For Gods or men? A reappraisal of the function of European Bronze Age shields.

Molloy, Barry

Introduction

From heraldic coats of arms to the badge of US police officers, the shield is a powerful symbol of authority and strength. In Western society it is an assertive if non-aggressive representation of power, a fossil from the heart of combats throughout the millennia where edged-weaponry dominated the battlefields of Europe. Today the shields of the Bronze Age in Europe are still visually striking artefacts seen in museums around the continent, highly decorated and pleasing to the eye. Though dull and tarnished with age when rediscovered, the shields of metal would once have been polished to a rich golden hue.

In important research carried out nearly five decades ago, John Coles conducted experimental tests on replica shields of the Bronze Age, innovatively combined with a catalogue of the then known shields of Europe, so integrating methodical analysis with practically informed interpretation (Coles 1962). Coles developed a two-tier model of functionality, arguing that leather and wooden shields were functional, whereas shields of bronze were not, a model which has been widely accepted.

The purpose of this paper is to re-examine the basis for this distinction using new experiments and metric data from the surviving shields. It will be demonstrated that a strict functional divide by material of construction is inappropriate and that both metal and organic shields could be highly effective articles of defensive weaponry. Conversely, it will be argued that shields of any material may have functioned in non-martial roles and that we need not conceive their multiple functions as being mutually exclusive. As with contemporary swords and spears, shields would have had a plurality of meanings in society and so any single item may at various times have been a symbol of identity, a tool of combat or a votive offering. Evidence for organic shields of wood and leather survives today only from Ireland, though artistic depictions from Iberia in particular indicate they were once much more widespread (Coles 1962; Harding 2007). Metal shields are found throughout much of Europe, and in particular in the British Isles, Germany and Scandinavia (Coles 1962; Raftery 1982; Osgood 1998; Harding 2000, 2007; Uckelmann 2006, forthcoming).

The shield in context

As a weapon on the field of battle, the shield was used in unison with offensive weapons. There was a wide range of attack weapons available to the Bronze Age warrior. The first swords emerged around 1600 BC, and were light, thin, rapidly deployable weapons offering a limited repertoire of cutting and thrusting attacks, typically dubbed 'dirks' and 'rapiers' in English language literature (Burgess & Gerloff 1981; Ramsey 1989, 1995; Molloy 2006, 2007). These were superseded around 1200 BC by a range of leaf-shaped swords originating in the Balkans and northern Italy (Cowen 1955, 1966; Catling 1961; Kilian-Dirlmeier 1993) which spread in various forms throughout most of Europe, from Greece in the southeast to Ireland in the north-west (Cowen 1951; Eogan 1965; Bridgford 1997, 2000; see also various volumes in the Prahistorische Bronzefunde series IV). More numerous again than the sword, the spear was a weapon which perhaps best typifies the diversity of the Bronze Age battlefield (Ehrenberg 1977; Hockmann 1980; Avila 1983; Ramsey 1989; Rihovsky 1996; Davis 2006). Spears came in all shapes and sizes throughout the Bronze Age, bur it was typical for two and three forms to be current in any given area at the same time, and these usually occurred in a variety of sizes (Ramsey 1989).

The Bronze Age axe is demonstrably a tool well suited to cutting down trees (Mathieu & Meyer 1997), bur its role as an implement of violence is suggested by its deposition in hoards containing nothing else but weapons and occasional personal ornaments (in Ireland, for example, see Eogan 1983: nos. 14, 17, 43, 76, 95, 98, 110, 119, 155). One also finds a wide variety of daggers and stone mace-heads which would have served well in combat. One thing which most of these weapons have in common is that they only required the use of one hand, and the shield is an obvious candidate to fill the other. Shields range in size from around 300mm to over 700mm in diameter (Coles 1962; Molloy 2006; Uckelmann forthcoming), and the materials used vary greatly in their weight and mechanical properties. Thus these were weapons which had immense diversity in their combat applications and, by extension, their modes of use.

Bronze Age shields

Radiocarbon dating of a wooden shield-mould from Kilmahamogue in Ireland puts the emergence of leather shields into the first hall of the second millennium BC (Hedges et al. 1991 ; Waddell 2000: 240). This is the era when spears increase in complexity and swords first emerge, so this date fits well with the martial milieu of the rime. Coles (1962) meticulously reproduced the first replica leather shields and found them to be highly effective defensive weapons. These shields were made from sheets of leather cut to a circular shape and impressed on a wooden former or mould, giving them their distinctive shape and decoration. This 'decoration' serves to corrugate the otherwise flat structure of the shield, giving additional structural support in much the same way that thin sheets of steel are corrugated for use in roofing to afford them lightness yet strength. Leather shields once moulded to shape can then be boiled or baked and impregnated with wax to further strengthen them. While two wooden shield formers are known from Ireland, only one actual leather shield survives, the example from Cloonbrin (Figure 1). This is a 4-5mm thick piece, measuring some 500mm across with three impressed ribs of 7-10mm depth and occasionally small bosses around a central boss (Molloy 2006). This latter boss covers the hand and is the most prominent and exposed element of a shield, so it is no surprise to find an additional 2mm-thick layer of leather sewn on over this to re-enforce it and so protect the hand of the user behind.

[FIGURE 1 OMITTED]

Leather shields were not the only organic forms to survive, and again from Ireland we have two examples of shields made from wood. Throughout much of history, from the Romans to the Normans, wood was a material of choice for shields as its fibrous structure absorbs blows and is resistant to cutting (especially against the grain). The Irish shields were made from alder wood, formed of single blocks carved from planks cut longitudinally from the trunk of the tree. The two surviving shields differ markedly in their form, the broad example is from Annadale, probably extending 655mm or more in its original diameter and averaging 13mm in thickness. This was a quite robust object weighing in at just over 1.8kg, but it was still only about half the weight of Anglo-Saxon and Viking shields of the same diameter (Underwood 2007:136). As with the leather shields, it had ribs decorating its outer face (purely decorative/symbolic) and a central boss giving ample protection to the handle and the hand of the wielder.

The wooden shield from Cloonlara is also made of alder and carved from a longitudinally split plank, but its proportions mark it as an unwieldy 'shield-shaped' lump of wood. It is a mete 450mm in diameter yet it is 30-40mm in thickness and it weighs nearly 4.5kg despite its very thin handle. As with the Annadale shield, this piece retains the raised concentric ribs interrupted by a notch seen on the Churchfield shield former (although U-shaped in this case rather than the V-shaped notches of the latter), though the [sup.14]C dating evidence places it several centuries later to around 1200 BC (Hedges et al. 1993).

Shields made from bronze survive in greater numbers in Ireland than their organic counterparts, and from the rest of Europe only shields made from bronze survive, though they come in a wide range of shapes, sizes and decorative motifs (Coles 1962; Uckelmann forthcoming). What is particularly interesting is that the different shields made from this medium impose very different performative patterns on the user, and indeed represent distinctly different combat modalities. Working initially from the Irish dataset, at the smallest end of the scale, there are three shields from Athenry, Lough Gara and Athlone (of Athenry-Eynsham type), which are all around 300mm in diameter, and are typically in the region of 0.8-1.5mm in thickness, and weigh around 1kg, not much heavier than a contemporary sword (Eogan 1965; Molloy 2006). In functional terms, the Nipperwiese shields which are found from Britain across to Poland are as thick as the Irish Athenry-Eynsham shields and while marginally broader, they indicate a wide currency of robust small shields in many areas of Europe (Coles 1962; Needham 1979; Raftery 1982; Uckelmann 2006, forthcoming).

The larger bronze shields are in the region of 650-700mm in diameter, markedly broader than the smaller variants, though in thickness they are somewhat thinner, falling in around 0.6-1.2mm in thickness (0.7-0.9 on average), with notable differences in thickness occurring within millimetres of any particular reading (e.g. the Loch Gur shield shown in Figure 2). Unlike the flat-rimmed Athenry-Eynsham shields, the Nipperwiese and Yetholm types have substantial rolled rims, many with the metal folded around a thick bronze wire. Using two Irish Yetholm examples, the rim thickness is thus increased to some 4.5mm on a shield from Barry Beg and 5.35mm on the Lough Gur piece, significantly reinforcing the rim and overall structural integrity of such shields. The Barry Beg shield at 1.868kg is typical of this class, and it is interesting to note how closely this weight correlates with the functional wooden shield from Annadale. The trade off between coverage, manoeuvrability and durability for bronze shields seems to be optimised around 0.7-0.8mm thickness and 1.5-2kg in weight; adding a mere 0.1mm thickness adds as much as 325g extra weight. Yetholm type shields were particularly popular in Britain and Ireland, though they are found as far afield as Denmark. The British Harlech shields were of a thickness comparable (Uckelmann forthcoming) to the replica Yetholm shield discussed below, though slightly less elaborately decorated. The Herzsprung type which was found typically in Germany and Scandinavia were similar to the Yetholm and Harlech types in many cases, although some few of this type are perhaps too thin to have been serviceable weapons.

[FIGURE 2 OMITTED]

The function of shields

A shield is a piece of defensive weaponry, not an element of defensive armour. The marked difference in this is that a shield must be manipulated by the user in order for it to perform its function, as its presence alone rarely affords sufficient protection. Another aspect of viewing this as a weapon is that it can be used not only to defend against attacks, but is a central component in making attacks, primarily through striking and control of space. Even large and potentially cumbersome shields such as the Roman scutum were highly versatile weapons allowing for the 'buzz-saw' of the Roman shield-wall using short swords, yet legionaries were also required to learn a number of offensive strikes using their shields (Coulston 2007: 42). A large shield offers a good degree of static coverage to the user by virtue of its presence, and is ideally suited to cooperative styles of fighting and forming a shield-wall as found in Bronze Age (Molloy 2006), Classical Greek (Van Wees 2004), Roman (Coulston 2007), Viking (Siddorn 2000), Anglo-Saxon (Underwood 2001) and Byzantine (Nicolle 1992) battlefield strategies. The role of larger shields in individual combats is seen from Roman gladiators to medieval single combats (Talhoffer 2000; Coulston 2007: 44-6). One common aspect of any of these contexts of combat is that the shields had sufficient manoeuvrability to allow the user to manipulate them to afford personal protection while also to thrust forward with them into an opponent's space.

To block an incoming attack a shield is rarely used as a static impediment but is rather used actively to redirect the force of the attack, loosely characterised as 'slapping' the attacking weapon so that the strike is received obliquely to the face of the shield rather than allowing its full force to be transmitted to its structure. It was desirable to avoid direct blows to the edge of the shield, since this increased the chances of splitting wood or cutting into leather or metal, because the maximum force of a strike lands on the minimum surface area. In battle, such blows could lead shields to fail, and the user would likely be killed. Such was the nature of combat.

Being 'pro-active' with a shield by stepping forward into an opponent's space with the shield leading can effectively throw them off-balance or inhibit their ability to make an aggressive strike. Leading with the shield edge (rather than its face) presents a thin almost blade-like element with which to strike at an opponent, particularly to the face or neck. This is a mode of use very well suited to shields of smaller diameter, as exemplified by the medieval targe or buckler, a shield typically in the region of 300-400mm in diameter. These shields must be used in very dose coordination with the offensive weapon and require a broad panorama of movement in order to protect against attacks, having to protect a sweep of 180 [degrees] and more. The medieval descriptions of their use place them not only in single combat, but in mass attacks, as illustrated by Machiavelli's description of the battle of Barletta:

'When they came to engage, the Swiss pressed so hard on their enemy with their pikes, that they soon opened their ranks; bur the Spaniards, under the cover of their bucklers, nimbly rushed in upon them with their swords, and laid about them so furiously, that they made a very great slaughter of the Swiss, and gained a complete victory' (Machiavelli 1560: 66).

New experiments with replica shields

In Coles' earlier trials (1962) one of the important deductions was that some bronze shields from Britain and Scandinavia in particular were simply too thin to be used effectively in combat. The 0.3mm example manufactured by Coles after the shield from Coveney Fen was easily cut through by a Bronze Age sword, and from this it has been widely perceived that bronze was unsuitable for the manufacture of effective shields or armour.

[FIGURE 3 OMITTED]

On the other hand, at 0.3mm, the Coles replica was, approximately two to three times thinner than a typical Yetholm type shield, and four to five times thinner than most AthenryEynsham and Nipperwiese shields. While extremely thin Bronze shields may not have been functional in combat, the thicker ones were well suited to this end. Several examples from Ireland and Britain still bear physical testimony to their use in this environment, e.g. the Long Wittenham shield, the Barry Beg shield (Figure 3) and the Cloonbrin shield (Needham 1979; Osgood 1998; Molloy 2006). To examine further the question of potential performance of shields in combat, the author carried out experiments using replica shields, one of leather and three of copper, complemented by use-wear analysis of surviving pieces from Ireland.

Leather shield

At 3.5-4mm, the leather shield was slightly thinner than the Cloonbrin shield (original 4-5mm). After repeated right-handed strikes (c. 25 strokes), particularly to the upper right-hand quadrant (looking face-on) the leather began to degrade and bend back on the arm of the user. However, small and light shields of this form are typically used to actively deflect blows so that the blade's edge runs along the face of the shield. Cutting attacks parried in this fashion lightly incised the face of the replica leather shield without severe damage. Damage was inflicted when strikes were made squarely to the edge of the shield, cutting into it by up to 57mm (Figure 4). The four types of damage inflicted in testing were: (1) degradation of the leather; (2) incisions to the face of the shield; (3) punctures as a result of stabbing attacks; and (4) cuts through the perimeter of the shield along the edge.

[FIGURE 4 OMITTED]

A use-wear study of the single surviving example from Cloonbrin exhibited all four forms of the damage discussed above. When the shield is held in a fist grip (the hand roughly parallel to the ground), the upper right-hand quadrant of the Cloonbrin shield has considerable leather degradation similar to the replica shield (compounded by subsequent deterioration of the material). In this same quadrant there is a cut to the edge of the shield similar to those observed on the replica shield and coming at the correct trajectory to be inflicted by a right-handed user. Slices and a few minor puncture marks are visible to the face of the shield also mirroring the damage on the replica shield.

The Bronze shields

Three copper shields were made, two with rolled rims and one with a flat rim, to a mean thickness of 0.9mm, reflecting the majority of shields from Europe with thickness in the range of 0.6-1mm (Molloy 2006; Uckelmann forthcoming). The form of one replica followed Athenry-Eynsham type shields (320mm diameter), another, the typical Nipperwiese design (378mm diameter), while the final piece imitated Yetholm varieties (710mm diameter). The replica shields were manufactured from pure copper, a material mechanically similar but slightly less hard than true bronze piece (tin-bronze being unavailable). The shields were tested against replica bronze swords manufactured using authentic alloys and dimensions by the author and Neil Burridge of Bronze Age Craft (www.bronze-age-craft.com/swordcasting.htm).

The copper shields tested were more effective than the leather one in most regards. All three pieces stood up to extensive impact testing with no damage rendering them unusable. The smaller two in particular have been subjected to repeated tests and public demonstrations (notably at the Sixth World Archaeological Congress) since their manufacture, and apart from needing minor 'cosmetic' repairs, they are still perfectly functional. As with the leather shield, cuts and thrusts to the face of the shields when deflecting blows only inflicted minimal damage. The Yetholm shield provided such a broad area of contact that most cuts merely incised the face slightly, though some caused a slight buckling of the face of the shield.

It should be noted that in the interests of safety, all shields were held stationary, an unrealistically static mode of use which maximises the impact force of a strike because there is no relative movement allowed between the two weapons. Cuts to the edge of the unrolled examples cut 11-15mm into the edge of the shield, but did not render terminal damage. On the shields with a rolled edge, an important purpose of this device became clear. When the sword impacted on the edge of the shield, the broader contact area meant that the sword did not cut into the shield (Figure 5) but the energy was dispersed by the rolled edge rolling further and denting (rather than being cut). It is notable that on the shield from Barry Beg, there is a wire running inside this rolled edge which is c. 2.3mm in diameter, making the rolled edge of 5.3mm external thickness almost solid metal.

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

Spear thrusts and throws sometimes penetrated the face of the replica copper shields, but the spear never passed through the face of the shield to represent a threat to the user, an effect similar to that observed on the Nipperwiese shield from Long Wittenham (Needham 1979: 115). Likewise, bronze- and flint-tipped arrows shot from a 35lb bow at 10m distance did not penetrate through any of the shields. The area of the shield which most suffered in the case of the copper pieces was the boss area--when strikes landed here they caused more significant damage and heavily dented the boss (Figure 6), indicating that organic padding or leather gauntlets would have been likely accessories for the shield hand. It is notable that on the shields from Lough Gara and Barry Beg in Ireland, the boss is completely missing. This may be the result of combat use rather than post-depositional activity, though it is difficult to tell given the absence of the damaged portion. The Barry Beg shield has two other areas of damage caused by weapons thrusts, and at least two areas (now partly corroded) at the edge of the shield which appear to have been damaged by cutting attacks. It is very likely that this damage may be combat related, though the non-combat damage on the South Cadbury (Coles et al. 1999) and Thames shields (Coles 1962: 187, Yetholm no. 2) are reminders that not all apparent damage to a defensive weapon came during combat.

Discussion

It is risky to make assumptions about the 'value' of shields in Bronze Age society based on our appreciation of their workmanship or aesthetics, much as we cannot do so for contemporary swords as we know these were often simply recycled for their intrinsic metallic worth (Bradley 1990: 12). Starting with pre-fabricated copper sheet and with modern hammers and saws, it was possible for the author to manufacture a piece similar to the Lough Gur shield in seven hours. Given the greater expertise of ancient smiths and the use of specialised tools and formers, from ingot to finished product may have taken as little as three or four days in the Bronze Age. This is less rime than a contemporary sword, and requires less precise skills, from my experience making both weapon types, so it would be facile to argue that only the more powerful, wealthy or influential warriors had access to bronze shields. We must therefore be careful that the aesthetic impact of the shield on the modern observer, coupled with the comparative rarity of their survival, does not create a false sense of the value of these objects in prehistory. It is difficult in this sense to consider them a signifier of differential access to resources, and consequently emblematic of ranking within or beyond a peer-group of warriors.

On the other hand, the wooden shield from Cloonlara was a very heavy lump of wood with little intrinsic value, the product of a minimal investment of labour and was nonutilitarian as a weapon. It would be fair to say that in this case, we have an organic shield which was manufactured for a purpose other than combat and that in this sense it was symbolic representation of a shield, without venturing into the question of to what end this would have served socially. The wooden Annadale shield on the other hand had clear combat potential in purely mechanical terms. The story of the role of the shield in society is more complex than such a division between 'definite weapon' versus 'symbolic object' and this emphasises that we cannot simply use construction material to underscore categorisations or contexts of use.

Conclusion

It has been demonstrated that bronze shields were potentially effective in combat, while accepting that not all examples which survive today were designed to this end. Whatever their materials, shields of different sizes were used in different ways (Figure 7). While the Athenry-Eynsham and Yetholm shields were each made from bronze, their modes of use were very different indeed. It is true that the sheet metal shields were likely to fail through protracted combat use, bur this goes too for the wooden and leather shields, and one in six of the Irish swords have breakages which rendered them useless (Eogan 1965; Molloy 2006). Weapons could and did fail in battle and their mechanical limitations would have affected the modes of combat that prevailed in the Bronze Age. A holistic approach to the study of Bronze Age weaponry should place all categories of weapon in an integrated framework, as they were originally designed to work with and against each other in Bronze Age society. This does not denigrate the role of shields as symbolic items, bur underlines the fact that symbolism and display of this nature must be referring to a well understood martial reality which empowered the symbols with effective meaning. Thus we do not need to take an 'either/or' approach to the functions of shields based on their fabric, but can appreciate them as being functional and meaningful on many levels.

[FIGURE 7 OMITTED]

Acknowledgements

This research was carried out under funding from the Irish Research Council for Humanities and Social Sciences. I am much indebted to Marion Uckelmann for sharing important data on all European Bronze Age shields with me prior to its comprehensive publication in the Prahistorische Bronzefunde series, and for many helpful recommendations for improving this article. I am also very grateful to Joanna Bruck for her advice on an early draft of this paper. Thanks are also due to Anthony Harding, Alan Peatfield, Kristian Kristiansen, Philip de Souza and the anonymous reviewer of this paper who have provided much appreciated help and advice in this undertaking.

Revised: 5 January 2009; Accepted: 5 May 2009; Revised: 11 May 2009

References

AVILA, R.A.J. 1983. Bronzene Lanzen- und Pfeilspitzen der griechischen Spatbronzezeit. (Prahistorische Bronzefunde V/l). Munich: C.H. Beck.

BRADLEY, R. 1990. The passage of arms. Cambridge: Cambridge University Press.

BRIDGFORD, S. 1997. Mightier than the pen? (An edgewise look at Irish Bronze Age swords), in J. Carmen (ed.) Material harm: archaeological studies of war and violence: 95-1l5. Glasgow: Cruithne Press.

--2000. Weapons, warfare and society in Britain 1250-750 BC. Unpublished PhD dissertation, University of Sheffield.

BURGESS, C. & S. GERLOFF. 1981. The dirks and rapiers of Great Britain and Ireland (Prahistorische Bronzefunde IV/7). Munich: C.H. Beck.

CATLING, H. 1961. A new sword from Cyprus. Antiquity 35:115-23.

COLES, J.M. 1962. European Bronze Age shields. Proceedings of the Prehistoric Society 28: 156-90.

COLES, J.M., P. LEACH, S. MINNITT, R. TABOR & A. WILSON. 1999. A Later Bronze Age shield from South Cadbury, Somerset, England. Antiquity 23: 33-48.

COULSTON, J. 2007. By the sword united: Roman fighting styles on the battlefield and in the arena, in B.P.C. Molloy (ed.) The cutting edge: studies in ancient and medieval combat: 34-52. Stroud: Tempus.

COWEN, J.D. 1951. The earliest swords in Britain and their origins on the continent of Europe. Proceedings of the Prehistoric Society 17: 195-213.

-- 1955. Eine Einfuhrung in die Geschichte der bronzenen Griffzungenschwerter in Suddeutschland und den angrenzenden Gebieten. Bericht der Romisch-Germanischen Kommission des Deutschen Archaologischen Instituts 36: 53-155.

-- 1966. The origins of the flange hilted sword of bronze in Continental Europe. Proceedings of the Prehistoric Society 32: 262-312.

DAVIS, R. 2006. Basal-looped spearheads: typology, chronology, context and use (British Archaeological Reports International Series 1497). Oxford: Archaeopress.

EHRENBERG, M.R. 1977. Bronze Age spearheads from Berkshire, Buckinghamshire and Oxfordshire (British Archaeological Reports 34). Oxford: British Archaeological Reports.

EOGAN, G. 1965. Catalogue of Irish bronze swords. Dublin: National Museum of Ireland.

--1983. The hoards oft he Irish Later Bronze Age. Dublin: University College.

HARDING, A. 2000. European societies in the Bronze Age. Cambridge: Cambridge University Press.

--2007. Warriors and weapons in Bronze Age Europe. Budapest: Archaeolingua.

HEDGES, R.E.M., R.A. HOUSLEY, C.B. RAMSEY & G.J. VAN KLINKEN. 1991. Radiocarbon dates from the Oxford AMS system (Archaeometry datelist 17). Archaeometry 35: 305-26.

-- 1993. Radiocarbon dates from the Oxford AMS system (Archaeometry datelist 23). Archaeometry 39: 247-62.

HOCKMANN, O. 1980. Lanze und Speer im spatminoischen und mykenischen Griechenland. Jarbuch des Romisch-Germanischen Zentralmuseums, Mainz 27: 13-158.

KILIAN-DIRLMEIER, I. 1993. Die Schwerter in Griechenland (ausserhalb der Peloponnes), Bulgarien und Albanien (Prahistorische Bronzefunde IV/5). Stuttgart: Steiner.

MACHIAVELLI, N. 1560. Arte of Warre (translated by E Whitehorne, 1905). London: David Nutt.

MATHIEU J.R. & D.A. MEYER. 1997. Comparing axe heads of stone, bronze, and steel: studies in experimental archaeology. Journal of Field Archaeology 24: 333-51.

MOLLOY, B.P.C. 2006. The role of combat weaponry in Bronze Age societies: the cases of the Aegean and Ireland in the Middle and Late Bronze Age. Unpublished PhD dissertation, University College Dublin.

--2007. What's the bloody point: swordsmanship in Bronze Age Ireland and Britain, in B.P.C. Molloy (ed.) The cutting edge: studies in ancient and medieval combat: 90-111. Stroud: Tempus.

NEEDHAM, S. 1979. Two recent British shield finds and their continental parallels Proceedings of the Prehistoric Society 45:111-34.

NICOLLE, D. 1992. Romano-Byzantine armies fourth to ninth centuries. Oxford: Osprey Publishing.

OSGOOD, R. 1998. Warfare in the late Bronze Age of north Europe (British Archaeological Reports International Series 694). Oxford: Archaeopress.

RAFTERY, B. 1982. Two recently discovered bronze shields flora the Shannon basin. Journal of the Royal Society of Antiquaries of Ireland 112: 5-17.

RAMSEY, W.G. 1989. Middle Bronze Age weapons in Ireland. Unpublished PhD dissertation. Queen's University Belfast.

--1995. Middle Bronze Age metalwork: are artefact studies dead and buried?, in J. Waddell & F. Shee-Twohig (ed.) Ireland in the Bronze Age proceedings of the Dublin conference, April 1995: 49-62. Dublin: The Stationery Office.

RIOVSKY, J. 1996. Die Lanzen, Speer- und Pfeilspitzen in Mahren (Prahistorische Bronzefunde V/2). Stuttgart: Steiner.

SIDDORN, J.K. 2000. Viking weapons and warfare. Stroud: Tempus.

TALHOFFER, H. 2000. Fechtbuch (translated and edited by M. Rector). London: Greenhill books.

UCKELMANN, M. 2006. Schutz, Prunk und Kult--zur Funktion bronzezeitlicher Schilde. Anodos: Studies of the Ancient world 4-5: 243-9.

--Forthcoming. Die Schilde der Bronzezeit in Nord-, West- und Zentraleuropa (Prahistorische Bronzefunde III/4). Stuttgart: Steiner

UNDERWOOD, R. 2001. Anglo-Saxon weapons and warfare. Stroud: Tempus

--2007. The early Anglo-Saxon shield: reconstruction as an aid to interpretation, in B.P.C. Molloy (ed.) The cutting edge: studies in ancient and medieval combat: 134-44. Stroud: Tempus.

VAN WEES, H. 2004. Greek warfare., myths and realities. London: Duckworth.

WADDELL, J. 2000. The prehistoric archaeology of Ireland. Wicklow: Wordwell Press.

Barry Molloy, UCD School of Archaeology, University College Dublin, Belfield, Dublin 4, Ireland (Email: barrymolloy@gmail.com)