Crowd othismos model

If you've read the paper below then you already know this, but I will summarize the key points and contrast them with the other two alternative views (simplified of course).

Orthodox view: Othismos is literally a mass pushing match with the aim being to push opponents back until the cohesion of their battle-line breaks. In the most widely proposed form, the clash of hoplite phalanxes progressed along the following pattern: Both phalanxes charged at a slow run from about 200 meters apart. Hoplites move directly into shield-on-shield contact from the charge using the momentum to smash their shields together like rams and stabbing with spears underhand like cavalry lances. Spears are often shivered and opposing ranks become to some extent interlaced. This is followed by intense infighting with swords as ranks reform through a process not well explained and the ranks behind the front-rankers begin to push forward. It is this pushing phase that is labelled “othismos” (pushing) and the pushing is done side-on to the man in front with the left shoulder in the bowl of the shield. This othismos continues until one side gives way and collapses. Once one side collapses the victors pursue (but not too far) and the losers sustain many casualties.

Heretical view: There are many variations of this, so I shall present it as I think most consistent with the hoplite panoply. The progress of battle differs from the orthodoxy in that there is no running charge directly into combat. They note that the run would cause disorder in the ranks that this would be counter to the whole idea of forming ranks in the first place. Combat occurs at spear's length, perhaps with shields overlapped, perhaps not, in a phase known to the Hellenistic Greeks as doratismos. Fighting might then progress to infighting with shield-bashing on an individual, uncoordinated scale. While the front-rankers and the one or two ranks behind them fight, the men to the rear provide only moral support and make ready to step over their corpses to take a place in the battle line. The advance of the phalanx is figuratively labelled “othismos” as we might speak of an armored "push" of tanks and mechanized infantry. Fighting occurs until one side gives way due to mounting casualties and morale failure. As before, the losing side suffers as the victors pursue.

Othismos-crowd model: This view incorporates elements of both of the above. In my view there is a running charge from a couple hundred meters as in the view of the orthodoxy, but with spears held up overhand. This does not lead directly into othismos, for there is no crash of hoplites seeking to use momentum to bash into their foes, and thus no interlacing of opposing ranks. Front ranks pull up from the charge prior to shield contact with their foes, then close to the approximately 5' distance from their foes that reflects the reach of the dory, entering doratismos as in the heretic's view. They then spear fence en masse as envisioned by the heretics, but at a spacing of something less than 1 m frontage, their aspides just overlapped. Only the second rank uses their spears in support of the front rank.
At some point there is a shift in phase to fighting at less than spear range. Closing to less than spear range would be natural if a hoplite’s spear was broken, for he would then find himself with the choice of standing in the ranks unable to reach the enemy, leaving the ranks to move in close, or staying in the ranks, but pressuring the men beside him to move in as a group. As more spears are lost this urging in close magnifies until the line moves. This describes how the move into close range could emerge without any central command, but it is possible that men were purposefully led into close range fighting.
It is possible to move from advance to othismos without a spear-fencing stage, but it is unlikely that this occurred as the more recent formulations of the orthodoxy put forward, with men running into each other in an uncoordinated manner. In order to maximize force, the men must pack their files tight before they hit the enemy phalanx rather than run in directly from the charge. Spartans would have had an easier time of this due to their slower advance.
However it occurs, when the opposing units are in close contact, the rear rankers close up behind those in front for physical support. The weapons used by hoplites are well designed for this type of fighting over the top of the shield rims. The ranks continue to tighten until the men are belly to back with the men in front and behind. How long this takes can vary with training in a polis or over time, but eventually this compact mass enters the othismos phase of battle. If the opposing front rankers are already shield to shield, then the progression to othismos is gradual and fluid. If they are at a distance, then there is a short and shuffling charge by the whole mass (as seen in the videos on this site). The two phalanxes now function like crowds, generating intense force as they push against each other. Men in the middle of the mass do not control their own movement, but ride the waves of flesh, all the while fighting and defending with sword or broken spear in their upraised right arm. The pressure is enough that these men would be asphyxiated without the aspis. If shields break under the pressure men die, unless they can breathe within the bowl of the overlap of the man to their right's shield. There may be lulls in the combat where the opposing phalanxes loosen this tight level of packing due to exhaustion, the front men may still be fighting or they too may pull apart. Eventually one side gives way and the same pursuit seen in the other models occurs.

As you see, my model incorporates the possibility of extended initial doratismos with literal othismos. My othismos is a far more brutal thing than that described by the orthodoxy. The momentum of men charging at 5 mph produced forces that are similar to the maximum seen in the crowd-crush. More importantly, individual collisions are also instantaneous, lasting a few miliseconds, while the pressure within crowds can be maintained for extended periods and is far more lethal. Men pushing side-on with the shoulder in the shield cannot generate the enormous crowd-like crushing forces that men pushing belly to back can. As the pressure builds, side-on men will collapse to belly to back if there is room to, and if there is no space between them and the men beside them, then they are vulnerable to asphyxia since their diaphragm is not in the belly of the shield.


To a buzzard circling the battlefield othismos would look like this, with only the rear rankers able to push sideways.



If we compare a side-on to belly to back postures of hoplites n the phalanx, you will see the difference in the density of the crowd packing, though the side-on man have a more narrow frontage. Under pressure these sideways men will collapse to face forward, absorbing the force of the men pushing them from behind instead of transferring it to the man in front of them. The pushing forces generated by a crowd can be greater than the smashing force of an initial running collision of shields, and there is no disruption of order from interlacing front ranks.



As I am able I will add more comparative details of the mechanics. So far I have not leaned heavily on primary sources, but I am putting together a post in which I shall interpret the words of ancient authors to support my position, just as both the heretics and orthodoxy have used alternate interpretations of many of the same quotes to support their position and rule out the other.

How phalanxes clashed

To date there has been no hoplite reenactment groups with enough men and the will to go at it full speed to properly recreate phalanx on phalanx combat. There are reenactors of other periods who have "discovered" the same crowd mechanics that I elucidated below. The following is a video that my friend Giannis Kadoglou found thst shows how large groups of armed and armored men collide. At about 20 seconds into the video you will see a close approximation a clash of shallow phalanxes. Note that as I described, they stand facing their opponents, not side-on as had been widely proposed and now seems generally accepted.

The Aspis: surviving hoplite battle. Part 2

Those who support a literal interpretation of the othismos as a pushing match have pointed out that the hollow design of the shield allows the left shoulder to be placed within it both to support the weight of the shield and to allow for a sideways pushing stance, while the flattened face provides a broad surface for pushing against the men in front of you. While correct in some details, this scheme fails when we apply a realistic model of othismos mechanics.

John Keegan, in The Face of Battle, noted that a crowd is the opposite of an army when he applied crowd psychology to formed men, and that crowd-like behavior signaled immanent defeat, but the ancient Greeks harnessed the force of a panicked crowd and turned it into an offensive weapon. The modeling of how force is generated in crowds is in its infancy, but the destructive potential is shown by the many tragic deaths caused by crowds colliding during sports events or fleeing in panic.

The outcome of a collision of ranked hoplites is not simply a matter of the number of men on either side. Most of the force applied by the rear ranks will be simply absorbed by the mass of their own men in front of them. In order to maximize the pushing force of a crowd, the distance between bodies must be minimized to the point that individuals lose control of their own movement and the group becomes one mass pushing in synchrony. In crowds of this density, shock waves are produced that can tear off clothing, lift people off their feet, and propel them 3 m or more through the air. These forces are generated by a domino effect of people leaning against each other and pushing in the same direction at once, and have been shown to exceed 1000 lbs of force and bend metal retaining structures. Death occurs in these conditions due to compressive asphyxia when the diaphragm is crushed and breathing is impossible.

Men must protect against asphyxiation if they are to subject themselves to these forces for the duration of battle. This is the aspis’ primary function and a role for which it is uniquely designed. The shield’s large diameter arose from the need to hold the shield across the front of the body, its flat rim resting on the upper chest and thighs, while the depth protected the diaphragm and allowed the hoplite to draw breath. The central position of the porpax ensured proper alignment, but left about a third of the shield extending beyond the hoplite to the left. As individuals with their shields tight to their chest came up behind the overhanging shields of men to their right, overlapping right over left, a phalanx assembled like building blocks. The job of rear rank veterans, who could push with their shoulders, was not simply to keep men from fleeing battle, but to keep them packed belly to back and as tightly as possible. The lethal zone in a crowd of this density extended well back into the phalanx, so the risk of death by asphyxiation was shared more equally among ranks than the danger from weapon strikes.

Along with the characteristic aspis, a second element of the panoply associated with the emergence of hoplites is the sauroter, a specialized butt-spike for the spear. The sauroter has been linked to phalanx combat through its use as an auxillary weapon, but this role was secondary to its use as a staff in steadying a man in the rear ranks and allowing him to add the strength of his right arm in pushing.

Weapons could still be used in the press of othismos, as the raised right arm would have just enough room to brandish a weapon in an overhand strike in the “V” formed by overlapped shields. The downward stabbing strike of a spear would require very little range of motion to be deadly, while the point-heavy chopping swords commonly used relied on a snap of the wrist more than a broad slash. The most deadly weapon in this press would be the short Laconian dagger stabbing in a downward strike from above. If the othismos gradually became the phase of battle that decided hoplite battles, this may explain the abandonment of body armor and enclosed helms for the high-peaked pilos that protected from overhead strikes. Any benefit of armor in the crowd would be outweighed by the need for increased stamina and the ability to breathe freely and hear commands.

That the Spartans developed a specialized weapon for the othismos is perhaps indicative of their role in perfecting this phase of combat. We are told that the Spartans did not excel in combat because of the martial arts teaching of hoplomachoi, but because of their singing and dancing. The reason for this is obvious if we accept the model presented above. Accidental synchronicity of effort is what builds lethal shock waves in crowds, so men who have trained to coordinate their motions through group dancing and rhythmically chanted songs will have an advantage in producing and amplifying forces in the othismos. Theban success in battle was explained in part by their skill in wrestling which develops kinesthetic sense and ability to read and anticipate motions.

Thebes met Sparta in battle again at Leuktra in 371 B.C. in the ultimate othismos battle. In an outcome presaged by Koronea, the Thebans countered Spartan skill with mass. The 12 ranks of Sparta collided with 50 ranks of Theban hoplites, who had added Boetian merchants and baggage-carriers to the rear of the phalanx. Our model of othismos helps us understand what happened next. The deep Theban formation did not crash into the Spartans and immediately drive it from the field. What followed was an almost tidal play of crowd against crowd. The synchronized Spartans could push back the Thebans, as they did to claim their wounded King, but each time they did this they packed them tighter, forcing them into a coordination that they may not have achieved on their own. There may have been long lulls where exhausted men simply fought for breath in the loosening crowds. Epaminondas’ called-for “one more step” was in reality a shuffle, but the Thebans eventually gained ground in a ratcheting advance that broke the Spartan ranks and their hegemony.

The othismos turned a phalanx into more than sum of its ranked hoplites. To fight within such a crowd was to submerge one’s individuality, and facing one was like standing against single scale-breasted beast, a many-headed hydra of down-thrusting weapons. Perhaps it is fitting that the states who mastered it were both descended from Heracles.



Paul M. Bardunias is not a historian, but an entomologist who studies group behavior in social insects- termites and ants. On the theory that one Myrmidon is as good as another, he is applying concepts from his background in biology and crowd behavior to an examination of the evolution of Greek weapons and tactics. He was born to the topic, his family comes from Sparta, but this is his first publication specifically on Greek warfare. He lives and works in Hollywood, Florida, USA.


Further reading:

J. Fruin, The Causes and Prevention of Crowd Disasters, in: R. A. Smith, J. F. Dickie (Eds.), Engineering for Crowd Safety. Elsevier, New York, 1993.

A. K. Goldsworthy, The Othismos, Myths and Heresies: The Nature of Hoplite Battle. War in History 4: 1, 1-26, 1997.

V. D. Hanson, The Western Way of War: Infantry battle in Classical Greece. New York, Hodder & Stoughton, 1989.

J. Keegan, The Face of Battle. London, Cape, 1976.

R. D. Luginbill, Othismos: The Importance of Mass-Shove in Hoplite Warfare. Phoenix, 48:1, 51-61, 1994.

N. Secunda, Greek Hoplite 480-323 BC. Osprey Publishing Ltd., 2000.

A. Snodgrass, Early Greek Armour and Weapons. Edinburgh University Press, 1964.

The Aspis: surviving hoplite battle. Part 1

The Aspis: Surviving Hoplite Battle.


Agiselaos II of Sparta stood victorious on the battlefield of Koronea in 394 B.C. when he learned that the Thebans had defeated his allies and were looting his baggage train. Showing more bravery than tactical sense, he formed his phalanx directly across the Theban line of retreat rather than taking them in the flank or rear as they passed. Sources do not reveal how many ranks of hoplites faced each other in this struggle, but the Spartans likely formed in 12, while the Thebans may have formed as they had at Delium, in 25 ranks, for a similar breakthrough attempt. The Spartan force, “…crashed against the Thebans front to front: and setting shields against shields they pushed, fought, killed, and were killed.” (Xenophon, Hellenica 4.3.19). The Spartans were victorious, but the Theban loss obscures the fact that their extra-deep phalanx broke through the Spartan phalanx. A description of the aftermath of the battle conveys the brutality of the clash: “a weird spectacle met the eye… the earth stained with blood, friend and foe lying dead side by side, shields crushed to pieces, spears snapped in two, daggers bared of their sheaths, some on the ground, some embedded in the bodies, some yet gripped by the hand.” (Xenophon, Agiselaos 2.1.14).

What Xenophon described is the result of a phase of hoplite battle termed Othismos. The term implies “pushing”, and has traditionally been taken to mean a clash between the massed ranks of hoplites where the goal was literally to push back the other formation until it broke due to a loss of cohesion and failure of morale.

The mechanics, and even existence, of this mass-scrum have been hotly debated over the last half-century in a series of papers whose titles imply nothing less than a protestant revolution. One proponent of the “heretical” view points out that no other army fought through such a pushing match: “Should it be proved that the othismos really was a contest of massed shoving…It would be necessary to explain how the Greeks were able to fight in this unique way, and why they did so.” (Goldsworthy)

The answer to Goldsworthy’s challenge lies in understanding the hoplite shield, which is often referred to as a hoplon and seen as lending its name to the hoplite. However, the term hoplon can refer to any implement of war. The specific term is aspis and sometimes rendered as Argive aspis, although the connection to Argos, either as originator or mass- producer, is unclear. The shield may have originated as early as the late 8th or early 7th century B.C. and survived in nearly identical form for at least three hundred years.

Round to slightly oval in outline, normally 80 cm to 1 m in diameter and about approximately 7.5 kg weight, its characteristic features are a convex profile and a robust, offset outer rim. A surviving Etruscan example in the Museo Gregoriano of the Vatican that appears similar to less well preserved examples from Greece was constructed of 20-30 cm poplar wood planks glued together horizontally to form a solid block, then turned on a lathe to form the characteristic shallow dome shape that left the flattened center of the core thinner than the curving outer edges. This shield has a diameter of 82cm, a depth of 10 cm and a 4.5 cm off-set rim reinforced with wooden laminates. Construction methods may have changed with time because shields represented on the 7th century Chigi vase appear to be made of layers of thin wooden laths. The shield’s inner face was usually covered with leather and sometimes highly decorated. The face of the Vatican shield is covered in a seamless .5 mm layer of bronze that overlaps the inner face of the rim by 4 cm. This fully bronze-faced front of the shield is commonly portrayed, but archaeological finds show that bronze plating of the rim alone, perhaps with a bronze blazon, was more common. Some depictions show additional metal bands reinforcing the inner face of the shield.

In use the aspis was supported by a double grip system. At the center of the shield’s concave interior was a bronze loop, the Porpax, through which the left forearm was placed to bear the weight of the shield. The porpax was secured to the shield’s inner face by long vertical bronze straps. This structure could be one piece or made so that the armband was removable, as the Spartans are said to have done to render their shields unusable to rebellious Helot serfs. A second grip, the antilabe, was for the left hand and was placed near the rim. The grip was a pair of metal staples through which a rope passed that traveled through four or more metal rings spaced around the inside perimeter of the shield.

This peculiar shield’s evolution is not well understood. Round, bronze-faced shields were common in the region prior to its appearance, but these were generally single-grip and did not display the twin features of domed shape and off-set rim. Assyrian infantry carried large, deeply convex shields of bronze, but they had a single grip and were cone-shaped. Urartian bronze shields, which have been excavated at Toprakkale near Lake Van in Anatolia, were domed with offset rims. Their diameter, 77 cm to 1 m, differed little from hoplite shields. These had a triple-grip system that has been interpreted to be a single hand grip and two points of attachment for a neck strap or telamon, but may have influenced Greek design.

Much attention has been paid to the convex shape of the aspis. Convex or sloped armor presents a greater thickness of wood to be penetrated by weapons impacting at oblique angles. This fact alone might explain the convex Assyrian shields, for their cone shape is the optimal profile to maximize this effect. The Greek shield is less well adapted for this, since the greatest slope is relegated to a small area at the outer edge, while the broad face of the shield is a shallow curve. Curvature also insures that chopping weapons will be impact on a broad area rather than biting into the rim of a shield. The semi-cylindrical examples of the Roman scutum may be made for this, but in the aspis the curvature by being confined to a narrow band at the edges is too extreme to maximize this form of protection and the robust rim would seem to negate its necessity. That the aspis is thinner at its center than at the edges differs from other shield types and indicates that protection from penetration was not the only factor in its design.

The uniqueness of this shield is apparent if we analyze it from a structural engineer’s perspective. A domed profile can support great weight without collapsing and indicates a load-bearing function. The dome of the aspis is shallow and less efficient, so increased force on the flattened surface will cause the rim will be thrust outward. Trusses or supports that resist this outward force are needed to keep the shallow dome from popping inside out like an umbrella in the wind. In the aspis, this was accomplished by its thickened rim. The off-set design presents the maximum thickness of wood against the force attempting to push the rim outwards. A bronze sheath adds to this as the tensile strength of the metal resists stretching. On some depictions, narrow metal reinforcements are seen around the inner surface. These add little protection, but would aid greatly in supporting the integrity of the load-bearing dome. The enigmatic rope that runs around the inner surface and forms the antilabe may have had its origins as a cable truss to add additional support to the inner face of the dome. The Greeks used such cable trusses, upozwmata, to keep their ships from bending up in the center, or "hogging," from pressure on the hull.

My article on hoplite combat

This is the article I wrote for Ancient Warfare magazine that spawned this blog, the issue is now out of print so I have decided to link to it. It presents a new model of Othismos and hoplite warfare which differs greatly from previous attempts at reconstruction.

http://www.hostalis.net/documents/theaspis.zip

You will see that most of the posts I have made here have been groundwork for this model. I will go into greater detail in future posts.

Back to domed shields

Let's review a bit. The aspis was a dome shaped shield. Domes are supremely suited to transferring force applied to their face around to their base. On a micro-scale this explains why pieces of armor like the Napoleonic cuirasse at right are curved. Because some portion of force applied to a curved face is transferred to the area around it, a curved plate makes stronger armor than a flat plate.

A blow to a curved plate will compress the metal around the site of the impact as in figure A. An impact on a flat or bowl shaped plate, as in B, will stretch the adjoining metal rather than compressing it. Thus only the tensile strength of the metal resists the impact. Compressive strength is always higher in metals than tensile strength.

So on a small scale we can see that curvature helps resist impact. This is probably not a driving force behind the evolution of the aspis because the front of the dome, precisely where we would want this property maximized is flattened.

Once again we must seek a reason for the flattened dome shape.

The aspis rim did not evolve as a shoulder rest

There was a comment to my last post from a fellow I know and who's opinion I respect. Rather than address it as a comment, I'm going to expand it into a post. He stated:

"One criticism: you leave out the obvious advantage of supporting a shield on your forearm and shoulder or neck- that it reduces the strain on the arm. Clearly its much harder to support a 15 lb shield for an hour at the end of your arm than it is to support it close to your body and resting on your shoulder. The Roman scutum shows that this isn't an insurmountable problem,
but it is significant."

This is an advantage. I need to address it specifically because resting the shield on the shoulder has has been put forth by the pro-othismos (pushing match) authors to partially explain why the shield has its characteristic concave shape. Once I am finished presenting my new model for othismos in the next few posts, I will have to deconstruct the current, incorrect pro-othismos arguments. The development of the double-grip as a means of bearing the weight of the shield I will adress in coming posts, because it is tied to the shield's true function, but I'll discuss the idea of the rim as a shoulder-rest.

The apis can be rested on the shoulder. The logic behind citing this as a reason for the concavity and rim is as follows. The aspis is heavy and held in a double-grip that does not allow it to be put down easily, so a rim was added to allow it to be rested on the shoulder. There are problems with this logical progression.

The aspis is heavy, but much of the weight is added by the thickest part of the shield, which is the thickened shoulder section near the edge and the off-set rim. Thus the means of carrying the weight is responsible for a good portion of it.

Throughout Greek history we see many heavy shields. Large Mycenaean shields appear to have aided in carrying this weight by the use of a shoulder-strap or telamon. The smaller, but still robust shields of the later Macedonians is also equipped with a shoulder-strap. Were bearing the weight of a heavy shield the only concern, the use of a simple strap makes far more sense than an unique, heavy shoulder rest rim structure. Resting the shield on the shoulder is surely something hoplites did to bear its weight, but this function arose as a consequence of the heavy, offset rim, not as its source.

As an analogy I put forward the wearing of corinthian helmets pushed back off the face and held in place by its cheek-pieces. In high school we used to wear our football helmets in this way whenever possible because, even with modern ventilation systems, the difference in heat is enormous. No one would suggest that the cheek-pieces evolved in order to be worn in this manner, the protective function to the face and neck is obvious. It is only the fact that the benefit of the aspis' unique shape is not obvious that allows such conjecture.

As an aside there is an excellent reminder of the power of fashion in the design of military equipment. In Italy, magna graecia, we see a type of helmet known as the Italo-corinthian that is designed to look like a corinthian helm worn in this way, but cannot be worn pulled down. There is no functional reason for this, the faux eye-slits surely weaken the structure, but the force of fashion is strong.