This is the second part of a six-part series taking a military historian’s look at the Siege of Gondor in Peter Jackson’s adaptation of Return of the King. We discussed the operational objectives and logistics of the Army of Mordor last time. This time, we’re looking at what Gondor’s defense plan is, how plausible it is, how likely it is to be effective, and also some of the tactical mechanics of how it plays out.
Defense in Depth
Let’s start by laying out a theoretical term: defense in depth. Instead of trying to ‘hard stop’ an attacker with a single, maximally strong defensive line, defense in depth seeks to slow down or damage an attacker while yielding space. One of the great virtues (but not the only one!) of such a defense is that it turns friction into an ally. Armies are hugely complex things, involving many moving parts (people, equipment, animals, etc). Friction (pedantry note: here in the sense used by Clausewitz) is simply the tendency for things to begin to go wrong with that system as it moves and fights. As Clausewitz says (drink!), “Everything in war is very simple, but the simplest thing is hard.”
You can think of it this way: when an army first jumps off on the offensive, it has had time to plan and prepare. Positions very close to the army are under good observation, so information is more accurate (and there has been time to sort out the inaccurate reports). Everyone is in the right position, everyone’s weapons are in good condition, everyone is fed and rested. As soon as the first step forward is taken, that begins to break down: key specialists are killed, equipment breaks, soldiers get tired, scared, lost, bored. Intelligence is swallowed in the fog of war. An attack is thus at its most dangerous at the very beginning, before it is worn down by friction. An attacking army is in the most danger at the end of an assault, exhausted and worn down – such a moment is the perfect time to counterattack, if forces remain to do so.
Defense in depth seeks to exploit this tendency, maintaining a measure of defense pressure on the enemy to inflict attrition, delay and friction, but with a flexible enough defense to enable the defenders to repeatedly withdraw to new lines, preserving a potent force for a potential counterattack.
Gondor’s Operational Plan
Now that I’ve lost half of my readers with theory, let’s hit the punchline: Gondor’s plan is to delay via a defense in depth. In the film, Gondor’s armies have four major defensive lines (counting from the earliest to make contact): 1) Ranger forces in Ithilien which aim to wear down enemy forces with hit-and-run tactics, 2) an initial line at the Anduin, anchored by defensive positions at Osgiliath and Caer Andros, 3) the open Pelennor Fields and 4) the city of Minas Tirith itself. Minas Tirith then replicates this system in microcosm, as it consists of seven (!) concentric rings of fortified city, with each ring having its main gate on the opposite side of Mount Mindolluin from the previous, forcing the enemy to zig-zag their way up. To take the city, any army from Mordor has to punch through each of these zones in sequence.
Nothing about this plan is necessarily impossible from a historical standpoint, but I do find it necessary to note that, with seven walled city rings, Minas Tirith is substantially more fortified than any walled city I am familiar with. Many walled cities had citadels – secondary defensive fallback positions within the city. Castles often had two, or even three, concentric rings of defenses. But seven such rings? Around an entire city? For comparison, the best defended city in Europe throughout the Middle Ages was Constantinople, with a single double-wall circuit (double walls meaning a lower forward wall set in front of the main wall). But then, the defenses of Minas Tirith were built by the Men of Westernesse at their height, so they can be a bit fantastic.
Book Note: Gondor’s defensive systems are more complex in the books, and this matters for assessing Denethor’s leadership during the battle, as we’ll see. In the books, there is another defensive ring, the Rammas Echor, a wall which encloses the Pelennor Fields. The Causeway Forts, a pair of fortifications in the Rammas Echor, challenge any approach down the road from Osgiliath. A large stone wall like this would be a significant undertaking, but is perfectly plausible – Hadrian’s Wall in Northern England covers much more space.
The Rammas Echor and its Causeway Forts significantly change the playing field in a number of ways. First, it limits the danger of forward placing forces in Osgiliath. Even though the circuit is far too large to be held long-term against any army large enough to force a crossing at Osgiliath, by delaying the advance of any attacker, the Rammas Echor screens any retreat. Moreover, it provides another opportunity to inflict attrition and confusion (read: more friction) on an advancing force without risking a pitched battle in an open field, or (in the film) a cavalry charge into a rock.
Gondor’s defense-in-depth system makes particular sense because it aims not only to inflict losses on enemy forces but to delay them. Gondor’s most populated lands are around the Bay of Belfalas (places like Lebennin and Dol Amroth), the latter of which is around 200 miles from Minas Tirith. Edoras, where the riders of Rohan might be looked for, is just as far. The longer the forces of Gondor can keep the roads to these regions open, the more reinforcements they can receive. Likewise, as discussed last time, an enemy attack in force is likely to be in a tricky supply situation, where unexpected delays might prove fatal.
We’re going to skip around a bit to keep this clear – we’ll come back to the assault on Osgiliath in a moment, but first I want to talk about the beacons.
You may be waiting for me to now say that the Beacons of Gondor is a silly, a-historical thing that could never happen. Nope. Not only is this system plausible, it existed, on a similar scale, in the 9th century Byzantine (read: Eastern Roman) Empire. The system stretched more than 400 miles from the frontier to Constantinople, and consisted of fire signals set on high ground at intervals of 30 to 60 miles.
This was particularly important because of the way the Byzantine Empire’s army was organized into themata or themes. Each theme was a combined military and civil administrative district, with its own small field army that could respond to local raids – however a theme’s army would be insufficient to respond to a major invasion. In the event of a large attack, the beacons rippling back to the capital would bring the tagmata, the main imperial field army, which tended to stay wherever the emperor was (to discourage rebellion in the themes). The tagmata could then roll out to confront the invasion, picking up theme forces as it moved (forces which – because of the beacons – would already be ready). It was an effective system and despite the Byzantine reputation for decline, the period from the 9th century to the 11th century was a period of Byzantine reconquest and consolidation (until the Battle of Manzikert in 1071).
What is baffling is Denethor’s decision not to light the beacons. In the film, Gandalf and Pippin have to do this themselves. What is more incredible is that this is happening as Osgiliath is under attack. This is far, far too late. Osgiliath is, after all, only around 30-40 miles from Minas Tirith, whereas Théoden’s rescuing armies – in the film, these are the audience for the beacons – have far further to travel. Denethor’s reasoning is that he knows that Aragorn is with the Rohirrim and does not wish to bring him to Minas Tirith, which would require Denethor to give up lordship. Given the situation, this is foolish to the point of madness, something Gandalf points out in dialogue.
I should also note that it takes too long for the beacon system to run its course in the film. When the first beacon is lit, the scene is in the evening, and we then see beacons lit at twilight and into the night, before the last beacon being visible at Edoras in afternoon (look at the shadows from the West). In fact, the Byzantine beacon system was reputed to be able to send a signal to Constantinople in a single hour.
Book Note: In the books, Denethor is nowhere near this stupid (this will be a theme). First off, the beacons are lit on the night of March 7th (Pippin notes it was three nights since he had touched the Palantir, RotK 20), before the Witch King had even departed Minas Morgul (which happens on the 9th). The Eyes of the White Tower are not blind, indeed!
The audience for the beacons is also different: Denethor is not summoning Rohan with them, but his own dominions. In the film, Gandalf implies this is not done (“Where are Gondor’s armies?”), but in the books, this question is answered – the threat of a black fleet sailing up the Anduin has forced many of the troops in the interior to be held there against that threat (RotK 46). Nevertheless, smaller forces from all over Gondor arrive in the city before the battle is joined in earnest, most notably Prince Imrahil of Dol Amroth and his force of knights, who seem to make up the sortie we’ll discuss momentarily.
The beacons thus face south of Ered Nimrais (the mountains separating Gondor and Rohan), not north towards Rohan. To summon Rohan, Denethor must send a messenger with a ritual ‘red arrow’ – Hirgon (the messenger) meets Théoden at Dunharrow with the arrow on the 9th. Assuming the arrow was sent on the same day that the beacons were lit, Hirgon will have had to ride quite hard, although the distance is not impossible. Because of Gandalf, the armies of Rohan were, in fact, already mustering and the army was set to ride out on the following day, and cover the distance to the Pelennor over the next five days.
This in turn neatly explains why none of Rohan’s infantry – in evidence at Helm’s Deep under the command of Erkenbrand of the Westfold – are at the Battle of the Pelennor Fields. Théoden has to cover nearly 200 miles in five days – 40 miles a day is on the outside edge of the possible campaign speed for agrarian cavalry (Steppe warriors, like the Mongols, could make up to 60, but the Rohirrim are not nomads). It seems likely that Théoden’s riders brought spare horses (as almost any knight would – expect at least three horses to sustain a knight on campaign (pack horse, charger and a riding horse)), and likely spared their chargers by riding down the others to move so quickly. Théoden also has the advantage of moving over maintained roads and known lands for most of his trip and – as we’ll see later – he makes excellent use of local knowledge and native guides.
That said, the lighting of the beacons is one of my favorite sequences in all of film. I love it, I love the rising music, the gorgeous mountains and the mounting sense of “oh yeah, the bad guys are in for it now.” So I understand the desire to have this on screen and have the main characters involved, rather than merely something Pippin sees at a distance while riding past.
Meanwhile, at Osgiliath, bad tactics
The tactics the orcs employ here is a night-time crossing of the river in small-boats and using the mist to cover the advance. I think I will surprise some of you when I say I actually find this plausible. The garrison at Osgiliath is small – the movie does not make its size clear, but it is obviously hundreds, not thousands. This is a force the orcs might expect to be able to overwhelm with a fairly small advance force. George Washington was able to make a night-time crossing of the Delaware in bad weather (Dec 25th/26th, 1776) with around 2,400 men (roughly half of the army was unable to cross on schedule) to ambush 1,500 Hessian troops at Trenton the next day. Looking further back, historically, the Romans made an opposed crossing from Italy to Sicily to open the First Punic War in what seem to have been mostly small craft – at around 3.5 miles wide, this was a far longer crossing than what the orcs here intend.
What is extremely rare are opposed landings, which almost never happen before the modern era. That said, this is a decent exception – in the book, Denethor notes that the Anduin is not easy to cross elsewhere (too wide and swift) and moreover, the presence of repairable bridges make Osgiliath the obvious point. Mordor’s numerical superiority and preparation are such that they can count on being able to force the crossing, albeit with heavy losses.
In contrast, the position of Gondor’s forces at Osgiliath is a bit puzzling. Though the ruins of the city itself provide a measure of natural fortification, especially against missiles, there is no evidence of any effort to improve those defenses. The Romans are perhaps the most famous for fortifying their camps – Roman armies built a fort of wooden palisades around their camp every night – but almost any pre-modern army of any significant organization would fortify a camp it expected to be in for any length (or in proximity to enemy forces).
What might such impromptu fortifications look like? Roman camps were surrounded by a wooden palisade of close-set stakes (Polybius notes that Roman palisades were unusually robust, compared to their Greek equivalents, Plb. 18.18.1-17), around which would be set a trench (the agger) – the trench was not to provide cover, but to foul approaching enemies. Material from the agger was used to build up the height of the wooden palisade, and the agger itself was often filled with obstructions and hazards like sharpened sticks.
Gondor’s positions at Osgiliath could easily be so enhanced, stringing palisades and ditches along the waterline between the stone ruins. Faramir’s lack of field fortification here is careless.
Mordor’s forces attack in a large number of small boats over the river, attempting surprise. They are foiled by a watchman (credit to Gondor for good watches), who is shot through the breastplate. To be clear: there is no range at which an arrow will penetrate a decent steel breastplate (see here, for instance), so this man should be fine. Nevertheless, surprise is blown and Faramir opts to meet the enemy on the waterline as they disembark. That makes sense: you do not want to give them time to form up and get organized.
What does not make sense is how he springs his ambush. He has his soldiers hide behind the ruins, letting the first rank of enemies past, before charging the waterline. This does have the benefit of cutting off the first wave of attackers, who are now trapped between Faramir’s front line and his reserve, but it creates a chaotic melee in which both sides are sure to take heavy casualties – casualties which Faramir can ill afford.
What would better tactics here look like? Surprise actually benefits Faramir here little – his enemies are equipped and expect a fight. But troops making the crossing in poorly protected (uncovered) smallboats are vulnerable to all kinds of missile fire – not only arrows, but javelins or even heavy thrown stones (especially if you have slings available). Faramir’s first order should be to get his archers on those ruins and firing as quickly as possible.
Second, there is no need to cut the enemy off – by virtue of making the crossing, they have cut themselves off. Since the orcs are disembarking their boats, they are out of formation already – if they were to slam into Gondor’s plate-clad heavy infantry deployed in close formation on the beach, they’d be cut to pieces. Engagements between lighter infantry in loose formation and heavy infantry in good order tend to be shockingly one-sided.
This is a common error on Hollywood battle scenes: everyone breaks out of formation into a pell-mell melee. I have actually been asked in class, by students, quite seriously “what happens when this all dissolves into a chaotic melee” to which I had to dash their preconceptions by noting that (we were discussing hoplites), if you are winning, it never does. And if your formation breaks, chances are you have already lost. Most close-quarters infantry rely on tight formations to be effective. The large shields and spears carried by Gondor soldiers – akin to the kit of a Roman triarius – would clearly work best in a shield-and-spear wall.
Since simple numbers mean that Faramir is still sure to be overrun, he should probably deploy close to the main road out of the ruins – aiming to smash the first wave of orcs at the center and delay the bridge-crossing, but withdrawing before his flanks are turned by the landings further north and south which he hasn’t the forces to oppose. As his tactics stand now, Faramir’s ‘riverside ambush’ inflicts losses on the enemy, but also wastes most of his own forces.
Book Note: The books do not take us to the action as Osgiliath and we only hear about it second hand. What we see first instead are the Nazgûl attempting to run down (fly down?) Faramir and four of his horsemen (RotK 90), reporting on enemy movements in Ithilien. This is actually a clever use of fantasy airborne assets – rather than rushing them uselessly into clouds of arrows, the Witch King is using their tremendous speed and mobility to try to cut the line of communications between Minas Tirith and its forces. Alas for them, Gandalf has brought his AAA (did you not know? His staff packs Bofors!)
Denethor then recommits to the defense in depth strategy, sending Faramir back to Osgiliath with reinforcements (RotK, 98-9). What we are told next is that the small boats of the crossing ‘swarmed across like beetles’ (a description that makes me think they were covered to protect against arrows), but that the orcs took heavy losses making it across and that the attack might even have failed but for the supernatural power of the ‘Black Captain’ – the Witch King (RotK 99).
Charge the ruins! Perhaps the stones will run away?
We pick up with Faramir as he retreats to the city (we’ll deal with all of the Rohirrim’s scenes when they arrive). Continuing the film’s tendency to undermine its own characterization of Faramir as the finest captain of Gondor, the retreat is headlong rout, with no rear-guard. His disordered mass of cavalry and infantry retreating over open ground (trust me, we will get to that in a later post – why is Minas Tirith surrounded by grassland?) gets predictably savaged.
Nevertheless, Gandalf’s magic flashlight saves the rout and the troops get back to Minas Tirith. For some reason, while all of this is happening, the orc army does not actually advance, but just opts to chill out in Osgiliath, giving Faramir the time to ride all the way to Minas Tirith and then ride all the way back. That is, at minimum, a 60 mile round trip and should have taken him at least two days – the problem here is that Peter Jackson has (as we’ll see) moved around events that occur in a different order in the books, so Tolkien’s plausible timetable gets mangled into something that makes little sense.
Denethor now orders Faramir to retake Osgiliath. Oddly, a line from an earlier point in the books (“I will not yield the river and Pelennor unfought”) is brought forward to here, where it makes no sense – Faramir has already fought for, and lost, the river. Denethor then launches Faramir on a cavalry charge against a city, albeit a ruined one.
This attack makes no sense on multiple levels. First, the distance is nonsense – Osgiliath is thirty-some miles away. Faramir could ride half the day, camp, sleep out the night, and still be too far away to charge in directly. Second, the terrain is nonsense – cavalry need open spaces to operate in, but the enemy has not left Osgiliath. These guys are going to have to dismount and fight on foot when they get to Osgiliath and we’ve already seen that a much larger infantry force failed to hold the city. Surely one would wait for the orcs to get into the open field before attempting to charge them. Thirdly, the size of the cavalry force that is sent out is ludicrously small.
Finally, the overall purpose of the charge makes no sense. As I’ve discussed elsewhere, cavalry charges are all about morale impact. Large bodies of infantry in good order cannot simply be run over in the field (we’ll get to it!), so instead a charge aims to break the morale of the defenders, causing them to scatter and then be easily run down. But a ruined city cannot panic and run away. The orcs Faramir is charging cannot flee even if they wanted to – they have their backs to the river. Moreover, while standing against a charge in open ground is hard (horses are scary!), standing against a charge while behind a stone ruin is relatively easy. “The horse cannot get to me here, I am two stories above the ground, in a well-built stone building,” is an excellent morale builder.
Narratively, Jackson wants us to see Denethor’s callousness. As a piece of film, the scene is magnificent, with Denethor’s eating creating a brutal commentary on his gluttonous waste of life (Dan Olsen lays this out in more detail as an aside in the middle of an interesting discussion of the Triumph of the Will and how Nazi propaganda continues to shape the perception of Nazis in popular culture). But as tactics, we have to assume that Denethor is either stunningly incompetent or already a touch mad at this point (something Gandalf more or less says earlier in the film).
That’s not a charge. Now this is a charge.
Book Note: This sequence runs very differently in the book, both because of the more detailed military topography of the battlespace, but also because Denethor is not actually a mad fool (yet).
Faramir’s return to the city (RotK 90) comes before the assault on Osgiliath. So instead of sending Faramir on a suicide charge against some stone walls, Denethor is instead sending Faramir out with what reinforcements can be spared to Osgiliath.
This changes the character of Denethor’s decision. He is aware (Faramir has indeed just told him) that any troops sent at this point to defend the crossing are likely to suffer high casualties. To ‘spend’ (as he puts it) his own sons so is perhaps callous and certainly ruthless, but it has sound sense to it. Faramir is the finest captain he has, and this is a difficult but necessary assignment. No one – not even Faramir – suggests calling back the forces at Osgiliath or further north at Caer Andros (which has, in fact, at this point already fallen, but this is not known yet).
Whereas in the film, Gandalf advises Faramir to give up the attack, in the books, he merely urges caution and for Faramir not to die in the field refusing to retreat (a suggestion clearly made by Denethor’s emphasis on ‘the manner of your return’ – win or don’t come back), advice which Faramir follows.
This order also makes sense of the transit time – Faramir is not charging 30 miles into a stone city, but leading a force along well-controlled roads to reinforce a friendly outpost – he is not in combat until two days later, and not wounded until a day after that.
We see little of the fighting at this point, but from what we do see, Faramir’s handling of his forces is quite good. The orcs swarm the Anduin in small boats (RotK 99), Faramir bleeds them for crossing and then withdraws, apparently in good order, to the Rammas Echor and the Causeway forts. Faramir holds there long enough for retreating forces from Caer Andros to reach him (RotK 101).
Faramir then begins withdrawing troops as the Army of Mordor begins blasting holes in the Rammas Echor (RotK 100). Faramir stays with the rearguard, aware that the Witch King’s power of fear might rout his army, and that he needs to retreat in good over across the Pelennor, inflicting losses as he goes. By keeping part of his army intact and ready to fight, he forces his enemy to advance in slow, plodding battle formation – remember, this plan is all about friction and delay. Faramir is also able to send wounded and exhausted soldiers ahead.
The rearguard, one mile out is “a more ordered mass of men…marching not running, still holding together,” supported by Faramir’s cavalry (RotK, 102). Holding together a retreat like this is incredibly difficult – the inherent psychological urge in humans to run would be nearly overpowering – suggesting that Faramir really is the masterful captain he’s cracked up to be. The retreat, harried by enemy cavalry and ring-wraiths, finally breaks into a rout ‘scarcely two furlongs’ from the city (440 yards), when Faramir is injured.
Denethor – having coolly observed all of this – now releases his sortie (which he had held prepared all morning) like a boss. In breaking through the defenses and pursuing Faramir’s forces, the vanguard of the army of Mordor is strung out and has its formation all broken up (there’s that friction again). Denethor’s sortie reminds me of the Battle of Hastings (1066) – there William (the Conqueror)’s Norman knights had charged the Anglo-Saxon line, which had held and when a cry went up that William was killed (he wasn’t), had retreated. The less disciplined Anglo-Saxon fyrdmen (the levy infantry) charged down the hill after them in a pell-mell pursuit much like the one Faramir just lured the orcs and Southrons into. Thereupon William, having gotten control of his forces, promptly wheeled them around and charged the Anglo-Saxons out in the open, annihilating them. Infantry out of formation is extremely vulnerable to cavalry.
The sortie is crushing, as you would expect it. Gondor’s cavalrymen had surprise, and a short distance to charge. The enemy cavalry is exhausted and all of their forces (they have supporting infantry) are out of formation. Denethor is quick not to let his men get overeager and he calls the charge back swiftly. Good command there – successful cavalry charges have a tendency to overreach. At Magnesia (190 BC), for instance, the Seleucid cavalry, after breaking the Roman left flank, proceeded to charge straight off of the field towards the Roman camp, both removing them from a battle in which they were sorely needed (the Romans were triumphant in the center and on the right, winning the battle) and leaving the Seleucid cavalry isolated and vulnerable to counterattack. Denethor makes no such mistakes.
Once again, we see Denethor’s willingness to risk even his own son for a military advantage by holding his sortie to the last available second, but unlike in the film, here that decision makes clear military sense. Denethor is willing to coldly sacrifice his son for a military advantage, but he is not mad. Like both of his sons, he is a masterful commander, even if he lacks his younger son’s wisdom.
We then pick up again with the orc army having reached Minas Tirith and Denethor viewing them from atop the city. We’ll come back to what the orcs are doing here next week. But for now I just want to note the comedy of Denethor declaring at this point, “Rohan has deserted us. Théoden has betrayed me!” which is pretty rich given that you didn’t call them. Denethor then rants like a madman (something he does not do at this juncture in the books) and gets clubbed by Gandalf while his sworn guards do not-a-damn-thing.
The Perils of Film
Overall, I think this is one of the least satisfactory parts of Jackson’s adaptation from a military historian’s perspective, but I can understand why he restructured it this way. Much of the tension in the books is carried by dialogue – we are told how badly outnumbered Faramir is, how many men he has lost, and how callous and demanding Denethor is being. In a filmic medium, however, we need to see this and we need to see it much more quickly and more clearly.
The way film treats time is also creating demands for Jackson. In a book we can accept jumping around chronologically from chapter to chapter and so Tolkien has a lot of freedom to weave what becomes four parallel narratives (that is, 1) Gandalf and Pippin in Minas Tirith, 2) Merry, Théoden and Eowyn with the riders, 3) Aragorn, Legolas and Gimli and 4) Frodo and Sam) together in large, chapter-length chunks that are easier to follow. But Jackson has to shift between these parallel threads in short snippets (often just a few minutes) and still keep his audience clear on what’s going on – an audience he cannot count on having read the books.
That said, I think it unfortunate that much of Faramir and Denethor’s skillful handling of the battle gets a bit lost in the need to move through the necessary emotional beats. Denethor, especially, is a complex character in the books – he opposes Mordor and never considers betraying the good guys, but he is also vain and arrogant. His arrogance is perhaps most shown in his use of the palantír, a weakness which Sauron uses to drive him to despair and eventually madness. Nevertheless, Denethor, as Gandalf notes, sees further than other men, and has a deep cunning to him (RotK 26). The film, however, reduces Denethor to a mad villain almost immediately.
In contrast to the film, the book shows Faramir and Denethor’s handling of the battle as nothing short of a masterful execution of defense in depth. At each stage, the army of Mordor is forced to sustain casualties and disorder to surmount one set of defenses, only to be presented with fresh defenses and troops. At the end of it, Denethor’s sortie shatters Mordor’s vanguard and buys the escape of Faramir’s force. Thus for all of their pains and delays, the Army of Mordor faces a Minas Tirith fully defended, having lost the chance to destroy a good part of the army of Gondor in the field.
One thing I truly love about the Lord of the Rings in general is that it does not rely – as so much fiction does (looking at you, Game of Thrones) – on the ‘good guys’ making stupid mistake after stupid mistake in order to create tension. Instead, Gondor executes its plans admirably, and yet it is so outmatched in military might that it remains in peril. Tolkien’s familiarity with medieval literature – especially Anglo-Saxon – shows through here, as he has a solid grasp on the mechanics of the battle. He can make his heroes militarily clever, because he has taken the time to understand these very mechanics.
Next week, the orcs arrive at Minas Tirith, Amazon-Three-Week-Shipping (no rush!) finally delivers a siege, and we look at how one goes about taking a fortified city.