Stilicho: The Vandal Who Saved Rome

by Ian Hughes

  Vegetius, 1.20

  Up to the twentieth century the passage has been used as evidence that the later army was no longer equipped with metal armour. This was reinforced by details on sculpted columns, military arches, and other stonework which survives. In these, it was noticed that the troops seemed to wear ‘moulded’ armour formed to resemble human musculature. This was interpreted as being leather armour moulded to form the same shape as earlier breastplates as used by Greek hoplites.

  However, more recent work has overturned this acceptance, and has, for example, shown that the sculptures in many cases have small holes drilled in them to make the appearance of mail armour. Furthermore, it is possible that Vegetius is describing an actual petition from the scholae palatinae, a guard unit, that they be excused from wearing armour.60 In an attempt to emphasize the poor quality of the then-current army, Vegetius simply expanded the request to include the whole of the army, not just a single unit. On the other hand, the remains of copper-alloy scales found at Trier, along with remnants of mail at Trier, Weiler-la-Tour and Indepenta, the latter of which date to the late-fourth or early-fifth century, show that armour was still in use at the time, a point reinforced by Ammianus Marcellinus, who gives many references to individuals wearing armour.61

  Finally, there is the evidence in the Notitia Dignitatum. The drawings that accompany many of the posts included in the document illustrate some of the equipment made in the imperial fabricae (factories). These include the items such as helmets and body armour, previously thought to no longer be needed. As a result of these insights, it is now generally accepted that the late Roman army wore heavy equipment equal to that of their predecessors. However, one more caveat needs to be remembered. In these days of mass-production, it is assumed that as soon as a new model of weapon or defence is introduced, the older one is classed as ‘obsolete’ and withdrawn. In the imperial fabricae, the process of production was time-consuming and expensive. As a consequence, items that were no longer ‘fashionable’ would continue to be issued until stockpiles were used. The overall result is that many items of equipment from earlier periods may have continued in use on a small scale and so remain invisible in the archaeological and sculptural records. This is especially the case with items such as bows, the manufacture of which could take over a year.

  Missiles

  The missiles used by the Roman army can be divided into long-range and short-range weapons. Long-range weapons included bows, slings, crossbows, and a variety of artillery, although larger equipment was usually reserved for use in siege warfare.

  Bows

  The most common form of missile weapon was the composite bow, as evidenced by the repeated use of the title sagittarii (archers) in the Notitia Dignitatum.

  Slings (fundae) and staff-slings (fustibali)

  Slings and staff-slings (slings attached to a 4 ft/1.18 m long stave) were also used. However, for whatever reason they were never to form a large proportion of the military establishment, being restricted to a few skirmishers supporting the combat troops.

  Crossbows (arcuballistae)

  A little-used weapon in military circles, although it is more commonly depicted in mosaics portraying hunting, is the crossbow, which may be the weapon described by Vegetius as arcuballistae.62 Other versions of the weapon included the cheiroballista described by Heron, which was prepared by placing the end on the ground and pressing on the stock until the string was drawn and a bolt/arrow fitted into the weapon. As with the slings and staff-sling, the crossbow remained little-used by the army.

  Artillery

  Artillery had been used by the Roman army since at least the second century BC. Later variants included the manuballista (previously ‘scorpio’ (scorpion)) of Vegetius, a torsion engine capable of accurately firing projectiles for a long distance.63 The army also produced a version mounted on a cart for ease of transport, known as the carroballistae, which is shown on Trajan’s column. There is little doubt that these weapons were used in the field, but the regularity and form of their deployment remains open to doubt.

  Finally, there were the artillery used only for siege warfare, such as the onager (wild ass). These were used only in siege warfare, as they were large and difficult to set up.

  Short range

  Short-range missiles include mainly those thrown by hand prior to contact. Those with the longest range included a variety of darts, whilst for shorter ranges there was a variety of javelins of various weights and range.

  Darts

  Included amongst the darts were types called plumbatae, mattiobarbuli/martiobarbuli, plumbatae tribolatae and mamillatae. All of these weapons were meant to be carried by the infantry and thrown at the enemy as the range closed. In the case of the plumbatae the dart had a lead weight behind the head to aid in penetration, whilst the plumbatae tribolatae is claimed to have had three spikes emerging from the lead weight so that if it missed a target it still posed an obstacle by presenting a sharp point to an unwary foot or hoof.64

  1. A selection of spear and dart heads. (after Bishop and Coulston/Stephenson).

  Javelins

  There were a variety of thrown weapons that come under the loose category of ‘javelin’. These include types called the spicula, the hasta, the pila, the iacula, the verruta and the tela. Despite prolonged investigation, it is clear that the differences between these weapons are unknown.65 Vegetius suggests that each man should be issued one heavy javelin (spiculum) and one light javelin (verrutum).66 However, it should be noted that these different names may be describing weapons that are almost identical. For example, it is known that the older pilum existed in a variety of forms, with some being heavier than others. It is possible, therefore, that Vegetius’ report of a spiculum, which is usually accepted as the newer name for the pilum, actually describes the heavier variety, whilst the verrutum is referring to a lighter version of the same weapon.

  As a result of these deliberations, it should be noted that although some writers refer to all of these weapons being derived from older variants, the new terminology might be different words used to describe the same thing.

  Close-combat weapons

  Swords

  Earlier Roman infantry had been heavily trained in the art of using the short sword known as the gladius hispaniensis. On the other hand the cavalry had been issued with the spatha. For unknown reasons, over the course of time the infantry stopped using the gladius and by the time of the later Empire the whole army appears to have used the spatha. Vegetius also attests to the use of a shorter sword which he calls the semispatha.67 There have been a variety of swords found in the archaeological record that are smaller than the spatha and so might in fact be the sword described by Vegetius. Unfortunately, apart from the name he tells us nothing about the weapon, so any correlation between archaeology and Vegetius remains speculation.

  A long sword, usually thought to be of Germanic origin, the spatha had probably been used by the cavalry due to the need for a weapon with longer reach when fighting from horseback. The spatha was a long, double-edged sword which varied from between 0.7 to 0.9 m in length.68 Although such weapons are usually described as being used in a ‘slashing’ motion, the spatha had a point that made it suitable for thrusting as well.

  2. Late Roman sword from Köln (after Bishop and Coulston).

  Spears

  It is commonly assumed that hand-held, shafted weapons came in two types: those used as missiles and those retained for use in hand-to-hand combat. However, it is clear from ancient sources and modern re-enactors that there was little, if any, difference between the two types of weapon. This leads to the obvious conclusion that whether they were used as missile or hand-to-hand weapons was determined more by circumstances than by weapon typology. Therefore, any of the weapons described in the section on close-range missiles as ‘javelins’ could also be used in combat should circumstances dictate. Furthermore, spears such as the spiculum, which had a large part of its shaft encased in iron, would be ideal in com
bat as the iron would protect the wooden shaft from being sheared by enemy swords.

  Yet alongside these variants are the types that would be classed simply as ‘spears’, used by both infantry and cavalry, ranging between 2 and 2.5 m in length (see, for example, Plate 26: the Stilicho diptych).69 These weapons could either be thrown a short distance or retained for combat, and it is possible that their use can be interpreted as a change in fighting styles from earlier periods, as will be discussed below.

  Others70

  Alongside these weapons others were in use, although in most cases to a lesser degree. Many burials included a short, single-edged knife, which by this time appears to have replaced the earlier broad-edged dagger, the pugio. The change in the design of the dagger may simply have been a recognition that its use as a utility tool far outweighed its employment as a weapon. As a consequence, it became simpler and was only sharpened on one side.

  3. Detail from the sculpture at Gamzigrad. Note that the infantryman appears to have a ridge helmet and that the cavalryman is wearing a pilleus Pannonicus and carrying an axe.

  Alongside traditional Roman weapons were others which were either of unknown origin or were Germanic imports. For example, there is evidence that some at least of the Roman cavalry used conventional axes, as mentioned by Ammianus and Procopius and shown in the stela from Gamzigrad and the Column of Arcadius.71 Unfortunately, these are limited examples of the use of axes and so their distribution remains a mystery. There is also the use of maces, as mentioned by Theophylact.72 Germanic imports included weapons such as the seax (a single-edged long knife) and the francisca (throwing-axe), which were slowly being introduced into the Empire, and attested later is the use of the lasso, following Hunnic practice.73 Again, the extent of their use in the Roman army remains unknown.

  Defensive equipment Helmets

  The earlier use of helmets with single-piece bowls spun from a single piece of metal disappears before the middle of the fourth century. Their place is taken by two new forms. The most common of these are the styles termed ‘ridge helmets’. Possibly deriving from Persian helmets, they are first found in archaeological deposits dating to the early fourth century, possibly c. AD 325.74 This date is confirmed by a coin from the reign of Constantine I (306–337) which appears to show Constantine wearing a ‘stylized Berkasovo helmet’, and by the sculpture found at Gamzigrad dating to the end of the third century (figure 3).75

  4. Front and rear views of a ridge helmet (after Stephenson).

  There are several slightly different styles, all – as in the ‘Berkasovo’ example just cited – named after the place where they were found. There are many finds from around the Empire, but probably the most important were the up to twenty examples found at Intercisa (modern Hungary). There was a variety of styles involved, including the extraordinary version with an integral metal crest known as ‘Intercisa 4’. Yet all of these were made using a similar technique, which was to manufacture the bowl as two separate pieces before joining them with a strip of metal along the crown, which gave them their distinctive ‘ridge’ appearance.

  5. A more complex ridge helmet from Conçesti (after Stephenson).

  It would appear that many of these helmets had attachable crests, and it may be that the ‘integral crest’ of Intercisa 4 was either cheaper than buying a separate crest or may have been a way of distinguishing officers from other ranks.76

  6. The Intercisa 4 ridge helmet and the Leiden spangenhelm. The differences in construction and the large metal crest on the Intercisa 4 are clear.

  At some point in the fourth century, if not earlier, the Roman army adopted another form of helmet, the spangenhelm. Named after the spangen, the plates that joined the separate parts of the bowl together, they may be dated as early as the Tetrarchy (c.293–312), although this date is uncertain and they may only have been introduced in the fifth century.

  The reason for the change from one-piece bowls to ‘ridge’ helmets (and possibly spangenhelms) is unclear. Earlier claims that this was due to the expansion of the army under Diocletian and the need to supply equipment that was cheaper and easier to make have recently been questioned.77 It has been pointed out that the new manufacturing method required accuracy in order to join the two halves of the bowl properly, so making them difficult to manufacture. Furthermore, the fact that many of them have traces of silver, gilt and/or paste gemstones attached results in the end product actually being quite expensive.

  These claims do not take into account the fact that the process of spinning iron can both weaken it and lead to irregularities in the bowl. This may account for the need to reinforce earlier, one-piece bowls across the brow. Furthermore, unless the manufacture was extremely well controlled, the ensuing bowl could be slightly off-centre and so weak down one half. This would have resulted in a high wastage of material as sub-standard bowls were returned to the forge for remaking. The new methods produced bowls that did not need brow reinforcement and were of a more uniform thickness and quality, since they are easier to work and toughen than the one-piece skull.78 Although looking to modern eyes, with computer-driven accuracy, as if they are a step back, in production and quality they may actually have been an improvement on earlier helmets.

  Although a little strange to modern eyes, there are artistic representations of troops wearing coifs.79 These appear to be allied to extremely long coats of mail. How they were manufactured or what form they took in reality is unknown, as none have ever been found in the archaeological record.

  A final piece of protective headgear was the ‘pilleus Pannonicus’ (Pannonian hat). This was a round, flat-topped cap. The earliest depiction is from the coins of Constantine I and from the Arch of Constantine. When seen in detail it is depicted as being ‘brown and furry’ and was probably made of felt.80 Of little defensive value in itself, its use amongst the military may have started due to its being used as a helmet lining. It may then have become a symbol for members of the army who were not wearing their helmets. In the porphyry sculpture of the four tetrarchs in Venice, the four emperors are wearing the ‘pilleus Pannonicus’, possibly as a sign of their affiliation with the army.

  Body defences

  It is sad that due to decomposition or rust the vast numbers of helmets that were once common in the Empire have now been reduced to a mere handful. The case is even worse when it comes to body armour. Therefore, what follows is largely conjectural and could be supplanted at any time by new archaeological finds.

  Mail

  Simply put, mail armour was made by making lots of small rings of iron and joining them to make a flexible, but rather heavy, form of protection.81 As has already been noted, there have been isolated finds of mail dating from Stilicho’s era, enough to show that it was certainly still being made. Monumental evidence suggests that the mail came in two lengths: either a ‘short’ version covering the shoulders and coming down to around mid-thigh, or a ‘long’ version which reached to the elbow and the knee.

  7. Details from the Arch of Galerius. The figure on the left, who might be the emperor, appears to be wearing a ‘muscled cuirass’, whilst that on the right appears to have holes drilled in the stone to represent ring mail.

  Scale

  Scale armour is made by lacing small overlapping plates of metal onto a fabric or leather structure. The result is a form of protection roughly equivalent to that of mail, but without the large amount of flexibility that mail offers. It would appear that the shape of the scale ‘shirts’ took the same form as those of the mail shirts already described.

  ‘Muscle cuirasses’

  It is common for depictions of muscle cuirasses, reminiscent of the styles found in ancient Greece, to be represented on large monuments in this period such as the Column of Arcadius. Historians in the past claimed that these were made of leather, possibly ‘cuir boulli’, mainly due to the style of the carvings appearing to depict movement in the cuirass, which would not be represented by metal.82 There is no evidence anywhere else to suppor
t this theory, and there are technical difficulties with some of the proposed materials.83 Therefore, unless further evidence comes to light, the use of leather remains a possibility, but no more than that.

  There are difficulties with accepting depictions of muscle cuirasses on such monuments at face value. The most obvious of these is that the monuments probably owe more to traditional Hellenistic forms of carving rather than representing contemporary models. For example, on the Column of Theodosius troops with muscled cuirasses are also depicted with pteruges (sometimes spelled pteryges: the ‘skirt’ of leather which commonly covered the wearer from the hips down, as well as over the shoulders), attic helmets and hand-straps near the rim of their shields. All of these items refer back to ancient hoplite practices rather than reflecting what troops of the time actually wore.84 Furthermore, a close analysis of the monuments has shown that some of these representations of muscle cuirasses actually appear to represent scale or mail armour. This would reinforce the theory that such monuments were carved with an eye more to the Hellenistic traditions of the past than to the accurate representation of contemporary armour.85

  A further problem is the manufacturing process. Metal cuirasses needed to be tailored to fit the individual. As a result, the use of such armour is often dismissed by historians as it is extremely time-consuming and expensive to make. However, as many of the monuments no doubt depict senior officials and dignitaries that could afford such luxuries, it is likely that some of these representations may be accurate. Therefore, it is possible to claim that muscle cuirasses may have been worn by rich, senior officers but not by the common soldier.