius Rufus 9.2.19). Arrian (a Roman general writing in the second century A.D.) dismisses the entire branch as obsolete (Arr. Tact. 19.6) and leaves it out of his tactical manual entirely on those grounds.

This negative opinion in turn seeps into the scholarship on the matter. This is in no small part because the study of Indian history (where war elephants remained common) is so under-served in western academia compared to the study of the Greek and Roman world (where the Romans functionally ended the use of war elephants on the conclusion that they were useless). Trautmann, (2015) notes the almost pathetic under-engagement of classical scholars with this fighting system. Scullard’s The elephant in the Greek and Roman World (1974) remains the standard text in English on the topic some 45 years later, despite fairly huge changes in the study of the Achaemenids, Seleucids, and Carthaginians in that period.

All of which actually makes finding good information on war elephants quite difficult – the cheap sensational stuff often fills in the gaps left by a lack of scholarship. The handful of books on the topic vary significantly in terms of seriousness and reliability.

So we’re going to take this discussion in three parts: Part I (this one) is going to look at how the elephant functioned in battle: how did it work as a weapon-system and why would anyone want to have it? Part II (next week) will then turn and ask the question: if elephants are such awesome weapon systems, why did the Romans defeat and then abandon them (and why did the Chinese never meaningfully adopt them)? Part III (the week after that) turns this question on its head: if elephants were as useless as the Romans thought, why did Indian kings keep using them?

The Elephant as Weapon

The pop-culture image of elephants in battle is an awe-inspiring one: massive animals smashing forward through infantry, while men on elephant-back rain missiles down on the hapless enemy. And for once I can surprise you by saying: this isn’t an entirely inaccurate picture. But, as always, we’re also going to introduce some complications into this picture.

Elephants are – all on their own – dangerous animals. Elephants account for several hundred fatalities per year in India even today and even captured elephants are never quite as domesticated as, say, dogs or horses. Whereas a horse is mostly a conveyance in battle (although medieval European knights greatly valued the combativeness of certain breeds of destrier warhorses), a war elephant is a combatant in his own right. When enraged, elephants will gore with tusks and crush with feet, along with using their trunks as weapons to smash, throw or even rip opponents apart (by pinning with the feet). Against other elephants, they will generally lock tusks and attempt to topple their opponent over, with the winner of the contest fatally goring the loser in the exposed belly (Polybius actually describes this behavior, Plb. 5.84.3-4). Dumbo, it turns out, can do some serious damage if prompted.

From Alexander (2004). Pictured: A group of confused Macedonians, about to witness an exciting, possibly fatal, wildlife exhibit.
On the left, you can see a man preparing to jump off the side of the elephant. I am not aware of any evidence that this was ever a fighting technique. It certainly seems suicidally stupid.

Elephants were selected for combativeness, which typically meant that the ideal war elephant was an adult male, around 40 years of age (we’ll come back to that). Male elephants enter a state called ‘musth’ once a year, where they show heightened aggressiveness and increases interest in mating. Trautmann (2015) notes a combination of diet, straight up intoxication and training used by war elephant handlers to induce musth in war elephants about to go into battle, because that aggression was prized (given that the signs of musth are observable from the outside, it seems likely to me that these methods worked).

(Note: In the ancient Mediterranean, female elephants seem to have also been used, but it is unclear how often. Cassius Dio (Dio 10.6.48) seems to think some of Pyrrhus’s elephants were female, and my elephant plate (seen above) shows a mother elephant with her cub, apparently on campaign. It is possible that the difficulty of getting large numbers of elephants outside of India caused the use of female elephants in battle; it’s also possible that our sources and artists – far less familiar with the animals than Indian sources – are themselves confused.)

Thus, whereas I have stressed before that horses are not battering rams, in some sense a good war elephant is. Indeed, sometimes in a very literal sense – as Trautmann notes, ‘tearing down fortifications’ was one of the key functions of Indian war elephants, spelled out in contemporary (to the war elephants) military literature there. A mature Asian elephant male is around 2.75m tall, masses around 4 tons and is much more sturdily built than any horse. Against poorly prepared infantry, a charge of war elephants could simply shock them out of position a lot of the time – though we will deal with some of the psychological aspects there in a moment.

A word on size: film and video-game portrayals often oversize their elephants – sometimes, like the Mumakil of Lord of the Rings, this is clearly a fantasy creature, but often that distinction isn’t made. As notes, male Asian (Indian) elephants are around 2.75m (9ft) tall; modern African bush elephants are larger (c. 10-13ft) but were not used for war. The African elephant which was trained for war was probably either an extinct North African species or the African forest elephant (c. 8ft tall normally) – in either case, ancient sources are clear that African war elephants were smaller than Asian ones.

Total War: Rome II’s war elephants appear almost precisely twice as tall as a standard infantryman (look at the third man from the front, whose helmet doesn’t have a high crest), which isn’t ludicrous, but it is still a bit too large. Still more accurate than most – also, credit for a fairly realistically conservatively sized tower.

Thus realistic war elephants should be about 1.5 times the size of an infantryman at the shoulders (assuming an average male height in the premodern world of around 5’6″), but are often shown to be around twice as tall if not even larger. I think this leads into a somewhat unrealistic assumption of how the creatures might function in battle, for people not familiar with how large actual elephants really are.

Henry Motte’s depiiction (1878) of Hannibal’s war elephants crossing a river is the headline image for Wikipedia’s war elephant’s article. These elephants – three times the size or more of the men on the rafts – are much too large. They also appear to be African bush elephants, which were not used in war (the big ear flaps, combined with the size is the give-away).
Also, if you plan to live through your river crossing, my advice: don’t be in the tower on the war elephant during the crossing.

The elephant as firing platform is also a staple of the pop-culture depiction – often more strongly emphasized because it is easier to film. This is true to their use, but seems to have always been a secondary role from a tactical standpoint – the elephant itself was always more dangerous than anything someone riding it could carry.

There is a social status issue at play here which we’ll come back to in the third post. The driver of the elephant, called a mahout, seems to have typically been a lower-status individual and is left out of a lot of heroic descriptions of elephant-riding (but not driving) aristocrats (much like Egyptian pharaohs tend to erase their chariot drivers when they recount their great victories). Of course, the mahout is the fellow who actually knows how to control the elephant, and was a highly skilled specialist. The elephant is controlled via iron hooks called ankusa. These are no joke – often with a sharp hook and a spear-like point – because elephants selected for combativeness are, unsurprisingly, hard to control. That said, they were not permanent ear-piercings or anything of the sort – the sort of setup in Lord of the Rings (pictured below) is rather unlike the hooks used (you can see one of the actual hooks shown on the Etruscan plate at the top of this post).

The elephant hooks shown in Lord of the Rings: Return of the King are quite wrong. Here they are metal hooks permanently attached to the animal’s ears and pulled on by ropes – in this case, the mahout has been speared by Eomer and falls off the side of the elephant, dragging it into a collision with another…which perhaps neatly illustrates why actual war elephants were not controlled this way.

In terms of the riders, we reach a critical distinction. In western media, war elephants almost always appear with great towers on their backs – often very elaborate towers, like those in Lord of the Rings or the film Alexander (2004). Alexander, at least, has it wrong. The howdah – the rigid seat or tower on an elephant’s back – was not an Indian innovation and doesn’t appear in India until the twelfth century (Trautmann supposes, based on the etymology of howdah (originally an Arabic word) that this may have been carried back into India by Islamic armies). Instead, the tower was a Hellenistic idea (called a thorkion in Greek) which post-dates Alexander (but probably not by much).

From Alexander (2004), some very stylish elephants. As Scenes of war elephants go, this is actually one of the better ones – the elephants are of realistic size and attack with supporting arms. Those towers on the elephant’s back, however, are several centuries (possibly fifteen centuries) too early.

This is relevant because while the bowmen riding atop elephants in the armies of Alexander’s successors seem to be lower-status military professionals, in India this is where the military aristocrat fights. We’ll talk more about this in the third post – indeed, it will be the focus on that post – but this is a big distinction, so keep it in mind. It also illustrates neatly how the elephant itself was the primary weapon – the society that used these animals the most never really got around to creating a protected firing position on their back because that just wasn’t very important.

In all cases, elephants needed to be supported by infantry (something Alexander (2004) gets right!) Cavalry typically cannot effectively support elephants for reasons we’ll get to in a moment. The standard deployment position for war elephants was directly in front of an infantry force (heavy or light) – when heavy infantry was used, the gap between the two was generally larger, so that the elephants didn’t foul the infantry’s formation.

From Alexander (2004), Elephants advancing with light infantry support. I’m not sure I would advance my light infantry in front of my elephants. Arrian – the best source on the battle – seems to set Porus’ infantry behind the elephants.

Infantry support covers for some of the main weaknesses elephants face (which we’ll get into more next week), keeping the elephants from being isolated and taken down one by one. It also places an effective exploitation force which can take advantage of the havoc the elephants wreck on opposing forces. The ‘elephants advancing alone and unsupported’ formation from Peter Jackson’s Return of the King, by contrast, allows the elephants to be isolated and annihilated (as they subsequently are in the film).

These elephants sure could use some light infantry support. Or any infantry support. But on the upside, that mahout on the left sure looks happy, with that big ol’ smile on his face.

Tusk or Terror?

But – and you knew this was coming – the main impact of war elephants is psychological. The Battle of the Hydaspes (326 BC), which I’ve discussed here is instructive . Porus’ army deployed elephants against Alexander’s infantry – what is useful to note here is that Alexander’s high quality infantry has minimal experience fighting elephants and no special tactics for them. Alexander’s troops remained in close formation (in the Macedonian sarissa phalanx, with supporting light troops) and advanced into the elephant charge (Arr. Anab. 5.17.3) – this is, as we’ll see next time, hardly the right way to fight elephants. And yet – the Macedonian phalanx holds together and triumphs, eventually driving the elephants back into Porus’ infantry (Arr. Anab. 5.17.6-7).

So it is possible – even without special anti-elephant weapons or tactics – for very high quality infantry (and we should be clear about this: Alexander’s phalanx was as battle hardened as troops come) to resist the charge of elephants. Nevertheless, the terror element of the onrush of elephants must be stressed: if being charged by a horse is scary, being charged by a 9ft tall, 4-ton war beast must be truly terrifying.

Yet – in the Mediterranean at least – stories of elephants smashing infantry lines through the pure terror of their onset are actually rare. This point is often obscured by modern treatments of some of the key Romans vs. Elephants battles (Heraclea, Bagradas, etc), which often describe elephants crashing through Roman lines when, in fact, the ancient sources offer a somewhat more careful picture. It also tends to get lost on video-games where the key use of elephants is to rout enemy units through some ‘terror’ ability (as in Rome II: Total War) or to actually massacre the entire force (as in Age of Empires).

At Bagradas (255 B.C. – a rare Carthaginian victory on land in the First Punic War), for instance, Polybius (Plb. 1.34) is clear that the onset of the elephants does not break the Roman lines – if for no other reason than the Romans were ordered quite deep (read: the usual triple Roman infantry line). Instead, the elephants disorder the Roman line. In the spaces between the elephants, the Romans slipped through, but encountered a Carthaginian phalanx still in good order advancing a safe distance behind the elephants and were cut down by the infantry, while those caught in front of the elephants were encircled and routed by the Carthaginian cavalry. What the elephants accomplished was throwing out the Roman fighting formation, leaving the Roman infantry confused and vulnerable to the other arms of the Carthaginian army.

So the value of elephants is less in the shock of their charge as in the disorder that they promote among infantry. As we’ve discussed elsewhere, heavy infantry rely on dense formations to be effective. Elephants, as a weapon-system, break up that formation, forcing infantry to scatter out of the way or separating supporting units, thus rendering the infantry vulnerable. The charge of elephants doesn’t wipe out the infantry, but it renders them vulnerable to other forces – supporting infantry, cavalry – which do.

Elephants could also be used as area denial weapons. One reading of the (admittedly somewhat poor) evidence suggests that this is how Pyrrhus of Epirus used his elephants – to great effect – against the Romans. It is sometimes argued that Pyrrhus essentially created an ‘articulated phalanx’ using lighter infantry and elephants to cover gaps – effectively joints – in his main heavy pike phalanx line. This allowed his phalanx – normally a relatively inflexible formation – to pivot.

This area denial effect was far stronger with cavalry because of how elephants interact with horses. Horses in general – especially horses unfamiliar with elephants – are terrified of the creatures and will generally refuse to go near them. Thus at Ipsus (301 B.C.; Plut. Demetrius 29.3), Demetrius’ Macedonian cavalry is cut off from the battle by Seleucus’ elephants, essentially walled off by the refusal of the horses to advance. This effect can resolved for horses familiarized with elephants prior to battle (something Caesar did prior to the Battle of Thapsus, 46 B.C.), but the concern seems never to totally go away. I don’t think I fully endorse Peter Connolly’s judgment in Greece and Rome At War (1981) that Hellenistic armies (read: post-Alexander armies) used elephants “almost exclusively” for this purpose (elephants often seem positioned against infantry in Hellenistic battle orders), but spoiling enemy cavalry attacks this way was a core use of elephants, if not the primary one.


Elephants in popular culture are often shown all on their own, smashing through formations of enemies. The smashing is accurate, but the all-alone is not: elephants existed within tactical systems which combined them, particularly with infantry. Trautmann (2015) notes that in India itself, the ‘fourfold’ army, with cavalry, chariots, foot and elephants, became a poetic paradigm for the correctly structured army, even after chariots had largely fallen out of use.

Nevertheless, elephants were potentially extremely powerful weapon-systems. Their key value lay not in the ability to shatter enemy formations off the field (although they could do that) or inflict massive casualties (but they can do that too), but in the disorder they caused an enemy. Against infantry, elephants were extremely disruptive to the tightly-packed formations required for success. Against horses, the natural aversion of horses to approach elephants could spoil cavalry charges and scatter horseman.

Thus the war elephant wasn’t a ‘battle winner’ so much as a dangerous complication thrown in the way of the enemy’s plan of attack. And at that, they were awesome.

Next week, we’ll look at the drawbacks. Why didn’t the Romans – or the Chinese for that matter – make much use of these awesome, awesome animals?

16 thoughts on “Collections: War Elephants, Part I: Battle Pachyderms

  1. Nice!! I guess you could explain, as a part of the drawbacks of using war elephants, the training they required, like percussion and rithms with drums, maybe during the night, or so I heard in a documentary…

    Well thanks as usual, looking forward to read more about most of the topics you address. Cheers!

  2. Found these interesting paragraphs in The Wonder That was India by Basham

    > The traditional divisions of the Indian army were four: elephants, cavalry, chariots and infantry; some sources add other categories, such as navy, spies, pioneers and commissariat, to bring the total up to six or eight. Of these elements the most important, from the point of view of contemporary theory, was the first.

    > Elephants employed in war are first definitely mentioned in the Buddhist scriptures, where it is said that king Bimbisara of Magadha owned a large and efficient elephant corps. They were trained with great care and attention, and, marching in the van of the army, acted rather like tanks in modem warfare, breaking up the enemy’s ranks and smashing palisades, gates, and other defences (p. 459f); a line of elephants might also act as a living bridge for crossing shallow rivers and streams. Elephants were often protected by leather armour, and their tusks tipped with metal spikes. The Chinese traveller Sung Yiin, who visited the kingdom of the Hunas in the early Oth century, speaks of fighting elephants with swords fastened to their trunks, with which they wrought great carnage, but there is no confirmation of this practice in other sources. As well as the mahout the elephant usually carried two or three soldiers, armed with bows, javelins and long spears, and advanced with a small detachment of infantry to defend it from attack.

    > The great reliance placed on elephants by Indian tacticians was, from the practical point of view, unfortunate. Though fighting elephants might at first strike great terror in an invading army unused to them, they were by no means invincible. Just as the Romans found means of defeating the elephants of Pyrrhus and Hannibal, so Greeks, Turks and other invaders soon lost their fear of the Indian fighting elephant. Even the best trained elephant was demoralized comparatively easily, especially by fire, and when overcome by panic it would infect its fellows, until a whole squadron of elephants, trumpeting in terror, would turn from the battle, throw its riders, and trample the troops of its own side. The pathetic Indian faith in the elephants’ fighting qualities was inherited by the Muslim conquerors, who, after a few generations in India, became almost as reliant on elephants as the Hindus, and suffered at the hands of armies without elephants in just the same way. Cavalry, though significant, was not up to the standard of that of many other early peoples, and the weakness of their cavalry was an important factor in the defeat of Indian armies attacked by invaders from the North-West; the decisive victory of Alexander over Porus in 326 b.c. and that of Muhammad of Ghor over Prthviraja in a.d. 1192 were both largely due to superior, more mobile cavalry. Mounted archers were a special danger to Indian armies.

    Also, not mentioned in the book (as it relates to pre-Muslim India), major battles in later era also. This is covered in the The Wonder That was India Part 2 by Rizvi

    > The two major defeats of Mu’izzu’d-Dm show that Rajput military strength was quite formidable; their fault lay in their adherence to fighting a defensive war and their inability to take full advantage of a victory. The ponderous war machinery of the Rajputs with their slow- moving elephants was no match for the mobility and lightning attacks of the Central Asian Turkic guerrilla warriors. Separated from their homes by long distances and committed to obtaining victory for their own enrichment, the Turkic troopers fought with vigour and desperation. No such personal stake was involved for the Rajputs. Political independence and nationalism in the twentieth-century sense meant nothing to the twelfth-century Hindus and Muslims. p. 25

    > Elephants were effectively used from the days of Mahmud of Ghazni, who developed an impressive stable which contained 1,000 elephants, tended by a Hindu staff. The elephants were lined up in the front and centre of the armies. When out of control they played havoc with the ranks, but their advantages were believed to outweigh the disadvantages, and they were always present. p 175

    And Babur’s military genius at Panipat also made use of elephants weakness

    > Ibrahim left Delhi to meet Babur with an army of 100,000 cavalry and 1,000 elephants. Babur’s army comprised about 10,000 troopers. He took up a position near Panipat, keeping the town and its suburbs on his right. Cannons mounted on wagons had been placed in front of his army. Between sets of wagons, tied together by strong rope, were gaps where small breastworks had been erected to deploy one or two hundred horsemen. On the left, and in other strategic areas, ditches had been dug. Babur’s cavalry, or ‘flying flanks’, was expert in making lightning attacks on the enemy’s flanks and rear. Ibrahim’s army depended on elephants and consisted of mercenaries. The battle took place on 21 April 1526. Babur’s flying flanks wheeled from right and left to attack the enemy’s rear, showering Ibrahim’s troops with arrows. Ibrahim sent reinforcements to the hard-pressed flanks. Groups from Babur’s right, left, and centre surrounded clusters of Indian troops, while the cannons discharged balls on them from the left-hand side. By the afternoon the Indian armies, which had no experience of these tactics, had been routed. Between 16,000 and 40,000 Indians were killed. Herd upon herd of elephants was captured. p 92-93

  3. How did elephantry interact with opposing elephantry? Was the locking of tusks and goring behaviors in the wild the main way trained elephants with their mahouts and riders would fight each other, or were they trained or equipped to be more sophisticated (perhaps in concert with their supporting infantry and riders)?

    It certainly seems like elephant vs. elephant would be a common occurrence given that both would be commonly placed at the front of the infantry line, which in battles not involving elephants would often themselves clash. How long would this contact between opposing elephants last? Would commanders and drivers attempt to position and maneuver their elephantry away from other elephantry, to wield them against the infantry and cavalry where they might be used to greater advantage and less likelihood of being tied up or killed?

    Also, I’m really glad that you’ve written this series on war elephants. It seems like, as you noted in your fifth post on the Siege of Gondor, elephantry is a decent real life proxy for how one might expect fantasy creatures such as trolls (or dragons or griffins) to interact with armies in fantasy settings. Your third post in this series on elephants as symbols of royal power is particularly interesting, given that fantasy settings that feature ridable griffins, dragons, etc. will often limit their use to great heroes, kings, and the like, being an exaggerated and dramaticized parallel to how chariots, horses, and of course elephants were all used by warrior-aristocrats to both physically and symbolically place themselves higher than those they considered their social inferiors.

  4. Bret, here are some belated proofreading corrections for this older post:

    and increases interest in mating -> and increased interest in mating
    a mother elephant with her cub -> a mother elephant with her calf
    The African elephant which was -> The African elephant that was
    effect can resolved for horses -> effect can be resolved for horses
    Greece and Rome At War -> Greece and Rome at War

    Hope these are helpful. Thanks for your whole series of posts. I will almost be sorry when I catch up and have to wait for the next!

  5. The picture of “Hannibal traverse le Rhone” has broken: WordPress complains “You can only process images up to 57671680 bytes.” So you might want to replace it with an embed of the Wikimedia version.

Leave a Reply