Collections: The Battle of Helm’s Deep, Part II: Total Warg

This is the second part of a series taking a historian’s look at the Battle of Helm’s Deep (Part I here), from both J.R.R. Tolkien’s The Two Towers (1954) and Peter Jackson’s 2002 film of the same name. Last time, we looked at the operations which led up to the fortress assault. This time, I want to take a closer look at a sequence which occurs in the films, but not in the books: the Warg attack and the resultant cavalry engagement.

But before we start, we need to have a serious conversation. Cavalry is called that in English because it is made up of cavallo (Italian, meaning horse, from late Latin caballus, meaning horse), which gives the Italian cavalleria (cavalry), which arrives in English via French cavallerie (and has nothing to do with Calvary; cavalry is dudes on horses, Calvary is a very particular hill). So if you ride cavalli, you are cavalry. If you ride camels, you are camelry. If you ride chariots, you are chariotry. If you ride elephants, you are elephantry. I promise I am not making these words up; these are, in fact, technical terms. My students find them hilarious, but that does not mean they are not technical terms.

Now I’m going to make some words up, because I don’t feel like writing “orc-warg-cavalry” two dozen times. Following the rules for forming these technical terms, if you ride wargs, you would be wargry, which I suspect in speech would sound something like ‘WAR-ger-ree’ (because that -ry terminator is standing in for a French –erie). So we’re going to call this unit type wargry, both because it makes my life easier, but also because I very nearly burst out laughing every time I write it, and damn it, this is my blog and I get to have fun here sometimes too.

And, before we dive in, as a reminder – if you like what you are reading here, you can support me on Patreon; and if you want to be updated when new posts appear, you can follow me on twitter (@BretDevereaux) or subscribe for email updates using the button below:

What To Expect, When You Are Not Expecting Wargs

Just to refresh, Peter Jackson’s effort to resolve some of his adaptation’s narrative problems means that the nature of Théoden’s column has changed fairly radically. He has some unknown number of civilians with him, carrying their belongings. Most of the baggage is food supplies; some good management there, getting a column like this to move that light would have been difficult and it speaks to Théoden’s workmanlike generalship (writing this, it brought to mind immediate comparisons with J. K. Stearns’ description – which I have just chanced to have been reading – of the refugee columns that streamed out of Rwanda in 1994 which were, shall we say, far less competently or humanely led). Stripping a column – either a convoy or an army – down to bare essentials was a common literary trope to show effective generalship for a good reason. So it’s a good sign everyone appears to be traveling pretty light.

And that is about the only nice thing I’m going to have to say about this sequence. Buckle up, this is going to get rough.

The sequence begins with Gamling and Háma riding forward to do some scouting, so let’s start by assessing Théoden’s scouting system.

What I would want to do here is estimate the size of Théoden’s entire column, because it would greatly inform his scouting procedures. But working with the film like it is a historical text, the historian in me has to caution insurmountable problems. Our best account (the books) does not record this minor engagement (that problem is, as an aside, so common in ancient sources – this or that skirmish will be recorded in one source for a war, but not the others, because it wasn’t deemed important enough, leaving us desperately short on confirming evidence and openly wondering if the skirmish ever took place, or if an author merely added it for ‘storytelling purposes’), leaving us with only this later, less reliable narrative.

We’re told that “the city must empty,” meaning that the convoy here includes most of Edoras’ civilian population, but we have almost no way to know how much that is. Edoras as we see it has something close to 100 houses (mostly quite large, we may assume they house extended families, as is common in agrarian pre-modern societies; the nuclear family as an organizing unit over the extended family is mostly a feature of modern societies), but the hillside obscures large parts of the town, leaving us guessing as to the total number. Assuming many of these fine big houses likely also include servants and laborers, we might imagine a ‘city’ of perhaps 500 households or so; perhaps 2,500 people all told within the walls. Which doesn’t actually strike me as unreasonable for a settlement like this, except that Edoras has a severe case of Lonely City: there should be quite a lot more settlement outside of the walls, which our flawed representational evidence (the technical term for ‘historical artwork that depicts something’ – like using manuscript illumination to talk about knightly arms) gives us no means to estimate (but must exist because these folks have to eat).

Edoras, with the refugee train filing out of it. The density of the column varies wildly from section to section and scene to scene. Meanwhile, it is hard to figure how many homes Edoras actually contains (we never see the other side of his hill, which may have more settlement).

The other problem, on top of the uncertain number of refugees, is the density of the column. In the film, what we see is badly strung out. Here I must admit I am no expert on refugee movements, but even uncoordinated mass movements of people tend to be much more densely packed than this. Moreover, Théoden is moving this column through potentially dangerous territory (remember, he already knows that Saruman’s forces are burning the Westfold, which is where he is marching), so he ought to be taking some care to keep the column tight and compact. In some cases, we see them moving about four abreast, but with fairly large intervals, but in other cases there are great giant gaps and individuals moving alone.

An example of the wildly variant density of the column. In front, they’re moving six or so abreast and fairly tightly packed, but up on the hillside, the train is single file with massive intervals.

In practice, I have to imagine the limitation here is just the number of extras for these scenes. For the film, the audience just needs to think “oh, that’s a lot of old, vulnerable people our heroes need to keep safe!” We don’t need exact numbers. But it makes calculating the road-space of the column a fool’s errand. On the one hand, with just Edoras (2,500 people), if they move with tight spacing, about four abreast, the entire column might only be around half a mile long (assuming around 4 feet of road space per row). On the flip side, what we see of average spacing might be as low as 2-4 feet per person, given the wide gaps between groups and people moving apparently nearly single file; if we take that and assume a rural population to match the city (say, four country farmers for every town dweller), we end up with a (probably unmanageable) 10 mile long column.

That said, it doesn’t really matter too much how big the column is, because in any case, Théoden’s arrangements are uncharacteristically sloppy. His scouting apparently consists of two men on horseback, and Legolas’ elf-eyes. On flat, clear ground, this would not be enough. In the rough country he is inexplicably moving the column through, it is even worse. What should his arrangement look like?

I should thank Eric Burke of the Army University Press for pointing me towards some good resources on 19th century convoy tactics. Often the most important research skill is knowing who to ask. Just don’t ask Eric about infantry skirmish tactics in the ACW unless you have an entire afternoon and enough alcohol to while it away.

Well, reaching back a bit, G.H. Dufour’s (1787-1875; Swiss Chief of General Staff and noted Red Cross founder) manual – translated into English by W.P. Craighill in 1864, for reasons that might easily be imagined given the date – gives guidance on how a convoy ought to be escorted (361-372, trans. Craighill); fortunately this is a matter wherein gunpowder changes little. Dufour notes the necessity of both infantry, because “troops of this kind can fight on all kinds of ground” (a point we’ll return to) and cavalry (for its speed and scouting ability). For Dufour, there really only the one kind of infantry (unarmored, firearm bearing men), but to translate into a pre-gunpowder setting, ‘infantry’ here probably mostly means ‘light infantry’ – heavy infantry (armored men in tight formation) tends to struggle in rough ground. He breaks the cavalry into three groups, an advance guard two to five miles in advance of the column, a second group at the head of the column, and a rear-guard. The infantry is also broken into three main groups, one at the head of the column, one in the center, and one in the rear, but also with “a few men detailed from the detachments of the main body to march along the sides of the road and compel the drivers to keep their places and distances [emphasis mine].” A.L. Wagner’s 1894 Organization and Tactics (441-455) makes much the same set of suggestions.

Théoden’s entire scouting operation, by contrast, consist of a pair of riders (Gamling and Háma) riding out just in front of the army – we know they didn’t go too far because Legolas is able to catch up to them on foot to finish off the orc they spot and is then able to shout relay (Legolas shouts to Aragorn and Aragorn shouts to Théoden) the message back. In fact, as far as we can see, Théoden has no forward scouting force and no flankers. His retinue rides around him, somewhere in the middle-to-front (but not the absolute front) of the column, occasionally sending pairs of scouts forward. The retinue isn’t even concentrated around him, but has to be rallied to the front of the column. This is unconscionably sloppy.

Pictured: Théoden’s entire scouting system: one guy on foot with good eyes and two guys on horses with bad eyes.

This is not only a scouting problem – which results, by the by, with Théoden being forced suddenly into a cavalry engagement on bad terrain – but also an organizational problem. Crucially, the absence of flankers and a rearguard limit Théoden’s ability to control the column itself. Note the emphasis above in Dufour’s advice on using flankers to keep the drivers (of wagons) where they need to be – and he has in mind military wagons. With a cloud of civilians – some likely old, sick, or young – this becomes even more necessary. The flankers and rearguard need to be watching for heat and fatigue casualties, for people falling out of the line, for obstructions (a collapsed marcher, or a broken vehicle) that might block progress. With so few troops, Théoden does not have time to leisurely ride with Aragorn and talk about how bad a foster-father he was – he should be rolling up and down the column more or less continuously, trying to make the most of his limited forces. Because large bodies of (untrained) people do not organize themselves, and if you want to have any hope of controlling them, you’ve got to be doing that.

Théoden, Aragorn and the cavalry force just hanging out in the middle of the column. In this scene, it’s clear that they are nowhere near the front of the column (where the cavalry vanguard should be) but also that they are just sort of moving along. Théoden should have a lot to be doing, trying to coordinate a column like this with so few soldiers.

Now that brings up the question of how limited movie!Théoden’s forces are, a topic we’ll return to later in the series, once we have all of the elements ‘on screen’ as it were. But for now, I want to note that even with Éomer gone, Théoden should probably have more riders about him than the couple dozen he appears to have. Rohan is a pretty big kingdom. For comparison Edward I’s household – not his army, just his household (we’ll explain this a bit later) – in 1297 included around 500 armored horseman. Philip the Tall (of France) in 1317 had 235 in his household (figures from Verbruggen, The Art of Warfare in Western Europe During the Middle Ages (1997)). And that force would almost certainly be supplemented by other mounted aristocrats (and their retinues) who lived near or in Edoras (to be close to the center of power), plus many of the townsfolk fighting as infantry. In the books, Théoden’s lightning column has little space to bring infantry from Edoras, but with a refugee force, there’s no loss in speed from bringing infantry. Éomer may have taken some of these men (more likely the local aristocrats than the knights of Théoden’s house) but surely not so very nearly all of them. Théoden’s force here ought to be larger. At the very least, he ought to have organized the fighting-fit poor of Edoras into infantry to free up his cavalry (we see a handful of infantrymen, but nothing like the number of military-aged non-aristocrats a settlement like Edoras should have).

In any case, Théoden’s sloppy scouting arrangements allow a mounted orc scout (scouting for a larger wargry force) to get very close to the column undetected. How close? Well, Legolas watches the riders leave the column, within just a few yards of the leading civilian elements. Following film time-stamps (while we shift camera-lines, the audio holds consistent as the screams are audible from the column, so we have a continuous timeline), Legolas reaches the orc just 26 seconds after the attack begins, and is navigating very rough, difficult terrain, slows down to fire, and never appears to be in a dead sprint. I’d hazard that the orc encounter is probably something like 100-150 meters in advance of the front of the column. If Legolas maintains Usain Bolt’s average run-speed over the 100 meter sprint (10.4m/s) – the fastest any human has ever run in ideal conditions, which these are not – that would put the encounter 270 meters out, giving us an effective maximum distance. Chances are the orc scout was within bowshot of the column before being discovered. An ‘encounter battle’ like this should be the exact thing all of Théoden’s arrangements are designed to avoid; this sort of failure is out of character for him.

Théoden rapidly rallies his cavalry to the front of the column (it’s strange that he is not armored and fighting ready himself, given that he is apparently leading the vanguard). As discussed above, there should already be a large body of cavalry at the front of the column, but evidently there isn’t. Nevertheless, the orc scout’s carelessness gives him the time to get organized and advance – although not very far. He engages the enemy wargry within bowshot of Legolas, who has advanced to his position – perhaps 200 meters in advance of the column, generously. We’ve discussed the impact of range on the lethality of archery here, for now we’ll assume Legolas’ bow is the equal of a longbow or a Mongol bow (were it any stronger, he would need superhuman strength merely to wield it) suggesting a maximum effective engagement range around 200 meters. The orcs look to close about half of that distance, leaving the cavalry collision to occur only about 300 meters (Legolas’ c. 200 meter maximum sprint, plus half of bowshot) from the front of the civilian column.

Théoden shouting “All Riders to the head of the column!” Why wasn’t there already a large body of riders at the head of the column? Why wouldn’t, at the very least, your own retinue be with you at the head of the column, where your banner is? This was so blandly standard in medieval armies that a ‘banner’ could be used as a synonym for a lord’s retinue, because they ought always be in the same place.

Opposing mounted units should never have gotten anywhere near that close without being spotted, especially on this sort of terrain (I mean, look at those sight-lines – this isn’t jungle, you can see for miles from the crests!). At the absolute minimum, Théoden should have ordered Gamling and Háma to ride forwards hours ago, dismount at this point and scale those rocks to the left side of the path and set a watch at the top. Of course, what should actually have been done, is that there should have been an advance cavalry screen at least two miles (3,218 meters) in front of the column.

Théoden is only saved by the carelessness of the orc scout, who engaged while still being undetected, rather than sensibly rushing to report back to his unit. Had the scout not abandoned his mission’s objective, Théoden likely would have only become aware of the impending attack as the enemy wargry crossed over the ridgeline, perhaps 150 meters from the front of the civilian column, leading to an unorganized cavalry engagement amidst the civilians in the front of his column, with the likely result that he, the column, and consequently the war, would have been lost.

Now, one may reasonably ask why all of this preamble before we get to the ‘good stuff’ of charging horses and clashing swords. But the fact is that success and failure in war are often found in this sort of preamble. Gathering intelligence and coordinating large bodies of people is very hard (and, as an aside, something players of strategy or tactics games are almost never really asked to do). Attempting to do this sort of thing ad hoc and make it up as you go along are almost certain to fail because the task is too big and too complicated.

Instead, this kind of thing seems simple, because the armies that do it already have a ‘playbook’ for how a camp, or a column is organized, how scouts are dispatched, how watches are structured, and so on. Our sources occasionally give us enough detail to see that playbook (e.g. Polybius book 6), sometimes not. In later eras, that playbook gets written down and codified, first in a tradition of military manuals (beginning in the broader Mediterranean in the Hellenistic era; there’s a parallel tradition in China at roughly the same time) and then eventually in codified form as ‘doctrine’ – a topic that will have to wait for another series to get into.

But what I want to draw out here is that something as apparently simple as moving an army down a road – the sort of thing which might get a sentence in an account of a campaign (Théoden then marched into the Westfold…) is in fact very complicated and also very easy to get wrong if you haven’t planned for it carefully. As Clausewitz says (drink!), “Everything in war is simple, but the simplest thing is hard.”

Warg and Blade: Wargband

All of which gets us to the cavalry engagement between Saruman’s wargry, and Rohan’s more traditional cavalry, and it is a bit of a bother. As with other points at which the film has (for reasons discussed last time) deviated from the books, Peter Jackson doesn’t have the benefit of Tolkien’s reading of medieval sources, nor his intuitive sense of how things work and is left on his own, and it shows in the engagement proper as well.

Théoden has no way of giving orders save for shouting “all riders to the head of the column.” Strikingly (for reasons we’ll get to below) his standard bearers don’t stay close to him; they should be stuck on him like glue, because the unit goes where the banner is, and the banner ought to be with the king (whose protection is his retinue’s job). But there are no horns or trumpets, nor a call-response order echo (put a pin in this, we’ll come back). The riders seem to advance in no particular order (while the wargry advance in what I can only describe as a pell-mell rush); Théoden seems to get the semblance of a line of cavalry at the top of the hill, only for it to fall into a rough mass as they rush in, waving mostly swords and axes. There’s a collision, with horses and wargs literally slamming together, and then a disorganized melee with many dismounted riders running about amidst the cavalry confrontation.

A well-ordered advance in lines around the banner, this isn’t.

This is, charitably put, not how good cavalry engages. And the one thing we are told, again and again is that the Rohirrim have not only good cavalry, they have the best cavalry. I can believe that Saruman’s inexperienced wargry – who mostly fight as fast-and-light raiders rather than as heavy cavalry – do a poor job maintaining formation, but not the knights of Théoden’s house (again, we’ll get to what that means as a unit a little later in this series), who ought to be some of the best trained and most experienced horsemen in all of Middle Earth. So what should it look like?

(What follows is mostly based on my understanding of medieval knightly cavalry because that is very roughly what the Rohirrim are modeled off of (particularly Hastings/Crusade era cavalry, given the equipment discussed below). The best summary for the relatively new and curious is probably the section in C. Roger’s Soldiers’ Lives Through History: The Middle Ages (2007). From there, Verbruggen, The Art of Warfare in Western Europe During the Middle Ages (1997) is a standard reference text, but harder for a beginner to follow. I’ll also reference here Keegan, The Face of Battle (1976), which is effectively the foundation-stone of this approach in modern English-language scholarship, but also a bit dated in some respects. There remains substantial debate about the mechanics of cavalry in general and knightly cavalry in particular – for the most part, I am going to try to remain on the ‘high ground’ of generally accepted things, though I will have to offer opinion on one or two matters where debate still rages).

Well, first off, they should be in formation. I’ve mentioned before that there seem to be, in effect, two major schools of cavalry formation: tight lines (most frequently, it seems, double lines, although more lines were possible, with separations between them) or looser formations – tight-and-deep and disordered-mass are the two things to be avoided. Well, the Rohirrim, stirrup-using armored heavy cavalry (with lances, we’ll get to it), are the tight line sort of cavalry. For a large group of cavalry (like at Pelennor Fields) you might well stack multiple lines behind each other with separations (though in the books, the Rohirrim’s advance over the Pelennor is so broad as to not very many lines). In this case, Théoden has only a couple dozen horsemen, so a single line would be appropriate, given the wider battlefield. Loose formations (an open order with intervals) was sometimes used for heavy cavalry, especially in the East, but first, that was still a formation with regular intervals and second, on the balance, I’d expect Théoden to form up tight anyway.

This scene, from Return of the King, is a better sense of how this sort of formation ought to form up, although I think I’d still like perhaps a bit more of an interval between lines, and – if our sources are to be believed – even denser spacing.
From the Song of Raoul de Cambrai, “The barons are so closely packed as they advance / that if you throw a glove on their helmets / it would not fall to ground within a mile.” (that same image occurs twice in the song, and also elsewhere in the genre – it was standard imagery).

There’s simply no excuse for Théoden’s riders rushing forward in such a sloppy formation. In actual battle, knightly heavy cavalry formed up first and advanced second. Since Théoden’s column has already begun moving away from the impending fight, he ought to have time to draw his riders up into a proper formation (in practice, these men – likely the knights of the king’s house – should be so used to drilling and fighting together that they should probably form up naturally out of habit, without needing to be told). The importance of that formation will become clear shortly.

They should also be organized. The popular image of the knight in battle often remains that of an individualistic warrior out on his own, or of the untutored brute smashing away without reason. But these were professional fighting men, and that vision, informed by some very old scholarship, is badly out of date. On the battlefield, they were broken into a set of more-or-less standard units. The most common, as far as I can tell, was the conroi – from Old French conreer, ‘prepared’ or ‘equipped,’ so the conroi are a set of men who trained together – a formation of anywhere from five to a couple dozen knights, organized around a single banner (there are a few different types; Rohan’s banners are, well, banners, whereas a gonfalon – a vertical hanging banner – was also common). The banner provided a crucial organizational aid (as we’ll see) and these units seem sometimes to have been simply called ‘banners’ (a knight who had received permission to carry his own banner and thus have his own unit was a ‘knight banneret,’ distinct from a knight bachelor, who would be one of the subordinate knights in someone else’s conroi and follow their banner). The eschelle or scara (both mean ‘ladder,’ the former in Old French, the latter in Latin) seems to have been a similar unit; I’ve only encountered it in reference to the military orders (like the Templars), but I don’t know if that is a general principle or just an accident of where I’ve seen it. In practice, Théoden looks to only have with him one, or perhaps two, conroi.

Book Note: We’ll come back to the organization of Rohan’s forces a bit later, once we have them all on the board, but it seems worth noting that the thousand horsemen Théoden has in the book would require larger divisions. A number of conroi would be grouped into a ‘battle’ (Old French batailles); a large army might be made up of a number of these batailles, often ten or fifteen. It seems that Rohan’s equivalent unit is the éored; at Minas Tirith, the Rohirrim are divided into more than three companies of this sort (he gives orders to three, but then bids others to follow the three in the lead). We ought to imagine Théoden’s army being subdivided this way, with a number of conroi (we do not, as far as I can tell, get the Rohirrim equivalent unit name, but doubtless there would have been one) grouped together into several éoreds, with each éored having a leader, and then each conroi also having a commander (likely a major landholding aristocrat).

Those conroi should be organized around their banners, which gets us into all of the command-and-control that is not happening in this scene. The banner was crucial, as it gave individual riders a point of reference to organize around, a place to return to if separated. With room for local variation, there were also systems of signals and commands. Since conroi were relatively small, many of these commands could be relayed by voice; for larger units, trumpets and horns might signal a general advance. The commands were about what you would expect – advance, stop, spur to a gallop, slow to a step, fall back; the exact words differed over languages. In at least some cases, these calls had either standard echos or responses to make sure the entire unit was paying attention (Verbruggen, 87). There do seem to have been standard orders, and a real concern that the formation – again, a tight line, stirrup to stirrup – not be disrupted in the advance.

(As an aside on battle-cries, because it really doesn’t fit anywhere else: it seems pretty clear to me that the battle-call of the dwarves, “Baruk Khazâd! Khazâd ai-mênu!” (“Axes of the Dwarves! The Dwarves are upon you!”) was probably been inspired by the long traditional battlecry of the Gurkhas, “Jai Mahakali, Ayo Gorkhali!” (Victory to Mahakali! The Gurkhas are coming!” It has a similar length and cadence, with the final clause heralding the impending arrival of fearsome foes)

The opposing wargry is even more poorly formed up, with massive, irregular intervals and a formation that is deep, rather than wide, which is the rough equivalent of going into a game of checkers by first making sure you have left nice, one-space intervals between all of your pieces.

So what should the approach look like? Théoden ought, once he has gathered his cavalry, to organize it around his standard (something he does in the books, RotK, 121, but not in the films), drawing it into a tight line. He ought to then order the advance at a trot, keeping his line dressed. Then the order to spur, first to the canter, then to the gallop. As they come into the gallop, the spears (we’ll get to them in a moment) – which will have been held pointing upwards to keep them out of the way – should come down into striking position. It’s likely that the formation naturally loses some rigidity in that last charging gallop, as stronger horses outpace weaker ones, but that is why the gallop is held until relatively close to contact – the aim is to keep a solid line (that this is ideal resounds from our sources, which talk a lot of serried ranks, tightly packed, saying things like “the wind could not blow between their lances” or how “it was not possible to throw a prune except on mailed and armored men,” quotes from Verbruggen, op. cit.). Given that the Rohirrim are exemplary horsemen and Théoden’s own conroi ought to be some of the best of them, we ought to imagine a tight formation, carefully controlled, even at the last charge.


The reader may be pardoned for having finished the previous paragraph in the hope that the next one would begin with a shattering collision of cavalry and wargry (this joke shamelessly adapted from Shattered Sword). But we need to talk about the equipment that both of these forces are carrying, because the film doesn’t quite have it right.

Let’s start with the Rohirrim. When we see them, we generally get a wide array of weapons. Gamling and Háma ride forward and try to defend themselves with one-handed arming swords (of mostly sensible design), but no shields, helmets or spears. When Théoden rallies his riders, we see a mix of swords, spears and axes. No one carries a shield, but some of the horses have shields still strapped in their carry positions. And those are, overall, the right sort of weapons, but in totally wrong proportion.

Charging Rohirrim. Ignoring Aragorn, I count 3 swords, 3 spears (two are hard to see), the banner pole and 2 axes. The correct count probably ought to be 8 spears and a banner.

Let’s deal with the first potential question: why no specialized lances? I should clear up my terminology: on horseback, the ubiquitous one-handed thrusting spear which occurs in almost every sedentary culture is often called a lance, but for the sake of specificity, I am going to call it a spear and reserve the word lance to mean the highly specialized high medieval cavalry lance (this is, should note, not the historical usage). Should we see that sort of lance here? I’d argue: no. That specialized high-to-late medieval lance – thicker and more reinforced and eventually made to utilize a specialized lance-rest constructed into the breastplate – strikes me (speaking as a scholar of arms and armor) as an evolutionary response to armored knights fighting other armored knights. It is the consequence of an arms race between offensive and defensive technology which has not yet taken place in Middle Earth, where the primary body defense is still mail (which a normal spear, held in the hand and delivered at a gallop, is perfectly capable of defeating). That armor-context is part of why I put the Rohirrim closer to the cavalry on the Bayeux Tapestry (showing the Battle of Hastings, 1066) than plate-armored late medieval knightly cavalry. Lance-rests and lances would be out of place on cavalrymen armored in scale and mail, riding on as yet largely unarmored horses.

Section of the Bayeux Tapestry, showing the Battle of Hastings (1066), with heavy cavalry using both overarm and underarm spear-grips. It should be noted that overarm grips dominate the total number of depictions in the tapestry.

But lancer cavalry – meaning cavalry using any sort of spear – is far older than the specialist lance, and – as an aside – far older than stirrups (I most emphatically do not ascribe to the stirrup thesis; Alexander charged his cavalry just fine without them, but it matters little here, as the Rohirrim have them regardless). And it isn’t that some of the Rohirrim should have spears, all of the Rohirrim should be using spears as their primary weapon. Swords, axes, maces – on horseback (and frequently on foot too) these are, with only a few exceptions, backup weapons.

The spear is just perfect for use on horseback. It can be easily wielded in one hand (good, because your other hand has the reins and your shield), it is long enough to deliver a strike both to the ground (because you are elevated and infantry targets are below you) and in front of your horse. The latter thing is important, as we’ll see when we get to impact, because it may happen that you end up stopped, or nearly stopped, facing or alongside a mounted opponent. Unlike an infantryman, you cannot necessarily close up, because your horses are in the way, so a weapon with reach is crucial. Finally, the narrow tip of the spear allows you to concentrate all of the energy you build up with your horse’s speed into a tiny, penetrating area of impact, which is ideal for defeating armor, particularly mail.

I count nine riders in this shot, five four spears between them. Five if you count the banner-pole, but you shouldn’t – the banner is a signalling device first.

But wait – what is a rider to do when he is actively accelerating himself towards the spear-points of his enemies which we just established are ideal for punching through his mail, scale or thinner lamellar armor? Carry a shield of course. Despite the fact that we see quite a lot of sensible Rohirrim round-shields (a kite shield might make a bit more sense for the horse-lords, but round shields on horseback show up all over Eurasia, so there’s no problem here; one assumes the round shield was chosen because of the Anglo-Saxon vibe the Rohirrim have), they almost never carry them, instead leaving them on their backs or on the horse (this is a lot more obvious with the charge of the Rohirrim at Minas Tirith).

Book Note: No such problems exist with the equipment as described in the books. Théoden engages the chieftain of the Southrons with his spear first and only resorts to his sword when the spear breaks; his shield also figures prominently (TT, 171; RotK 126). The Rohirrim’s shields and spears are likewise not absent (TT, 172; RotK, 121, 126).

I actually suspect this has to do with set-safety and stunt difficulty. Peter Jackson has to have a lot of riders on his set and few modern riders are experienced handling a horse with a shield in their hands (still fewer with the center-grip style of shield the Rohirrim use. This style of shield absolutely can be used on horseback – Gallic and Roman cavalry did so – but it may prove trickier than the later medieval shield suspensions). Add on to that that the Rohirrim prop-shields are overly thick and heavy and thus difficult to use effectively and you can see why, for safety reasons, they are left on the horse. Likewise, I am guessing that there was simply no way to make a prop-spear on a horse safe (it’s simply too fast to risk a wooden pole with any kind of rigid tip connecting with someone on the ground), which is why the Rohirrim spear-wielders almost always keep their points up, even when charging. Consequently, you want to limit the number of actors with prop-weapons they can’t use and keep them in the background.

All of which is no excuse for the weapons the wargry brings to the battle. While some of the orcs bring curved, relatively long one-handed swords (sensible!) many of them appear to only have much shorter swords or daggers, which are hardly ideal to use from warg-back. Worse yet is their choice of polearm, these things:

A selection of orc mounted weapons, what look like two long-hafted glaives, a bardiche and on the right, possibly a bearded axe? Also a terrible choice for use from horseback or wargback.

Now, those aren’t fantasy weapons from nowhere, so much as fancified versions of real weapons. The weapon held up in the center looks like a bardiche, while the two long blades held low look to be some sort of glaive (note: polearm nomenclature is confusing, complicated and wildly inconsistent from one area of the world to the next, and from one era to the next; I tend to lean on E. Oakeshott’s European Weapons and Armour (1980), just to use a set of terms that can be readily understood by others). These are, on the whole, pretty bad weapons to bring to a cavalry engagement, or really on warg-back at all.

The problem is that both of these are two-handed cutting weapons. But remember: you ought to have – as these orcs do – one hand on the reins of your mount. It should not be hard to imagine how awkward attempting to cut with these weapons one handed would be. And the bardiche is even worse, because you have to swing it while somehow keeping the long haft clear of your mount. Now, I should note that a short-hafted form of the glaive is sometimes seen used by cavalry as a two-handed slashing weapon:

Via Wikipedia, glaive/warbrand use from horseback from the Morgan Bible (folio 10v), c. 1240. Note the short haft of the glaive.

But so far as I can tell, depictions of this are extremely rare, and the glaive’s near-relative weapons (like the guisarme) seem to primarily be polearms intended for use on foot. In this case, the glaives the orcs are using seem to be long-hafted, which of course poses the same problem as the bardiche. These seem to me to be infantry weapons, largely unsuited for use from warg-back (oddly, this isn’t the only fantasy setting where someone thought that infantry polearms were cool and should be just as cool mounted). The orcs would have been far better off with simple thrusting spears. Oh yes and – given how lightly armored they are – shields.


Now we move to matters which are quite a bit less settled: what happens when these two formations meet. There’s considerable debate about what happens in just about any kind of shock engagement (that is, the violent meeting of close-order bodies of troops in melee, either mounted or on foot). In part, this is because we simply cannot effectively observe the question the way we can with, say, ancient ironworking techniques, nor could we easily observe and document the phenomenon in places where it was still practiced, as with subsistence agriculture, to develop clues. Attempting to create a modern test to replace that runs into problems with safety, especially when dealing with horses, which are plenty heavy and fast enough to seriously injure or kill someone without any weapons at all.

So we are left with the testimony of our medieval or ancient sources, often filtered through the somewhat (but often only somewhat) better sources for the use of cavalry in the modern period (which raises all sorts of awkward questions about how useful largely unarmored Napoleonic lancers are for understanding heavily armored knightly cavalry). And our ancient sources are on the one hand stylized artistic depictions (sculptures, manuscript illustrations, etc.) which often struggle to accurately display their subjects and often don’t care to be accurate in the first place. On the other hand, we have textual sources, which often struggle to use language to really describe combat; imagine trying to describe a fire-fight to someone who has never so much as seen a firearm used. And worse yet, over all of this is a layer of heroization and propaganda which runs through the sources for warfare and tends to further distort them from the reality on the ground. It leaves an uncomfortable amount of space for conjecture and theory, and thus debate.

So I cannot claim that my approach here is the be-all and end-all, but I will at least try to ‘paint within the lines’ of the broad scope of what the sources and the scholarship suggest at the moment of collision.

In essence, we might imagine two bodies of heavy cavalry closing on each other as a sort of ‘decision-tree.’ Our sources are really quite clear that cavalry closing to the attack spur to the gallop, so we can start there: both groups of cavalry are in fairly tight order (we’ll get to orcish indiscipline in a moment) and closing rapidly.

The first option is offered to us by simple physics: if neither formation looses up or slows down, they will collide. A gallop tends to be around 25-30mph, but since they are galloping at each other, the real collision speed is the sum of their velocities – so we have a c. 1,300lbs warhorse with a c. 250lbs of rider, armor, weapons and tack on it (effectively a single, 1,550lbs object), colliding with another c. 1,550lbs object at something like 50-60mph. The energy release of that collision is equivalent to a four-door Honda Civic wrapping itself around a tree at a little over 30mph (48.7kph for the metric among us). Being unhorsed in any situation risks potential injury – being unhorsed in a high energy collision as a tangle of metal plates and sharp weapons, as two 1,300lbs mounts come down around you is quite likely to cause catastrophic injury to the rider and almost sure to cause such damage to the horse that it will never charge again (a real problem because war horses were very expensive; even a wealthy knight might think twice before casually disposing of one).

We thus tend to think that this sort of collision did not happen often. Someone would slow down, turn aside, lose their nerve and avoid the almost certainly fatal crush. But in some ways, a cavalry charge is a game of nerves: coming in at a strong gallop looking like you are willing to risk that collision is a great way to panic your enemies, make them turn aside, and gain the advantage.

But what if no one loses their nerve, so the formations hold together? We tend to think both formations actually slow on the final approach, coming together at non-catastrophic speeds. At which point, there is an effort to ‘push through’ with swords, lances, spears, and so on, at close quarters, cutting and stabbing at enemy riders in front of you or two your side. Talking about this sort of press is tricky, because we find that our sources often like to represent as a collision what we later find out to have been a press (perhaps one of the most famous examples of this is the accounts of Waterloo, where multiple British officers maintained they had met the French at speed, but on closer inspection, it turns out to have been closer to a press, Keegan, Face of Battle, 147-9).

A press of cavalry, from MS. Bodl. 264, 74r, dated to 1344. Note how the horses are effectively stationary as they fight, at a stand still.

I think the press must have been common, because the sources place such strong emphasis on a dense, tight formation, which would be of greatest value in the press. If both forces meet, but one is loose and the other is tight and disciplined, the latter will present the former with a series of many-on-few melee combats, which ought to result in fairly rapid success (as anyone who has ever tried to fence multiple opponents alone with any close combat weapon will rapidly attest to the near impossibility of it, even with a shield). The spear is a crucial weapon here because the rider will often be striking over the head of his own mount at the enemy and his mount – reach is a crucial advantage, although in a very tight press, the greater precision of the sword dominates (thus the swords in the image above).

But did these formations ever come together at speed? This is a matter of some degree of uncertainty, but I actually think that some of the times they must have. If a charge wasn’t carefully kept regulated (something that must have been hard to do), the differences in the speed of the horses would open up gaps, loosening the formation. Alternately, men losing their nerve in the gallop may have slowed while their compatriots kept on, which would achieve the same result. In a large battle, this might be entire units – as some conroi slowed for the press and others charged home, again creating gaps which might be ridden through. Keegan raises another possibility, an ‘opening of ranks’ whereby the formation is intentionally loosened in the approach to allow the two bodies of cavalry (or wargry in this case) to pass through each other at high speed without the mounts colliding and he presents some Napoleonic era examples.

Some problems must at once be admitted. Doing this seems to cut directly against the emphasis on ‘tight’ knightly formations which resounds from the period literature. And I think there is a degree of artificiality (what Keegan phrases as ‘consent’) which the mind rebels against when it is framed that way. But I tend to think this happened. First, I do not think ‘consent’ is the right way to think about why – I think this happened as the natural response of men driving their horses forward and yet looking to avoid the catastrophic impact discussed above; it was later formalized into orders, but avoiding collision is a natural response for both humans and horses. And we see quite a lot of period artwork that shows individual or groups of horses, clearly at the gallop (all legs in the air at once) meeting head-on (Rogers, op. cit., 192 has a particularly good one which I cannot find online).

But I am most moved by the evidence of training for this, because this seems to be exactly the sort of thing that the quintaine (a charging target) or the tournament trained at (note that early tournaments trained this in units, not generally as individual contests – it was often a contest between two conroi). We see the same in modern lance and sabre drill, which survives just long enough to be documented in video (note esp. at 0:52-1:03 and 1:51-1:59) – notice how some of those sabre stances involve pointing the tip forward level with the rider (which means at another rider, not at a man on foot!), and then swiftly pulling it aside. If you are wondering what that second, silly looking movement is for, look into the modern sport of tent-pegging, which is an evolution of drills for spearing infantry from a horse – the flick to the right or left is to avoid losing your sword in the opponent after ‘giving point’ at high speed. I find it hard to imagine that all of these fellows, from the 11th century knight tilting at his quintaine to the late 19th century cavalryman, are training for a sort of combat that they never engage in.

At the same time, even in this meeting, keeping order was essential. It is not hard to imagine how one-sided an affair it might be if a conroi in good order charges against a conroi where some of the horsemen are fleeing, others are charging, and others are slowing – the formation in good order is going to scatter the other, almost immediately and likely inflict some losses (mitigated by the heavy armor everyone is wearing) while suffering very few themselves.

Likewise, the final option here is not a head-on-head engagement at all, but one in which a well organized conroi of cavalry catches another in the flanks or rear. With formation being so important and horses being so hard to turn, a formation struck in the rear or flank may have little choice but to scatter and flee. As Verbruggen argues, cavalry commanders were well aware of this vulnerability and both went to great lengths to protect their flanks (either with supporting units or, as often, with the terrain itself, anchoring a flank on some natural obstacle) and to try to exploit an enemy’s flank.

Conclusion: Towards a Better Charge

All of this is only a very brief summary of a complex topic, but it should by now be quite clear that despite Théoden and the Rohirrim’s reputation for cavalry excellence, their performance looks almost nothing like this. They attack in disorder, make little attempt at a formation and the entire scene devolves into a series of confused individual combats. What I think might be more instructive than continuing to beat up on this scene is instead to present a proposed revised scene more in keeping with what we’ve discussed, one with exemplifies the Rohirrim’s well-earned position as the ‘horse-lords.’ So what ought it to look like?

As the warning call comes from Aragorn, we ought to see the riders of Théoden’s house begin immediately coming forward and forming up around the king’s banner as by habit. This could be happening in the background while Théoden is giving instructions to Éowyn (so we don’t lose that important character beat), so that, by the time Théoden come up to the head of the formation, they are already in a line. The odd fellows out here ought to be Legolas and Gimli, who haven’t the right weapons or training for this; I might have them follow the line in and dismount when the fighting starts. If we want to get really complex, they might be backed up by some of the Rohirrim infantry, who might form up on that rocky ridge-line, taking advantage of the terrain (which honestly suits infantry far better than cavalry). Aragorn, by contrast, has adventured around the world, and might know this business well enough to form up behind the king’s banner.

Théoden then advances his cavalry at a trot to the top of that ridge we see. Each rider ought to have a spear in their right hand (held upright for now) and a shield in their left. The opposing wargry comes over the opposite ridge and – overeager and undertrained – breaks out into a charge straight-away, leading to their formation spacing out from the indiscipline. I might give the orcs spears and sabres (or just sabres, if we want to keep them visually distinct); the wargry is a light raiding force, not a heavy contact force, so sabre cavalry might be a good model (and, as an aside, woe-betide sabre cavalry that finds itself fighting lancer cavalry). Théoden orders his riders to advance, first to the trot, then spur into the gallop only as he gets close, maintaining a nice, dense stirrup-to-stirrup line as he does it. Théoden’s conroi of perhaps 25 is outnumbered, but tighter and better disciplined.

The orc riders – who were expecting to raid nearly defenseless civilians – respond unpredictably. Some slow to avoid the collision, while others spur onward, maybe even a few try to turn away. Because there are more of them and their formation has loosened, many of the orcs find themselves well to the outside of Théoden’s conroi – still more opt to avoid the solid line of the Rohirrim and intentionally steer out to the right or left. As all of this jostling makes tatters of the wargry formation, Théoden’s forces hit the center, expert horsemen guiding their mounts through the gaps created, spears striking out at orc riders as they pass; many of those spears will break with the force of impact. Those riders whose spears break swiftly draw swords or axes.

As Théoden’s force clears his first pass, only a few of his riders are down, but the center of the field is a mess of orcs and wargs. But he can’t stop, because the groups that went to the left and right are now behind him. Théoden – master cavalry captain – swiftly gathers his riders again under his banner (the obvious place to reform) and wheels about, turning left to take one of the two pockets of wargs left on the field. The Rohirrim reform and turn faster, striking the warg riders before they can properly reform, but without the devastating force of the first charge, leading to a press as some of the wargs from the other group arrive. At the same time, a handful of warg riders make tries at Legolas and Gimli’s small clump of infantry, and are toppled to ruin in the rough terrain for their trouble. The press provides the opportunity for Aragorn to “take [his] little tumble off the cliff,” but by now the wargry’s numerical advantage is gone and their cohesion is shattered. Théoden’s tighter formation off-sets the obvious advantage that a warg might hold over a horse in the press, because his riders can protect each other with their spears (keeping the wargs’ teeth and claws at bay); the press is brief and the remaining orcs scatter and flee, leaving Théoden holding the field.

Our fight now hits all of the right beats – Théoden outnumbered, ordering Éowyn away, a fearsome charge of wargry, Rohirrim victory with losses, Aragorn falling off a cliff – but now does so in a way which demonstrates Théoden’s leadership ability and the prowess of the riders of Rohan. It shows them winning through tactics, discipline and carefully honed riding skill, rather than a confusing melee where they win because they are the good guys. And it gives Gimli and Legolas something to do in the fight that makes sense, if we include the supporting infantry.

That said, I think given this alternative scenario, it should be obvious why they didn’t do this. Making this work on film would take a lot of planning for a fight which is just supposed to be a small prelude to the real action at Helm’s Deep; even more planning to do big aerial shots if you want the audience to be able to keep track of where everyone is and what they are doing. All of that would get expensive, fast. Moreover, complex wheeling maneuvers with tighter-order cavalry carrying spears and shields would probably have been too unsafe a thing to ask your riding stunt-fellows, much less your main stars, to do. But this is 2002, and you just can’t CGI the whole thing, because while your CGI stands up OK in the long-distance wide-shots, it isn’t good enough for the close-up work (something the near-contemporary Star Wars prequels will test to their great detriment).

So I get the difficulty, but it is too sad, because there are just very few good scenes of cavalry engagements at any scale.

Next week: all of the armies arrive at Helm’s Deep, and we talk about how to organize your pre-modern army so it can actually function!

157 thoughts on “Collections: The Battle of Helm’s Deep, Part II: Total Warg

  1. Thank you so much for your great work.

    There was one point that I don’t think is quite up to the latest research:
    “A gallop tends to be around 25-30 mph, but since they are galloping at each other, the real collision speed is the sum of their velocities”

    Of course, I am not aware of anybody attaching measuring tools to two armored knights during a real collision, but there is a lot of testing of Honda Civics and other vehicles. Your later statement describes the results of two Honda Civics hitting each other head on at 25-30 mph quite nicely. Thus, not the sum of their velocities.

    “The energy release of that collision is equivalent to a four-door Honda Civic wrapping itself around a tree at a little over 30 mph”

    It’s a small point of pedantry, but students might notice.

    Here’s the mythbusters video, but there is more serious research available.


  2. Hello, contemporary media portrays mounted characters in the Romance of the Three Kingdoms as wielding spears and other polearms one-handed during the charge, but two-handed during what I think counts as the press. See this youtube video of a fight scene from the 1994 TV series as an example:

    Would cavalry using polearms two-handed in such situations be historically accurate?


  3. Excellent analysis, and a great outline for an alternate version of the scene, though I get what you mean about the impracticality of staging a more elaborate and realistic cavalry charge.
    Makes me think of the cavalry charge scene from Kingdom of Heaven, where they did have the resources to stage a bigger clash of cavalry, and it looks very impressive! There are however some strange maneuverings in that scene that I struggle to see a reason for; as they ride out from Kerak to buy time for the civilians to get to safety the knights form into a line, then the reform into a column, then suddenly they’re back into a half-formed line again.
    Any thought?


  4. Love the SHATTERED SWORD reference.

    The best cavalry charge I’ve seen on film is in Olivier’s 1954 version of RICHARD III; the armour is very costume-y, and we don’t get to see the actual moment of impact, but the horsemanship and music really give you the feel nonetheless.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s