The Unknown Warrior (5 page)

As far as Britain is concerned, excavations at the fort of the II Legion at Caerleon, Wales, have produced numerous sword fittings. These include a bone-ribbed
gladius
handle. This piece, some 86mm long, though broken, was believed to have come from a Hadrianic–Antonine context – the early to mid-second century
AD
. Bone scabbard chapes were also found on this site (Greep, 1992: 188–9). Another Welsh site yielded a
gladius
; the iron sword from Caernarvon, some 46cm long, had a badly corroded blade and cracked bone grip (de la Bédoyère, 2001: 28). A bone
gladius
handle was uncovered from a first-century ditch at Aldgate in London (
ibid.
: 28).

The military dagger or
pugio
was not only a front-line weapon, but also a tool of campaign, used perhaps for cutting food as well as opponents. Suspended from the left side of the legionary's belt, the dagger ranged in size from
c.
20 to 35cm (Goldsworthy, 2003: 134) with a scabbard that could be richly decorated. By the second century
AD
, the dagger had become less common as a weapon, so much so that it is not depicted on Trajan's Column in Rome (
ibid.
: 134). Having said this, the weapon still appears on tombstones of this period and the discovery of a military workshop at Künzing, Germany, the stock of which was buried in the third century
AD
, revealed some fifty-nine dagger blades and twenty-nine sheaths (Feugère, 2002: 128). In addition to the sword fragments discussed above, the remnants of the defeated Roman force at Kalkriese also included dagger parts: bronze rivets from the hilt of a weapon as well as a fragmentary iron blade. Iron daggers and their sheaths have been found in Britain, too – from Copthall Court, London (de la Bédoyère, 2001: 213; Feugère, 2002: 128) and from Exeter (I.R. Scott, 1991: 263 and fig. 120). Rarer, perhaps, was the beautifully decorated sheath that was found with the body of a Roman at Velsen in the Netherlands, discussed in detail below.

In addition to blade weapons, the Roman soldier used projectiles to deadly effect. Following the reforms of their commander, Marius, in the later years of the second century
BC
, infantrymen were each provided with a heavy throwing spear or
pilum
, The 2m-long
pilum
was formed from a heavy wooden shaft into which a long thin iron spear tang with sharp point was attached. A pointed iron ferule completed the weapon. Two methods of attaching the iron head to the wooden shaft were used: ‘Some of the iron heads ended in a socket, the joint reinforced by an iron collet fitting over the top of the wooden shaft, but the majority had a wide rectangular tang which slid into a groove in the wood and was fastened into place by two rivets' (Goldsworthy, 2003: 132). By all accounts, the metal tang of the
pilum
was designed to bend on impact rendering the weapon useless and either preventing the enemy from throwing it back or encumbering him should it stick in his shield. This weapon was in use until the third century
AD
and had, according to Goldsworthy, a range of
c.
15m requiring close discipline in its use (
ibid.
: 132). Smaller spears could also be used to provide a lethal barrage.

Archaeologically, it is only fragments of these
pila
that have been recovered. A
pilum
head and a couple of collets have been found at Kalkriese (Schlüter, 1999: 138; Cowan, 2003a: 25–6) in addition to several other iron spears. Feugère points to a number of other
pilum
finds from elsewhere in the Empire, including several shafts from the valley of the Saône in France, and others from Oberaden, Germany, with collets recovered from Hod Hill in Dorset (Feugère, 2002: 130). A socketed iron
pilum
head, 241mm long, was also part of the assemblage in the hoard from Corbridge discussed below (Allason-Jones and Bishop, 1988: 9). Forty-six throwing spears were also present in this hoard, many of which had been bound together and were fused when excavated, as were three catapult bolts and several ferrules and sockets illustrating the throwing arsenal available to the legionary (
ibid.
: 10–17). As with other pieces of the infantryman's equipment, spears were sometimes inscribed by their owners and thus we have traces of names of those who fought to maintain the Roman Empire. One first-century example from Bucklersbury House, London, proclaimed it was the ‘(property) of Victor in the century of Verus' (Collingwood and Wright, 1991: 51).

Although not part of the standard legionary equipment, slings were used effectively in sieges and add colour to the picture of Roman warfare. The shot from these was often of baked clay or lead and some Roman examples from the second and first centuries
BC
, like their Greek predecessors, bore inscriptions. This subset includes invective aimed at the enemy – for example, the sardonic
avale
or ‘swallow this' (Feugère, 2002: 160). Under battle conditions, sling shot could be manufactured with relative ease. At Velsen, in the Netherlands, a soldier poked his finger into local sand to provide a simple mould for the lead that would make the shot. In so doing, a cast of his finger, complete with nail, was preserved. These lead projectiles (
glandes plumbeae
) were also among the weapons available to the ill-fated soldiers at Kalkriese in
AD
9 (Schlüter, 1999: 138).

ARMOUR

In addition to countless portrayals on gravestones, triumphal arches and sculptures, fragments of Roman armour have been recovered from a multitude of sites – one of the most important of recent years being the battle site of Kalkriese, Germany. This site, dating to
AD
9 has yielded a large quantity of equipment discarded by or hacked from the bodies of legionaries defeated by German adversaries. Some of this material is fragmentary and may simply be elements of the armour neither taken as booty, nor set up as a trophy by the victors. Elements of two types of body armour have been found, namely of mail (
lorica hamata
) and of articulated plate armour (
lorica segmentata
). Of the former (probably, as Goldsworthy (2003: 126) states, the most common early body protection), a fine fastening hook was excavated with a human face depicted upon it and ending in an animal's head terminal, while the latter is attested by hinges, buckles and a part of a breastplate made of both iron and bronze, some 18.8cm in length (Schlüter, 1999: 136–7; Patscher and Moosbauer, 2003). Roman soldiers were also provided with an apron to protect the groin – parts of the silver and bronze fittings of such protection having been found.

London also produced a piece of
segmentata
, which may have been damaged in conflict: a fragment of breastplate from the Bank of England seems to have suffered a blow just below one of its hinges (Bishop, 2002: 83).

One should remember that, although very similar, Roman equipment was not totally uniform – certain pieces would vary and would have been adapted or handed down. Armour was worn over, or stitched onto a leather tunic, as without it the legionary would have found the protection far too uncomfortable to wear.

Articulated plate armour has been found on many fortress sites, too. Exeter has produced a series of the essential fittings for this armour – buckles, hinges and rivets – although, as the authors point out, ‘they do not, unfortunately, add up to much more than would be required for one shoulder' (Holbrook and Bidwell, 1991: 244).

One of the most sensational finds of Roman armour came from the fortress site of Corbridge, northern England. Uncovered in July 1964, a hoard of Roman equipment in the remains of a wooden chest was excavated. This chest had been placed in a rectangular pit, which had been dug through destruction layers of an earlier building. Archaeologists were able to date the deposition of this hoard to around
AD
122–138. As well as the projectiles mentioned above, various tools were present, including an iron pick axe, chisel, bow or frame-saw, knives, pulley block, hinges, nails, lamp and bracket, and the remains of a sword scabbard and writing tablets (Allason-Jones and Bishop, 1988: 53–60). The armour was probably the most important part of the collection, enabling the reconstruction of the full layout of one pattern of Roman armour. Many elements of
segmentata
were present, including breastplates (up to 90mm wide and 95mm deep), shoulder guards and collar plates. The breastplates even held traces of the mineralised remains of internal leather straps and textiles (
ibid.
: 23–51 and fig. 53).

Mail alongside a probable armour plate with a hinge, a buckle and a rivet was found in the excavations of the Roman Gates at Caerleon, Wales (Evans and Metcalf, 1992: 166). Verulamium (St Albans), Kingsholm (Gloucester), London and Richborough, Kent, also produced elements of
segmentata
fittings.

Sauer (2000) discusses its presence in the early Imperial fortresses of Vindonissa (Germania Superior) and Oescus (Moesia Inferior) among others including his own excavations at Alchester in Oxfordshire.
Lorica segmentata
seems to have gone out of fashion in the third century (Goldsworthy, 2003: 128), though as it has been found at Kalkriese,
AD
9, this type of protection was in vogue for more than 200 years.

One must be careful not to assume that finds of armour in a fort or barracks represent the presence of an armoury; it may have been left either as some type of votive deposit or, in some cases, of clearing away material no longer required. This is the interpretation given to the armourers' hoard found at Corbridge. Many of the objects were damaged and either awaiting repair or scrapping and were abandoned along with the fort (Allason-Jones and Bishop, 1988: 109).

In addition to segmented armour hinges, Alchester produced segments of scale armour (
lorica squamata
), which was a form of protection for which we also have iconographic evidence (Sauer, 2000: 25). A further British example of scale armour, along with mail, was found at the fort of the II Legion at Caerleon (Brewer, 2000: 30; de la Bédoyère, 2001: 98). In terms of finds of scale armour, perhaps the most important discovery has come from Carpow in Scotland. Here, not only were scales (each
c.
1.3cm wide by 1.5cm long) uncovered in a pit within the Severan fortress, but also the textile backing and leather binding, which had been preserved as a result of corrosion products from the bronze scales. The scales were fastened by means of a cord running over a pair of holes on each scale, with linen yarn passing through these holes and affixing an underlying cloth (Wild, 1981; Dore and Wilkes, 1999). Large numbers of late first-century protective scales were recovered from the excavations at Newstead in Scotland (Goldsworthy, 2003: 128). Carlisle has also yielded some splendid examples of scale armour. Research on this material is under way, but already conservation analysis has revealed that, although ‘the armour is mainly iron, copper alloy coatings and wire have been found on the scale armour, and copper alloy rivets on some of the laminated limb defences' (Jones and Watson, 2003: 11).

Fragments of intact mail have been found on Roman sites on the European mainland, too. In addition to the segments from Kalkriese, Feugère (2002: 100) points out examples of links and attachments from Chassenard, France, and Künzing, Bavaria, and compares them with the depiction of such a protective coat on Vachère's statue of a Gallic nobleman (Goldsworthy, 2003: 30).

Leather armour was almost certainly worn by the early Roman soldier, but our archaeological understanding is dictated by preservation of such organic remains. Waterlogged or desiccated conditions are required if such materials are to survive. Out of the scope of this study, as it is from the third century, the fortress site of Dura Europos in Syria retained leather limb armour (James, 2004: 122–4).

With all the choices of protection available to the Roman legionary, it may come as a surprise that, as Cowan highlights, some front-line infantrymen fought without armour. ‘Casear made use of such legionaries to fight as
antesignani
, that is lightly equipped legionaries (
expediti
), who probably skirmished with light missiles in front of the main battle line or reinforced the cavalry … A relief from the legionary headquarters building (
principia
) at Mainz shows two legionaries fighting in close order, equipped with
scuta
and
pila
, but apparently without body armour, suggesting that even the ‘heavy' legionaries could fight
expediti'
(Cowan, 2003a: 31).

Shields

The shield (
scutum
) that protected the early Roman infantryman was composed of laminated wood and hide with metal facings and shield boss. An example of this first-century
BC
type was recovered from Fayum in Egypt and was 128cm long and 63.5cm wide, 1–1.2cm thick and around 10kg in weight (Cowan 2003a: 27; Goldsworthy, 2003: 31). These shields were curved and oval in form.

A later, Augustan variant was more in the shape of a curved rectangle and had the great advantage of being lighter. This curved shape covered the majority of the soldier's torso and, when linked together, could cover a group of infantrymen from attacks from the side, in front and above. This ‘tortoise' formation is depicted on Trajan's Column in Rome. The only complete example of one of these sub-cylindrical shields was found at Dura Europos and many reconstructions are based on this example. The Dura shield from Tower 19,
c.
AD
250, was formed from plywood around 5mm thick and 5.5kg in weight. Neither shield boss nor its associated rivets survived, but, incredibly, its stunning painted decorations did (James, 2004: 176–9).

Infantrymen were carrying shields on the ill-fated mission at Kalkriese in Germany. When wet and heavy, this equipment may actually have been more of a hindrance than help as it encumbered the legionary. Shield fragments the metal fittings – have been recovered from this site, including a shield boss (
umbo
), shield grip (
ansa
) and edge-binding (Schlüter, 1999: 136). The former could sometimes be included in a list of offensive weapons used by Roman infantrymen, as part of their tactical panoply was to use this part of the shield to drive into their opponents.