Ancient History & Civilisation


Large Weapons, Small Greeks: The Practical Limitations of Hoplite Weapons and Equipment

Physical Characteristics of the Hoplite Shield

During the entire period when hoplites held sway over Greek land warfare, they were defined above all in terms of their primary offensive and defensive weapons, namely, the spear and the peculiarly characteristic shield; indeed, it is nearly impossible to conceive of the idea “hoplite” without these. Of these weapons, it was first and foremost the hoplite’s shield that was his defining characteristic; and it was this shield that effectively set him apart from any other troop type in the Greek world. Moreover, whereas all other items in the hoplite’s equipment were subject to differing degrees of change and development over time, the shield and the spear were the only items to remain essentially unaltered throughout the entire hoplite era. The fact that they did not undergo any larger-scale structural change is significant: evidently their design was eminently suited to their purpose right from the outset, and it continued to be so successful that it needed few or no adjustments later on.

Consequently, attempting an assessment of the shield’s measurements, weight, and handling characteristics is of crucial importance for an understanding of what could be done with hoplite armor and weapons. What knowledge can be obtained about hoplite shields is largely derived from two types of sources. First, although shields, unlike other arms and armor, were invariably made chiefly of perishable materials and thus have largely vanished, a few have been at least partly preserved. Among these is the famous Spartan shield captured by the Athenians during the fighting at Pylos in 425; an Etruscan shield of the hoplite type found at Bomarzo in Italy; the Basle shield, found in Sicily; and a bronze shield facing, recovered at Carchemish on the Euphrates.1 To this should be added numerous bronze shield covers excavated at Olympia.2 Second, there are many representations of shields in iconography—representations that are frequently quite detailed and revealing, and which therefore allow a great measure of accuracy in determining shield measurements.

Based on this, it may be laid down that the shield was circular, noticeably concave, and on average 90 cm to 1 m across. These characteristics are also particularly evident from the large amount of vase paintings of hoplites holding their shields in different ways, offering good views of the weapon from all angles.3

The shield core was invariably made of wood,4 and while almost all original shield cores have therefore long since disintegrated, there are a few archaeological shield finds with some wood still preserved. This is the case with the Bomarzo shield, inside which were found remains of wood identified as poplar, as well as with the Basle shield, which had a core made of willow.5 The fact that this type of wood was especially suitable for shields is corroborated by Pliny the Elder, in whose Naturalis Historia both poplar and willow, and in fact all hardwoods in the group aquatica, are described as the most suitable wood for shields (scutis faciendis aptissima), because they are not only tough but also comparatively pliable, and thus very resistant to breaking.6 Several Greek sources, however, indicate that willow (itea, also reckoned among the aquatica) was in fact the normal material.7 Its characteristics lent the wood durability and resistance against penetration while at the same time allowing the wood to contract somewhat in the event of a penetration (plaga contrahit se protinus cluditque suum vulnus), and so helped minimize damage from edged weapons sustained during combat. In addition to these characteristics, poplar is also suitable for shield making because of its very light weight compared with, for example, oak or ash.8

The shield core was made by fastening wooden laths to the rim, running from side to side. The laths would be joined to each other by a system of grooves and tongues that were glued together in order to achieve a maximum of structural stability.9 According to Adolf Rieth, some shield cores may even have been made from several very thin layers of laths, with each new layer running at right angles to the preceding one.10 It is obvious that this manufacturing process would have yielded a gain in resilience: a modern parallel is ordinary plywood, which is extremely durable and nearly impossible to break. While it is possible that the views offered of the inward-facing surfaces of shields on the Chigi olpe—those held by the advancing phalanx on the left-hand side—reveal just such a manufacturing technique, it cannot have been ubiquitous, as the Bomarzo shield had only one layer of wood. In this case, then, it was obviously advantageous to affix the handle grips in a way that ensured that, with the shield held correctly, the grain ranhorizontally across the shield, offering much better flexibility in case of simultaneous pressure on both sides of the shield (rather than simultaneous pressure on the “top” and “bottom” of the shield, an unlikely scenario). This measure would have increased the shield’s combat effectiveness in no small degree.11

Often the shield was covered by a very thin bronze sheathing (chalkōma), which afforded extra protection against the wood splintering, and helped dissipate the force of blows over a larger area.12 Even in its current state, the Bomarzo shield offers an excellent example of both the thinness of the bronze facing and the remarkable concavity of hoplite shields: “The bronze cover, which is about 0.5 mm thick, forms a shallow bowl about 10 cm deep and between 81.5 cm and 82 cm in diameter, including a rim which projects about 4.5 cm from the wall of the bowl all round.”13Nevertheless, many vase images of hoplite shields suggest that shield concavity was frequently even more pronounced. The inner surface of the shield was usually padded with a glued-on layer of leather, which was likely mounted in order to decrease wear and tear on the wooden core, and to prevent the wood snagging or pinching the bearer’s arm. By Henry Blyth’s estimate, however, the leather layer inside the Bomarzo shield is too thin to have afforded any real protection from penetration, and is therefore chiefly decorative.14

The bronze shield facing was apparently optional, but whether the shield had one or not, the shield rim (itys) was invariably reinforced with a bronze band to protect the vulnerable edges. The itys was thin enough to be wrapped around the shield edge itself, and there is at least one case of an ityscovering the shield rim halfway.15 This reinforcement was indispensable on any hoplite shield, and the reason is obvious: if the edge were insufficiently protected, it would be very vulnerable in combat. The joints of the laths would be open to blows, especially lateral sword cuts, which could separate the laths from each other and thus cause the shield to disintegrate.

Emil Kunze and Hans Schleif ‘s listings of the numerous hoplite shields dedicated at the sanctuary in Olympia show diameters between extremes of 80 cm and 100 cm (most between 90 and 100 cm); and the famous Spartan shield captured at Pylos in 425, somewhat bent out of shape, measures 95 × 83 cm. Interestingly, Rieth mentions a shield that is in other respects similar to the one examined, but which measures no less than 125 cm in diameter.16 The large diameter gave the shield a considerable surface area: between 6,362 and 7,853 cm2, according as the diameter was 90 or 100 cm.17

That hoplite shields were of a considerable size seems also to be borne out by a passage in Arrian’s Anabasis. Here, the garrison of Miletos, consisting of Milesians and mercenaries, panicked and abandoned their positions when the city was stormed by Alexander’s invading Macedonian troops in 334:

Then some of the Milesians and the mercenaries, attacked on all sides by the Macedonians, threw themselves into the sea. On their own inverted shields, they then paddled across to a small, nameless island near the city. Others boarded small vessels and hurried to get past the Macedonian triremes but were caught by them at the harbour entrance. The majority, however, perished in the city itself.18

The word aspis in itself of course does not necessarily betoken a hoplite shield; but the fact that the shields are inverted (ὑπτίων) for the purpose of flotation seems to corroborate this. If these were hoplite shields, they naturally had a wooden core; but what provided the buoyancy in this case was rather the principle of displacement. It is reasonable to assume that the soldiers in this case, as so often, had discarded most of their weapons and equipment, and so weighed considerably less. However, when Alexander turned his attention to the three hundred survivors after taking Miletos itself, he realized that they were ready to fight to the death (διακινδυνέυειν), a fact that prompted him to spare their lives. This seems to indicate that they may after all have kept their offensive weapons as well. Whether the soldiers kept their weapons or not, we should probably estimate that each shield-boat in this situation had to keep some 60–70 kg afloat on average.19

If these events actually took place as described by Arrian, a considerable displacement (and, consequently, shield surface area) was therefore necessary. Let us look at the physics involved. The density of the shield in question is extremely difficult to ascertain, since the amount of metal in addition to the wood is unknown, but it may be assumed that the overall density is less than 1. The density of saltwater is slightly higher than that of freshwater: a 5 percent saltwater solution has a density of 1.034, which is probably not too different from seawater. It is difficult to say with any exactitude without precise knowledge of the concentration of salt, however, so water density will not be factored in. At any rate, the actual buoyancy would not have been less than the result of the calculation. Apart from the shield’s own buoyancy, the amount of air contained within the shield produces a buoyancy equal to the amount of water it displaces without the shield edges submerging. Therefore, if the shield contains 10 l of air, it can displace 10 l of water, weighing 10 kg, before becoming submerged and sinking.

Now, the volume of the relevant geometrical figure, a spherical cap, is determined as follows:


, in which r is the radius of the base of the cap and h is the height (or depth) of the cap. This yields, for shields of the following dimensions, the volumes shown in table 8-1.

Apart from the shield’s own buoyancy and the possibly increased density of the seawater in question, the shield will float approximately the same amount in kg as the amount of liters of air contained. It may be seen, moreover, that small changes in height and radius yield rather greater volumes. It is reasonable to surmise that the hoplites in question had jettisoned anything but perhaps their spears and swords; and assuming an average body weight between 62 and 65 kg for an adult male,20 it is apparent that the dimensions of the Bomarzo shield are insufficient for it to float a man. It is also evident, however, that the carrying capacity greatly increases with slight increments in dimensions; and on this assumption it does not seem unrealistic for the Milesian garrison to have paddled to safety on their shields.21

Blyth, in his assessment of the Bomarzo shield, estimates the total weight of all its components at 6.2 kg.22 Of this, the weight of grips, straps, and fittings is assessed at some 0.7 kg, the rest consisting of the wooden core and the bronze sheathing. The weight of the original wood and bronze can of course be no more than an educated guess, because the materials have decayed considerably over the centuries; but considering that the average hoplite shield had a diameter of 90 cm, rather than only 82 cm as in this case, the combined weight of bronze and wood would have been closer to 6.75 kg. Rieth, in his examination of the Bomarzo shield, reckons with a very pronounced decay of the wooden components and consequently a rather greater mass in the original shield. If we follow Rieth’s assessments, then, we arrive at a shield even heavier than any derived from Blyth’s estimates, weighing as much as 7 or 8 kg. And this still only allows for one layer of wood in the construction of the shield core.23 That this weight is probably closer to the norm is brought out by the fact that the shriveled remains alone of the Basle hoplite shield, as examined by David Cahn, weigh as much as 2.95 kg—or nearly half of the estimated full weight of a “new” and intact Bomarzo shield by Blyth’s estimates.24


Shield radius



45 cm

10 cm

32.3 l

45 cm

20 cm

67.9 l

50 cm

20 cm

82.7 l

45 cm

25 cm

87.7 l

50 cm

25 cm

106.3 l

Using the Hoplite Shield

It seems a straightforward enough assumption that the shape and sheer size of such shields would have had consequences for the men bearing them in combat, but the worst problem with the hoplite shield would no doubt have been the considerable weight. On closer inspection, the sources also seem to bear this out.

It is interesting, for example, that merely being ordered to stand still while holding the shield (τὴν ἀσπίδα ἔχων) was considered a sufficient disciplinary punishment in the Spartan army,25 which was not otherwise known for being particularly squeamish when it came to maintaining discipline by means of corporeal punishment.26 The element of physical ordeal in this type of punishment has been downplayed,27 apparently because Xenophon, in whose Hellenika the punishment is mentioned, adds “[which] is regarded by distinguished Spartans as a great disgrace,” almost by way of an afterthought.28 There can be no doubt that kēlis certainly does mean “blemish” or “disgrace”;29 but its mention here does not mean that the disgrace visited upon military miscreants was not intended to be a harsh physical punishment as well. This interpretation is in fact borne out by a comparison with Plutarch’s Aristeides, where the Spartan king Pausanias’ harsh treatment of recalcitrant Spartan allies is discussed: “The commanders of the allies always met with angry harshness at the hands of Pausanias, and the common men he punished with lashings, or by compelling them to stand all day long with an iron anchor on their shoulders.”30

Now, if the punishment that involved standing while holding a shield was merely symbolic (a symbolic value which, one suspects, may not have been immediately appreciable to outsiders), then this case—similar in all respects except for the object actually held—becomes baffiingly pointless. The point was hardly lost, however, on those unfortunate allies who fell afoul of Pausanias and were forced to stand around holding iron anchors all day—even if they were otherwise unaware that this was a particularly shameful brand of dishonor in the Spartan army. The substitute punishment of flogging emphasizes this to an even greater degree: Plutarch evidently sees the anchor punishment as completely parallel to the floggings, the physical nature of which can hardly be denied. Standing τὴν ἀσπίδα ἔχων may very well have been humiliating or dishonorable for the offender (if he was a Spartan, at least), but it is evident that it was also a grueling physical ordeal. Xenophon’s remark that it was a kēlis for the Spartans should be seen in this light: the physical harshness of such a punishment was obvious to any contemporary reader; that it was also considered disgraceful in Sparta was what needed an explanatory remark.31

An amusing passage in Xenophon’s Anabasis points to the same conclusion. During Xenophon’s learning period as commander of the Greek rearguard on the retreat from Artaxerxes’ pursuing Persians, a detachment of hoplites commanded by Xenophon (himself on horseback) “race” Persian troops for possession of a hill commanding the only way forward. One of the hoplites, a certain Soteridas of Sikyon, reacts unfavorably to the strategos’ ill-timed pep talk: “It’s not fair, Xenophon! You are sitting on your horse while I’m wearing myself completely out with carrying this shield.”32 Xenophon, tacitly acknowledging his leadership blunder, jumps down, takes Soteridas’ shield from him, and tries to keep up. He is soon completely exhausted, since he’s also wearing a heavy cavalrythōrax; and the other hoplites therefore bully Soteridas to take back his shield. Whether the anecdote is true or not, it must have been acceptable to contemporary readers, so the vignette demonstrates the hoplite shield’s very real problem of sheer weight.

The Hellenika and Anabasis passages afford a glimpse of how the burden of the hoplite shield was perceived, despite the fact that—or perhaps precisely because—the question of weight in either statement is treated in a rather offhand fashion, as is natural with allusions to everyday experiences. Yet other essential characteristics of the shield would also have contributed to making it difficult to use in actual combat. The ungainly shape and the sheer size of the shield made a significant contribution to a quite exceptional awkwardness in handling. As we saw above, the shield’s mass was spread fairly evenly over a surface area of approximately 6,400 cm2 at a minimum, but could easily reach an area of as much as 7,800 cm2. Now, the greater its surface area, the more inertia and thus difficulty in handling any physical object; and one that is the size of a small bridge table is thoroughly unwieldy, no matter how it is held.

Despite these considerable drawbacks, it has frequently been claimed that the hoplite shield is no less apt for single combat than other shield types;33 but this claim is too optimistic. The shield seemingly afforded good protection against any kind of edged weapon brought to bear against it, but that protection came at a price. Its size, shape and weight were enough to require some sort of alleviation, and this need in turn dictated the well-known combination of the double grip and holding posture peculiar to this shield type, entailing a series of consequences for close combat. When in use, the hoplite shield—uniquely—was supported on no fewer than three points: the elbow (the porpax), the wrist (the antilabē), and, by means of using the unique “lip” of the shield edge to hang or rest its weight there, the shoulder.34 The left-side-forward posture not only seems natural to assume in combat; it also greatly relieves the strain on the arm and shoulder by taking advantage of all three support points. Furthermore, when the rim is rested upon the bearer’s shoulder, the shield is carried aslant, its lower rim jutting out before the hoplite. This has the additional advantage of enlarging the zone of protection considerably. The inclination of the shield would also have served to make spear and sword thrusts glance off the shield, although, as is in evidence from iconography, the upper edge of a shield supported on the bearer’s shoulder was also directly under his chin if he was adopting a sideways-on stance. This meant that a thrust delivered to the top half of his shield might glance off the polished surface and straight into his face or throat, just as anything that jolted the shield forcefully upward at the lower edge, such as a swift kick, must have directed the upper edge in the same direction. Nevertheless, vase images clearly bear out that this was in fact the normal grip and defensive stance with a hoplite shield; and people who have actually worn replicas of hoplite armor have assured me that this way of handling the shield is not only the logical but indeed the only possible way.35

Furthermore, the hoplite shield had certain design drawbacks compared with other, lighter shield types having a single central grip. The double-grip system dictated that the shield could be held with the left arm only, whereas a single-grip shield could easily be shifted from one hand to the other to ease the strain on arm and shoulder. The hoplite shield for this reason generated even greater strain on the left arm, and supporting it on the shoulder was an absolute necessity, not simply a convenience. For the shield to afford sufficient protection, it must also be held as far away from the body as possible, and aslant at an angle of approximately 45°. This increased the angle of deflection and kept penetrating weapons farther away from the bearer’s body, but it also increased the strain on the left arm considerably.

However, an even more serious drawback was the fact, to my knowledge hitherto overlooked in scholarship, that in order to use a double-grip shield properly, it can only be held out at half an arm’s length, since the forearm must of necessity be bent and held at right angles to the upper arm. The zone of protection therefore begins already at the elbow, and cannot be extended beyond it. The frontal range is thus drastically reduced. A single-grip shield, on the other hand, can be held out at a full arm’s length, or about twice as far from the body as the hoplite shield. This is important because it means that a hoplite shield’s surface area must necessarily be much larger than a single-grip shield needs to be: the single-grip shield, being held at a full arm’s length from the body, can afford to be much smaller while still offering the same degree of protection. It decreases the adversary’s angle of attack just as effectively as the much larger hoplite shield does, simply by being held at twice the distance from the body. Furthermore, by merely turning the wrist, the single-grip shield can be rotated to maximize the angle of deflection, even when the bearer reaches across to his own right side. Thus, incoming attacks can be countered earlier and perhaps “nipped in the bud” by merely parrying with the shield. That these exact advantages of a smaller, single-grip shield were well understood in antiquity is demonstrated by Diodoros, who comments approvingly on the Iberians’ use of such targes.36

A hoplite shield could of course also be moved about to a certain degree, but, owing to its awkward size and shape, not exactly briskly; and if it was moved about actively to parry or block incoming blows and thrusts, this was carried out with the shield very near the body: there was little time and room for secondary measures if a parry came too late or was misdirected.

Weight, shape, and size together thus made a hoplite shield very awkward to “wield”; and it is open to serious doubt whether anyone, no matter how strong or how well trained, was able to sustain its weight, let alone wield it, for any considerable amount of time during combat.

Comparison with a Modern Combat Shield

A major problem facing scholars trying to assess the combat aptitude of ancient weapons is naturally the scarcity of possibilities to try to handle them, let alone under anything resembling actual fighting conditions. Accordingly, the next best thing would be if it were possible to obtain this much-needed information from somebody who has actually tried using similar items.

Police forces around the world have regularly used shields against rioters throwing stones, bottles, or even Molotov cocktails; and Danish riot police have often seen action, particularly against squatters in the 1980s, but on many other occasions as well. The police are among the very few today who have any experience with handling a shield in combat, and the theory and practice of shield fighting employed by them is therefore very relevant in a discussion of what can and cannot be done with a shield. For this reason, I contacted the riot squad section of the Danish police academy in Copenhagen. The following is the distilled result of a long interview I conducted with Chief Inspector Claus Olsen of the Danish police, who supervises the combat training section and has taught riot control for many years, including the use of double-grip shields in phalanx-like formations.

Danish police riot control forces regularly used shields from the 1970s until recently, when they were almost completely abandoned in favor of more mobile and offensive tactics. The shield in question is rectangular with rounded corners and made of Plexiglas, and so its shape is possibly more reminiscent of a Roman legionary scutum. It is fitted with a double-grip carrying system that allows for ambidexterity, placed in the middle. The grips are affixed at approximately 45° to the vertical edge, so that the arm is inserted at an oblique angle. The shield measures 95 × 60 cm, or 5,700 cm2 and as such is roughly comparable with the surface area of a hoplite shield, but the weight is nonetheless kept down to a mere 2.74 kg. Despite the shield’s weighing no more than between 34 and 39 percent of a hoplite shield, however, it was considered a weapon suitable only for defensive fighting. Policemen would typically form defensive lines (termed “chains”), and stand so close that the edges of their shields actually touched. They then advanced to the combat zone and kept their position. The defensive character of these formations was underlined by the fact that policemen in combat gear would also be equipped with visored helmets, greaves, bulletproof vests, and thick, padded gloves.

According to Chief Inspector Olsen, the shield was deemed too heavy, large, and awkward to be wielded freely, and to be put to offensive use—so much so, in fact, that a provisional concept was devised for offensive action. The stationary shield line might under certain circumstances be supported by hastily summoned plainclothes policemen, who would be equipped only with modified standard shields. The modified shield is identical to the normal type, but is simply sawn off just above the middle near where the grips are affixed, so that a little less than half the shield remains. Much like a buckler, this lighter shield could be swung around with comparative ease; and unlike the large shield, the adapted version could therefore be used offensively, combined with a lack of body armor to ensure crucial mobility. These policemen, cowering behind the wall of shields held by the front line in full combat gear, would then be able to move around the chain, dart forward, and close with rioters who had ventured too close to the line. It should be noted, however, that this was a stopgap measure intended to enable police to arrest the most aggressive individuals, since police generally had no interest in actually clashing with the rioters but rather aimed at containing them and driving them away from crowded spaces, thereby protecting the public and property.

The stationary, defensive police line could thus benefit from the unarmored, lighter troops, who could prevent aggressive missile-throwing rioters from coming close to their position with impunity—something that was otherwise a possibility. In other words: policemen with shields, and in combat gear, were considered unable to fight hand-to-hand, whereas they were extremely well suited to braving barrages of thrown cobblestones and bottles. Individually, however, they could do little more than that, and the practice of leaving the line to pursue rioters was discouraged for two reasons: first, this threatened to disrupt the shield wall and endanger the entire position; second, although well protected, policemen were unfit for single combat because of the large, heavy, and unwieldy shield.

It may of course be objected that Danish police shields are rectangular and as such not comparable to hoplite shields; but this is immaterial, since riot squads of other countries’ police forces operate with riot shields of other shapes, among them circular, as I have witnessed myself in Greece.37Clearly, then, it is possible to make a satisfactory shield line, providing sufficient shelter for the members, with round shields.

The standard police shield, deemed too heavy and clumsy by well-trained and physically fit riot squad policemen, weighed not much more than a third of a typical hoplite shield. It seems unlikely that hoplites in bronze armor would have been able to do what fit and trained policemen cannot, or at least deem hopeless—namely, fight as duelists in serial monomachiai, wielding their three times heavier shields with ease against attacks from all corners.


Another factor worth considering is the physical characteristics of the men actually wearing the armor. When theories are put forward about what could and could not be done while wearing hoplite armor, frequently based on assessments of what “adult males” are capable of, the tacit assumption must eo ipso be that Greek men of antiquity are immediately comparable to modern Western men.38 However, whereas the weight and measurements of the surviving specimens of armor, and to a certain degree weapons, are naturally constants (leaving aside for the moment the effects of oxidation, corrosion, weathering, and other decay of the materials, and the corresponding compensation estimates made), there is no guarantee that the individuals who had to wear and use them were physiologically similar to modern men. The questions concerning bearer physiology are thus of crucial importance for understanding the “relative” weight of hoplite weapons and armor; yet the problem of this relation has seldom, if ever, been addressed. It is therefore well worth examining the available data supplied by skeletal remains from the Archaic and Classical periods.

John Lawrence Angel, who in 1945 examined skeletal remains exhumed in Attica, put the average height of the Greek male in antiquity at no more than 162.2 cm, and of the female at 153.3 cm. It should be pointed out, however, that these data accrue from a rather scanty sample material: 61 male and 43 female skeletons from Attica, as against a total of 225 datable males and 132 females in all of Greece proper.39 Similar results accrue from Angel’s 1944 analysis of all ancient Greek skeletal remains known at the time: here, the result is given as 162.19 cm for males, with a range between extremes of 148 and 175 cm. The result for females overall remains the same.40 Angel, whose interest was primarily “racial” analysis, lists crania from Attica, Boiotia, Corinthia, and Macedonia; but unfortunately he does not indicate the distribution of more complete skeletons, which may have formed the basis for the calculations.41 Nevertheless it must be assumed that the average measurements actually represent the average, geographically as well as chronologically.42

The comparatively scanty material notwithstanding, we would be well advised to keep in mind that, in the words of Lin Foxhall and Hamish Forbes, “this sample may be biased in favour of higher socio-economic groups since it is the graves of the comparatively wealthy that are most likely to receive attention from archaeologists.”43 If this is accepted, it follows that the average Greek male was in fact likely less well nourished, and the skeletons examined by Angel may well belong in the absolute upper percentile.44 Walter Donlan and James Thompson give the average height of Greek males in the Classical period as approximately 170 cm, with body weight between 65 and 67 kg.45 Unfortunately, however, their article gives no information about how these results were arrived at; so Angel’s data must assume priority.

The modern European or American adult male, on the other hand, measures approximately 179 cm on average.46 Determination of body weight is more complicated, as it depends to a large extent on a wide variety of other factors (age being but one); but the average weight of modern males between the ages of twenty and sixty is 80.97 kg.47 It is therefore certain that Greek men in antiquity were shorter than Western men today; and it is also highly likely that they were noticeably lighter, considering a diet consisting largely of cereals and pulse and to a certain extent vegetables.48

These extrapolations from a rather exiguous sample material have recently been confirmed in no small degree by the interesting findings from the survey of the expansive territory of Metaponto (ancient Metapontion) in southern Italy.49 In the course of this grand-scale survey, field examinations of the remarkably well-preserved necropolises at Pantanello, Saldone, and Sant’Angelo Vecchio were carried out.50 Here, Maciej and Renata Henneberg examined the skeletal remains of 272 individuals, 251 of which were excavated at Pantanello alone.51 The Hennebergs’ work on these and other necropolises of Magna Graecia now comprises approximately 1,000 individuals, easily “the largest and most comprehensive study of the mortal remains of a population in the Greek world.”52 In addition, while the human remains excavated from the necropolis spanned a period of several centuries (from the sixth to the third centuries BC), the majority could be dated to well within the hoplite era.53

Metapontion was a largely rural settlement, focused primarily on agriculture, and with a population who were very likely predominantly farmers.54 Accordingly, its inhabitants would have been very similar to the average rural population in the rest of the Greek world with respect to—among other things—nourishment, growth, build, overall health, life expectancy, and general physiology.

Now, it is scarcely unreasonable to assume that, on average, more than 50 percent of the population in the Greek world as a whole were occupied with agriculture, arboriculture, and the production of foodstuffs in some capacity, if possibly sometimes in other activities than farming as such.55 If that is the case, most Greek hoplite armies, drawn from the male citizenry, would likely have displayed the same ratio of, at a minimum, 50 percent farmers (or, at the least, “agriculturalists”). The implications of this should be clear: the Metapontine necropolis not only furnishes a representative selection of a typical population occupied primarily in agriculture, the backbone of any hoplite citizen army who likely possessed the physical qualities most wanted in hoplites—toughness, stamina, strength, and resilience—but the amount of material in question is comprehensive enough to be statistically significant.56 The examined data revealed that the average height of adult males was between 162 and 165 cm, that of females between 153 and 156 cm (estimates vary according to the applied method of reconstruction), and with a body weight of approximately 60–65 kg for males and 50–55 kg for females: in other words, the findings of earlier examinations were soundly confirmed in this respect.57 Given this uncharacteristically ample, significant sample material, as well as its thoroughly agricultural setting, the Metapontion necropolis furnishes an excellent opportunity for assessing the physical characteristics of the average Greek hoplite.58

Greek men in antiquity—including Greek hoplites—were thus significantly smaller than modern Western men. For this reason alone, what may perhaps seem comparatively light or small to us as moderns may in fact have been considerably harder and more cumbersome for smaller men to bear. Offensive, and especially defensive, weapons and armor would have been even more uncomfortable, heavy, and unwieldy to men frequently more than 15 cm shorter than the modern Western average. The shield, above all, some 90 cm in diameter, would have been even larger for such men, normally no taller than 165 cm: in most cases, the shield’s diameter measured considerably more than half of the total body height. In fact, even the lightest of Greek panoplies would have been a much heavier burden to bear when compared to the physical norm of the average modern Western male. All this serves to underline the fact that hoplite weapons and armor were not, by any stretch of the imagination, easily or comfortably manipulated or worn.

It should also be kept in mind that polis armies were normally composed of citizens of all ages between eighteen and sixty.59 Owing to this tremendously long obligation to perform military service, there must have been a great many older men in the phalanxes, perhaps even a majority.60 In the words of Victor Davis Hanson, “after all, thirty of forty-two age classes liable to military service were composed of men over thirty years of age.”61 While this is a priori true, it should also be taken into consideration that these age groups would have been exponentially depleted of members because of the increasing mortality resulting from both natural causes and participation in more campaigns and battles. There is no shortage of sources attesting to older men taking their equal share of the grisly work in the rank and file;62 and all other things being equal, the burden of arms and armor must have been a great deal harder to shoulder for men pushing sixty.

The sources to a large degree bear this out. Personal servants (hypaspistai) frequently carried the hoplites’ weapons63 and supplies;64 at least this seems to have been the case with higher-ranking and wealthier persons. Even when hoplites did hold their own shields, and when supposedly ready for instant action, they normally would not pick them up until the last possible moment: the command θέσθαι (τὰ) ὅπλα means to extract the arm from the twin grips and set the shield (and probably spear) down on the ground, leaning against the knees and ready to be picked up again quickly when the order is given. The phrase is unusually common: it is found forty-three times in Herodotos, Thucydides, and Xenophon alone,65 and so well known was this “stand at ease” position that the Athenian general Chabrias, according to later historians, on one occasion could display his contempt for the advancing enemy simply by ordering his mercenaries to remain brazenly in this position as the enemy approached (δέχεσθαι τοὺς πολεμίους καταπε φρονη κότως ἅμα καὶ ἐν τῇ τάξει μένοντας).66 We even find that the Phokians guarding the Anopaia pass at Thermopylae in 480—who, it must be supposed, should have been on maximum alert—only picked up their weapons after they saw the Persians approaching, surprising the advancing enemy with that strange sight: “Leaping to their feet, the Phokians were in the act of arming themselves when the enemy were upon them. The Persians were surprised at the sight of troops preparing to fight.”67 In fact, Herodotos’ choice of verb, ἐνέδυον (“put on,” rather than the expected ἀνέλαβον), implies that the Phokian hoplites did not even put on their body armor until the last possible moment—even in a “red alert” situation like this.68

In Euripides’ tragedy Herakleidai (datable to c. 430)69 the weight of a “full set of armour” (ὅπλων παντευχίαν) also seems to be a consideration. The old Iolaos, about to join battle, is advised by his servant to put his armor on in a hurry, since battle is near. All the same, the servant adds, “However, if you dread the weight of it, / go unarmed for the present, and when you reach the ranks / put all this on there. Meanwhile I’ll carry it.” With palpable relief, Iolaos quickly accepts (καλῶς ἔλεξας).70


The weight and completeness of armor was, if anything, reduced over the centuries, either by perfecting the metalwork techniques, by replacing bronze cuirasses with corselets made of other materials, or simply by increasingly discarding items that were apparently no longer needed. It is interesting to note that the full set of hoplite weapons and equipment was at its heaviest and most cumbersome in its earlier stages of development; and it is equally interesting that Euripides suggests that even by 430—at a time when most defensive body armor was well on its way to becoming obsolete71—a set of armor was still considered a significant burden (albeit, in this case, to an elderly man). Reconstructions based on the surviving remains of shields suggest that they were in fact very heavy, as much as 8 kg not being an unrealistic assessment for a combat shield some 90 cm in diameter. Contemporary literary sources testify to the considerable size and perceived burden of hoplite shields, an aspect that was exacerbated by the shield’s ungainly shape and the double-grip system, which, although securing good support for the bearer, also drastically reduced his range with the shield when compared with a lighter, single-grip shield. A comparison with probably the only modern use of shields “in anger” reveals that a double-grip shield of roughly comparable size was deemed too heavy and awkward to be used for any sort of soloist fighting, despite the fact that the police shield in question weighs a little more than a third of a hoplite shield.

The tacit but widespread assumption that Greeks in antiquity were physically immediately comparable to ourselves is refuted by analyses of skeletal remains from the Archaic and Classical periods, revealing that the average height of Greek males was between 162 and 165 cm compared with the 179 cm of modern Western males. Obviously the burden of weapons, and especially shield and armor, would have been even greater for such smaller men. Exacerbating this even further is the fact that quite a few old men must have been present in the average phalanx, since men were required to serve until they were sixty years old.

All this serves to underline the importance of assessing the physical characteristics of the weapons themselves in the debate on how hoplites and phalanxes functioned in combat. Furthermore, it seems to me that the evidence, such as it is, points in the direction of a defensive, closed-order system intimately connected with and based on the weapons themselves. The evidence thus suggests that the debate on hoplite fighting, unlike the hoplite phalanx itself, is not closed at all.


I wish to thank Curtis Eastin, Gregory Viggiano, and Donald Kagan warmly for inviting me to participate in the Origins of the Greek Phalanx conference at Yale in April 2008. It should also be pointed out that, owing to circumstances entirely beyond my control, I found myself with a matter of mere days in which to prepare a paper for the conference. It, and consequently this article, therefore consisted largely of material drawn from my (then forthcoming) monograph, Reinstating the Hoplite (2009). However, I have endeavored, as far as at all possible, to rearrange it and add fresh material and new data for this chapter.

    1. Blyth (1982) 9, 13–14; Rieth (1964) 104–5; Cahn (1989) 15–16; Shear (1937) 347; Boardman (1980) 75; Millard (1994) 288–89.

    2. Kunze and Schleif (1942) 70–93; Mallwitz and Herrmann (1980) 106.

    3. The concavity of the shield is apparent also from a number of literary sources: Hdt. 4.200.2–3 (in which an inverted shield is used as a stethoscope to listen for enemy sapping underground, a procedure also recommended in Aen. Tact. 37.6–7); Thuc. 7.82.3 (where four upturned [ὑπτίας] shields serve as vessels for coins confiscated from Athenian prisoners of war); Xen. Hell. 5.4.17–18 (where hoplites weigh their shields down with stones to prevent them from blowing away during a storm on a mountain); Plut. Mor. 241f 16 (where the shield [again, presumably, inverted] is thought of as a possible stretcher, in the famous anecdote of a Spartan mother saying to her son “ἢ τὰν ἢ ἐπὶ τᾶς”); cf. S Thuc. 2.39.1; Stob. Flor. 3.7.30 and Hammond [1979–80]). Several notable landmarks were called “the Aspis,” most famously perhaps a steep hill or slope near Argos with difficult access (ὀχυρὸς τόπος): “[Kleomenes] led his army by night up to the walls, occupied the region about the Aspis overlooking the theatre, a region which was rugged and hard to come at, and so terrified the inhabitants that not a man of them thought of defence” ([Κλεομένης] νυκτὸς πρὸς τὰ τείχη ἦγε τὸ στράτευμα, καὶ τὸν περὶ τὴν ᾿Ασπίδα τόπον καταλαβών ὑπὲρ τοῦ θεάτρου χαλεπὸν ὄντα καὶ δυσπρόσοδον οὕτως τοὺς ἀνθρώπους ἐξέπληξεν ὥστε μηδένα τράπεσθαι πρὸς ἀλκήν [trans. Perrin]): Plut. Cleom. 17.4–5, cf. 21.3, Pyrrh. 32.1–4.

    4. Blyth (1982) 9–12. A wooden core is also strongly suggested by Brasidas’ shield, which, when dropped from a ship, drifted ashore (Thuc. 4.12.1; cf. Diod. Sic. 12.62.4).

    5. Blyth (1982) 9, 13–14; Rieth (1964) 104–5; Cahn (1989) 15–16. There are adequate photographs in both Blyth’s and Rieth’s articles; but the best illustration of the Bomarzo shield remains Connolly’s drawing (Connolly [1998] 53).

    6. Plin. NH 16.209: “The trees that have the coldest wood of all are all that grow in water; but the most flexible, and consequently the most suitable for making shields, are those in which an incision draws together at once and closes up its own wound, and which consequently is more obstinate in allowing steel to penetrate; this class contains the vine, agnus castus, willow, lime, birch, elder, and both kinds of poplar” (Frigidissima quaecumque aquatica, lentissima autem et ideo scutis faciendis aptissima quorum plaga contrahit se protinus cluditque suum vulnus et ob id contumacius tramittit ferrum, in quo sunt genere ficus, vitex, salix, tilia, betulla, sabucus, populus utraque [trans. Rackham]); cf. Franz (2002) 128–29.

    7. Eur. Heracl. 375–76, Supp. 694–96; Cyc. 5–8, Tro. 1192–93; Ar. fr. 65. All these are metonymical uses where itea is simply used for aspis.

    8. See, e.g., (at table 1; accessed 02.09.2011).

    9. Blyth (1982) 9–13.

  10. Rieth (1964) 108 (citing Robinson who in fact only says “crossing pieces of wood,” however: Robinson [1941] 444).

  11. The situation, as illustrated by Blyth, can easily be imagined: if the shield bearer charged (or was charged by) two enemies at the same time, the shield might as well be caught between their shield edges as hit either frontally (Blyth [1982] 17 fig. 6).

  12. Hdt. 4.200.2–3; Xen. Lac. Pol. 11.3; Aen. Tact. 37.6–7; Polyaen. 1.45.2, 7.8.1; Cartledge (1977) 12–13. See, however, Snodgrass (1964a) 63–64. The Bomarzo shield’s remains of wood and leather were found inside its chalkōma: Blyth (1982) 1, 12; Rieth (1964) 101, 106.

  13. Blyth (1982) 5–6.

  14. Blyth (1982) 12; Cahn (1989) 16; Robinson (1941) 444. Another suggestion is that the leather increased comfort for the bearer, both when the shield was hung on the shoulder and when he pressed against it and cowered inside it during combat.

  15. Rieth (1964) 108. Rieth also mentions the finding of a shield at Olynthos: remains of wood, circumscribed by a bronze itys.

  16. Kunze and Schleif (1938) 70–74; Shear (1937) 347; Rieth (1964) 101. Cf. Paus. 1.15.4.

  17. If r = 45 cm, then A = p × 452 = 6,362 cm2.

  18. Arr. An. 1.19.4: ἔνθα οἱ Μιλήσιοί τε καὶ οἱ μισθοφόροι πανταχόθεν ἤδη προσκειμένων σφίσι τῶν Μακεδόνων οἱ μὲν αὐτῶν ῥιπτοῦντες σφᾶς ἐν τῇ θαλάσσῃ ἐπὶ τῶν ἀσπίδων ὑπτίων ἐς νησῖδά τινα ἀνώνυμον τῇ πόλει ἐπικειμένην διενήχοντο, οἱ δὲ ἐς κελήτια ἐμβαίνοντες καὶ ἐπειγόμενοι ὑποφθάσαι τὰς τριήρεις τῶν Μακεδόνων ἐγκατε λήφ θησαν ἐν τῷ στόματι τοῦ λιμένος πρὸς τῶν τριήρων·οἱ δὲ πολλοὶ ἐν αὐτῇ τῇ πόλει ἀπώ λοντο.

  19. I am very grateful to John R. Hale for pointing out the significance of this passage to me.

  20. See the “Physiology” section below.

  21. I am indebted to Christian Tortzen, Sebastian Persson, and Claus Glunk for these calculations.

  22. Blyth (1982) 16.

  23. Rieth (1964) 101. Donlan and Thompson (1976) 341 n. 4 give this measurement as the average weight of the shields in Olympia.

  24. Cahn (1989) 15.

  25. Xen. Hell. 3.1.9.

  26. See especially Hornblower (2002).

  27. Franz (2002) 269–70, criticizing Hanson (2000) 67; but see also Schwertfeger (1982) 263 n. 34: “Daß der schwere Hoplitenschild eine Last war, geht aus dem spartanischen Recht der klassischen Zeit hervor, wo der König im Rahmen seiner Disziplinargewalt gegen ungehorsame Soldaten die Strafe des ‘Stehens mit dem Schild’ verhängen konnte” (that the heavy hoplite shield was a burden is also evident from Spartan law of the Classical period, where the king, as part of his disciplinary measures, could punish disobedient soldiers with ‘standing with the shield’ ”).

  28. Xen. Hell. 3.1.9: ὃ δοκεῖ κη λὶς εἶναι τοῖς σπουδαίοις Λακεδαιμονίων.

  29. Cf. LSJ 9th ed. s.v. κηλίς.

  30. Plut. Arist. 23.2: τοῖς τε γὰρ ἄρχουσι τῶν συμμάχων ἀεὶ μετ’ ὀργῆς ἐνετύγχανε καὶ τραχέως, τούς τε πολλοὺς ἐκόλαζε πληγαῖς, ἢ σιδηρᾶν ἄγκυραν ἐπιτιθεὶς ἠνάγκαζεν. ἑστάναι δι’ ὅλης τῆς ἡμέρας (trans. Perrin, modified).

  31. Greek literature is rife with Spartans either exacting or threatening physical punishment: the sources are collected in Hornblower (2002) 57–60.

  32. Xen. An. 3.4.47–49: oὐκ ἐξ ἴσου, ὦ Ξενοφῶν, ἐσμέν· σὺ μὲν γὰρ ἐφ’ ἵππου ὀχῇ, ἐγὼ δὲ χαλεπῶς κάμνω τὴν ἀσπίδα φέρων (trans. Warner, modified). See also Lendle (1995) 187.

  33. See, e.g., Greenhalgh (1973) 73; Salmon (1977) 85 n.6; Krentz (1985) 60–61; van Wees ed. (2000) 126; Rawlings (2000) 246–49.

  34. See especially Hanson (1991) 68–69, (2000) 68; and cf. Franz (2002) 132; van Wees (2000) 128; (2004) 167–69. Van Wees’s sample of illustrations (figs. 3, 4a, 6, and 10) demonstrate the posture clearly.

  35. I have corresponded with the UK-based Hoplite Association (, a reenactment group whose members have kindly (and patiently) answered my questions. I am also grateful to Nino Luraghi, who, having actually tried on a replica, emphasized to me the impossibility of holding a modern replica of a hoplite shield, except with bent arm and supported on the shoulder.

  36. Diod. Sic. 5.34.5: “The bravest among the Iberians are those known as Lusitanians, who carry in war quite small shields which are interwoven with cords of sinew and are able to protect the body unusually well, because they are so tough; and shifting this shield easily as they do in their fighting, now in this direction, now in that, they expertly ward off from the body every blow which comes at them” (τῶν δ’ ᾿Ιβήρων ἀλκιμώτατοι μέν εἰσιν οἱ καλούμενοι Λυσιτανοί, φοροῦσι δ’ ἐν τοῖς πολέμοις πέλτας μικρὰς παντελῶς, διαπεπλεγμένας νεύροις καὶ δυναμένας σκέπειν τὸ σῶμα περιττό¬τερον διὰ τὴν στερεότητα· ταύτην δ’ ἐν ταῖς μάχαις μεταφέροντες εὐλύτως ἄλλοτε ἄλλως ἀπὸ τοῦ σώματος διακρούονται φιλοτέχνως πᾶν τὸ φερόμενον ἐπ’ αὐτοὺς βέλος [trans. Old-father, modified]).

  37. See also, e.g., (accessed 02.09.2011) for a photo of unspecified police riot control forces holding round shields, or (accessed 02.09.11) for a round Paulson BS-6 Lexan Gladiator Body Shield, manufactured by Paulson Riot Equipment.

  38. Such a rationale must be behind, e.g., Blyth’s assurance regarding the Bomarzo shield. In Blyth’s assessment, “the total is little more than the weight of a World War II rifle, a weapon which can be handled quite briskly by a trained man” (Blyth [1982] 17). For comparisons of this type to work, it is evident that ancient and modern men must be equal in terms of physical characteristics.

  39. Angel (1945) 284–85 and n. 25. The male skeletons dating from the Classical period—no more than three—are on average 165.4 cm high: Angel (1945) 324. Foxhall and Forbes (1982) 47 correctly warn against possible statistical insignificance due to the relatively small sample material. Insufficient though it may be, however, there is no better way to estimate bodily proportions of ancients, so the material at hand will simply have to suffice.

  40. Angel (1944) 334 table 2a.

  41. Angel (1944) 331 table 1.

  42. Garnsey (1999) 57–59 cites a number of analyses of ancient Roman skeletal material from, among other sites, Pompeii and Herculaneum. He circumspectly concludes that all that may be said for the average male height here is that it is “for the most part within the range 162–170 cm for men and 152–157 for women.”

  43. Foxhall and Forbes (1982) 47 n. 21.

  44. Teeth stemming from several skeletons excavated under the Stone Lion monument at Chaironeia (in all likelihood the human remains of Thebes’ “Sacred Band” elite force of three hundred hoplites, wiped out to a man during the battle there in 338) bear signs of linear enamel hypoplasia, an indicator of systemic stress during childhood; the reasons include severe malnutrition or illness. “The presence of these lines in multiple individuals indicates that even for the future military elite, childhood could at times be stressful in ancient Thebes”: Maria Liston (personal communication); and cf. Hanson (1999) 152–76.

  45. Donlan and Thompson (1976) 341 n. 4; the 170 cm apparently repeated in Stewart (1990) 1:75. Hanson posits some 1.67 m (Hanson 1991) 67–68 n. 14.

  46. (accessed 02.09.2011). The figure given is the average of German, Dutch, British, and American males.

  47. (accessed 02.09.2011).

  48. Foxhall and Forbes (1982) and Garnsey (1999) 17–21 suggest that 70–75 percent of the total consumption consisted in cereals. A postwar survey carried out in Crete demonstrated that no less than 29 percent of the calorie intake was olive oil: Allbaugh, cited in Garnsey (1999) 19 n. 12. Even for heavyweight boxers it was a rare occurrence, apparently deserving of mention, to venture outside the staple diet: Harris (1966) 88–89 relates the few instances; cf. Waterlow (1989) 6–9. See also Henneberg and Henneberg (1998) 512–14; Wilkins and Hill (2006) 114–39, esp. 120–21.

  49. Carter (1998), (2006).

  50. Carter (1998) 5–22, (2006) 21–22, 40–42.

  51. Henneberg and Henneberg (1998) 503–37, at 504.

  52. Carter (2006) 41.

  53. Morter and Hall (1998) 449–54 with table 8.2 and graph 8.2; Carter (2006) 22.

  54. Hdt. 4.15.2 (referring to the χώρη); Carter (2006) 9–15. The Metapontine χώρα was large, covering approximately 20,000 ha (or 200 km2): Fischer-Hansen, Nielsen, and Ampolo (2004) 279–80.

  55. For Attica (otherwise probably the most urbanized of the Greek poleis) see, e.g., Thuc. 2.14; and cf. Arist. Pol. 1256a 35–40, where there is a list of possible occupations. Aristotle then adds “The largest group of people, however, live off the land and off cultivated plants” (τὸ δὲ πλεῖστον γένος τῶν ἀνθρώπων ἀπὸ τῆς γῆς ζῇ καὶ τῶν ἡμέρων καρπῶν). The denotations and connotations of the word γεωργός are many and varied.

  56. Strangely, however, the ratio of buried females to males at Pantanello is almost 2:1, which is all the more puzzling as no predominantly male necropolis has been found on Metapontine territory: Henneberg and Henneberg (1998) 509; Carter (2006) 41–42.

  57. Surprisingly, these average height values are “the same as [those] for Italian peasants in southern Italy before World War II” (Carter [2006] 42; Henneberg and Henneberg [1998] 519–21 with tables 11.14 and 11.15, and 538–41 [appendix 11A.1]). Of other interesting information culled from the skeletons it may be mentioned that life expectancy was on average forty-one years for males, thirty-nine for females (Henneberg and Henneberg [1998] 509–14), and that there were found twelve cases of malaria and also—rather more surprisingly—signs in some individuals of antigens for Treponema pallidum, a form of syphilis (Jeske-Janicka and Janicki [1998] 557–59 [appendix 11A.5]). Moreover, there were high rates of dental hypoplasia, indicating either malnourishment or childhood disease, the latter being the likelier cause: Henneberg and Henneberg [1998] 517–19 and tables 11.10–13; Carter (2006) 42, and cf. above, n. 44.

  58. It is interesting, therefore, that when the Athenian relief force under Demosthenes put in at Metapontion in 413 on their way to Syracuse, they picked up a force consisting of no more than three hundred javelin throwers (and two ships): Thuc. 7.33.4–5. However, there may be more than meets the eye to the seemingly grudging assistance afforded Athens by Metapontion: see Hornblower (2008) 608–9.

  59. Xen. Hell. 6.4.17; [Arist.] Ath. pol. 53.4, cf. IG II 2nd ed. 1926.

  60. In Xen. Hell. 6.1.5, Jason of Pherai is made to boast of his well-trained, mercenary army: “[b]ut armies made up of citizens must include some men who are already past and some who have not yet reached their prime. And there are very few people in each city who keep constantly in good physical training. But no one serves in my mercenary army unless he can stand physical hardship as well as I can myself “ (ἀλλὰ τὰ μὲν ἐκ τῶν πόλεων στρατεύματα τοὺς μὲν προεληλυθότας ἤδη ταῖς ἡλικίαις ἔχει, τοὺς δ’ οὔπω ἀκμά ζοντας·· σωμασκοῦσί γε μὴν μάλα ὀλίγοι τινὲς ἐν ἑκάστῃ πόλει· παρ’ ἐμοὶ δὲ οὐδεὶς μισθοφορεῖ, ὅστις μὴ ἱκανός ἐστιν ἐμοὶἴσα πονεῖν [trans. Warner]).

  61. Hanson (2000) 90.

  62. Holoka (1997) 342; Hanson (20002) 89–95.

  63. Hdt. 7.229.1, 5.111; Xen. An. 4.2.20 (Xenophon’s personal hypaspistēs runs away with his shield, leaving him in a tight spot), Hell. 4.5.14, 4.8.39; Polyaen. 2.3.10; cf. Lazenby (1991) 89.

  64. Hdt. 7.40.1; Thuc. 2.79.5, 4.101.2, 7.78.2; Xen. Hell. 3.4.22, Cyr. 5.3.40, 6.3.4.

  65. Hdt. 1.62.3, 5.74.2, 9.52; Thuc. 2.2.4bis, 4.44.1, 4.68.3, 4.90.4, 4.91, 4.93.3, 5.74.2, 7.3.1, 7.83.5, 8.25.4, 8.93.1bis; Xen. An. 1.5.14, 1.5.17, 1.6.4, 1.10.16, 2.2.8, 2.2.21, 4.2.16, 4.3.17, 4.3.26, 5.2.8, 5.2.19, 5.4.11, 6.1.8, 6.5.3, 7.1.22bis, Hell. 2.4.5, 2.4.12, 3.1.23bis, 4.5.8, 5.2.40, 5.3.18, 5.4.8, 6.4.14, 7.3.9, 7.5.22; Diod. Sic. 11.5.4, 12.66.2, 14.105.2, 18.26.4, 18.61.1, 20.42.5, 20.88.8.

  66. Diod. Sic. 15.32.5; Polyaen. 2.1.2; Nep. Chabr. 1.1–2 (obnixoque genu scuto), and see Stylianou (1998) 297–98. The Athenians later erected a statue of Chabrias in just this position in the Agora, the base of which has probably been found: Arist. Rhet. 1411b 6–10; Nep. Chabr. 1.3; Anderson (1963) 411–13; Buckler (1972), esp. 474.

  67. Hdt. 7.218.1–2: ἀνά τε ἔδραμον οἱ Φωκέες καὶ ἐνέδυον τὰ ὅπλα, καὶ αὐτίκα οἱ βάρβαροι παρῆσαν. ὡς δὲ εἶδον ἄνδρας ἐνδύομένους ὅπλα, ἐν θώματι ἐγένοντο (trans. de Sélincourt).

  68. Cf. Thuc. 6.69.1; Xen. Hell. 4.8.37–39; Plut. Pel. 32.3.

  69. Wilkins (1993) xxxiii–xxxv.

  70. Eur. Heracl. 720–26: εἰ δὲ τευχέων φοβῇ βάρος,/νῦν μὲν πορεύου γυμνός,ἐν δὲ τάξεσιν/κόσμῳ πυκάζου τῷδ’· ἐγὼ δ’ οἴσω τέως (trans. Vellacott).

  71. Anderson (1970) 13–42.


Anderson, J. K. 1963. “The Statue of Chabrias.” American Journal of Archaeology 67:411–13.

———. 1970. Military Theory and Practice in the Age of Xenophon. Berkeley.

Angel, J. L. 1944. “A Racial Analysis of the Ancient Greeks: An Essay on the Use of Morphological Types.” American Journal of Physical Anthropology 2:329–76.

———. 1945. “Skeletal Material from Attica.” Hesperia 14:279–363.

Blyth, H. 1982. “The Structure of a Hoplite Shield in the Museo Gregoriano Etrusco.” Bollettino dei monumenti, musei e gallerie pontificie 3:5–21.

Boardman, J. 1980. The Greeks Overseas: Their Early Colonies and Trade, 2nd ed. London. Buckler, J. 1972. “A Second Look at the Monument of Chabrias.” Hesperia 41:466–74. Cahn, D. 1989. Waffen und Zaumzeug. Ausstellung Antikenmuseum Basel und Sammlung Ludwig. Basle.

Carter, J. C. 1998. The Chora of Metaponto: The Necropoleis, vols. 1–2. Austin, TX.

———. 2006. Discovering the Greek Countryside at Metaponto. Ann Arbor, MI.

Connolly, P. 1998. Greece and Rome at War, 2nd ed. London.

Donlan, W., and J. Thompson. 1976. “The Charge at Marathon: Herodotus 6.112.” Classical Journal 71:339–43.

———. 1979. “The Charge at Marathon Again.” Classical World 72:419–20.

Fischer-Hansen, T., T. H. Nielsen, and C. Ampolo. 2004. “Italia and Kampania,” in M. H. Hansen and T. H. Nielsen (eds.), 249–320.

Foxhall, L., and H. A. Forbes. 1982. “Σιτομετρεία: The Role of Grain as a Staple Food in Classical Antiquity.” Chiron 12:41–90.

Franz, J. P. 2002. Krieger, Bauern, Bürger. Untersuchungen zu den Hopliten der archaischen und klassischen Zeit. Frankfurt am Main.

Garnsey, P. 1999. Food and Society in Classical Antiquity. Cambridge.

Greenhalgh, P.A.L. 1973. Early Greek Warfare: Horsemen and Chariots in the Homeric and Archaic Ages. Cambridge.

Hammond, M. 1979–80. “A Famous Exemplum of Spartan Toughness.” Classical Journal 75:97–109.

Hansen, M. H., and T. H. Nielsen (eds.). 2004. An Inventory of Archaic and Classical Poleis. Oxford.

Hanson, V. D. 1991. “Hoplite Technology in Phalanx Battle,” in V. D. Hanson (ed.), Hoplites: The Classical Greek Battle Experience, 63–84. London.

———. 1999. The Other Greeks: The Family Farm and the Agrarian Roots of Western Civilization, 2nd ed. New York.

———. 2000. The Western Way of War: Infantry Battle in Classical Greece, 2nd ed. Berkeley. Harris, H. A. 1966. “Nutrition and Physical Performance: The Diet of Greek Athletes.” Proceedings of the Nutrition Society 25:87–90.

Henneberg, M., and R. J. Henneberg. 1998. “Biological Characteristics of the Population Based on Analysis of Skeletal Remains,” in J. Carter (ed.), 503–37.

Holoka, J. P. 1997. “Marathon and the Myth of the Same-Day March.” Greek, Roman and Byzantine Studies 38:329–53.

Hornblower, S. 2002. “Sticks, Stones, and Spartans,” in H. van Wees (ed.), War and Violence in Ancient Greece, 57–82. London.

———. 2008. A Commentary on Thucydides, vol. 3, books 5.25–8.109.

Jeske-Janicka, M., and P. K. Janicki. 1998. “Detection of Trepomena Pallidum Antigens,” in J. Carter (ed.), 557–59.

Krentz, P. 1985. “The Nature of Hoplite Battle.” Classical Antiquity 4:50–61.

Kunze, E., and H. Schleif. 1938. Bericht über die Ausgrabungen in Olympia, vol. 2. Berlin.

Lazenby, J. F. 1991. “The Killing Zone,” in V. D. Hanson (ed.), 87–109.

Lendle, O. 1995. Kommentar zu Xenophons Anabasis (Bücher 1–7). Darmstadt.

Mallwitz, A., and H.-V. Herrmann. 1980. Die Funde aus Olympia. Athens.

Millard, A. R. 1994. “King Solomon’s Shields,” in M. D. Coogan, J. C. Exum, L. E. Stager, and J. A. Greene (eds.), Scripture and Other Artifacts: Essays on the Bible and Archaeology in Honor of Philip J. King, 286–95. Louisville, KY.

Morter, J., and J. Hall. 1998. “Dating of Tombs,” in J. Carter (ed.), 449–54.

Rieth, A. 1964. “Ein etruskischer Rundschild.” Archäologischer Anzeiger 1:101–9.

Robinson, D. M. 1941. Excavations at Olynthus, Part X, Metal and Minor Miscellaneous Find: An Original Contribution to Greek Life. Baltimore.

Salmon, J. 1977. “Political Hoplites?” JHS 97:84–101.

Schwertfeger, T. 1982. “Der Schild des Archilochos.” Chiron 12:253–80.

Shear, T. L. 1937. “The Campaign of 1936.” Hesperia 6:333–81.

Stylianou, P. J. 1997. A Historical Commentary on Diodorus Siculus Book 15. Oxford.

Van Wees, H. 2000. “The Development of the Hoplite Phalanx: Iconography and Reality in the 7th Century,” in H. van Wees (ed.), War and Violence in Ancient Greece, 125–66. London.

———. 2004. Greek Warfare: Myths and Realities. London.

Warner, R. 1949. Xenophon: The Persian Expedition. London.

———. 1979. Xenophon: A History of My Times, 2nd ed. London.

Waterlow, J. C. 1989. “Diet of the Classical Period of Greece and Rome.” European Journal of Clinical Nutrition 43:3–12.

Wilkins, J. (ed.). 1993. Euripides: Heraclidae. Oxford.

Wilkins, J. M., and S. Hill, S. 2006. Food in the Ancient World. Malden, MA, and Oxford.

If you find an error please notify us in the comments. Thank you!