

What’s the Bloody Point?:

Bronze Age swordsmanship
in Ireland and Britain
Barry P.C. Molloy

introduction

Weapons are tools designed to inflict bodily harm on another being, to injure or
kill through force or finesse. The power of a sword to inflict harm on another,
even to take a human life, can be seen as the root of its symbolic potency in
ancient societies. This was a weapon possessed of qualities both highly esoteric
and brutally mundane, one which was to be celebrated for over three millennia
as the weapon par excellence of the warrior. The earliest forms of this weapon
in Europe emerged sometime around - bc, and it quickly achieved
widespread popularity in its various manifestations throughout the continent.
Thus it was far back in the Middle Bronze Age that the long lineage of the sword
was established.
In the centuries following this first introduction of the sword to the battlefields
of Europe, the weapons panoply of the Bronze Age warrior grew ever more
rich in style and variation. This near pre-occupation with martial arts, violent
symbolism and the activities of combat can be seen to permeate the social world
of this transformative epoch in human development. This paper will explore
how these martial developments evolved in the far northwest of Europe in the
islands of Ireland and Britain.
When analysing bladed weaponry, we can speak of the process of their
manufacture, their role as prestige items, as votive offerings or the importance of
their typological or metallurgical affinities and the communications networks that
these relate to. It is rare, however, to have a study dedicated to their use as tools
of violence. This is all too frequently restricted to simple ‘lip-service’ mention
in some articles, as Peatfield notes, because when they are referred to, ‘… the

90
 what’s the bloody point?: bronze age swordsmanship in ireland and britain

fearsome martial capacities of blades … are left curiously implicit’ (Peatfield :
 and see Kristiansen ).Yet this story of violence is one which is central to
their development, their very raison d’être, and it is one that is essential to explore
as it is inextricably linked to any other roles they may have played in society.
In this paper, I will not be writing about aspects of their non-martial or socio-
symbolic roles, this has been the subject of many excellent studies, and would
be beyond the scope of this paper (see for example Bradley, , ; Ramsey,
, ; Barber, ; Fontijn, ). I will write of just one story, one facet of
their social role that their extant material remains gives us close insight into, one
that the artefacts themselves bear physical testimony to through their scars and
ruptures. That is the question of how were these weapons wielded in the hands
of those ancient warriors who desired, perhaps even coveted, them so much?
The beginnings of this human story lie in the Early Bronze Age of Ireland,
where we have clear evidence for the use of weapons of interpersonal combat
(see O’Flaherty, this volume). It was during the Middle Bronze Age that weapons
specifically designed for killing human beings began to dominate the material
record of metalwork. Prior to this epoch, save the ‘halberd’, the weapons used in
warfare were those which people were familiar with from hunting – weapons
with which there was a more direct translation in the act of killing from animals
to men. With weapons such as Middle Bronze Age swords, the proximity and
sensual experience of killing was dramatically altered, and the social context of
killing can also be seen to shift accordingly. The directness of the physical and
emotional experience of the act of killing is accentuated by using purpose built
tools (see Molloy and Grossman, this volume).

middle bronze age swords and daggers

A discussion of definition must of necessity precede a discussion of use, hence I

will outline first some considerations which need to be adopted before analysing
the evidence for modes of use. From the tiny, simple double-edged knives of
the beginning of the Bronze Age, a wide variety of forms evolved. Some of
these were adapted further by the sixteenth century bc, with the lengthening
of the blade beyond what could be comfortably be called a dagger (see Burgess
and Gerloff  and Ramsey , ). The sword in its most archaic, even
embryonic form, had arrived.
In archaeological literature spanning back well over a century, these weapons
have been variously called daggers, dirks and rapiers with criteria for differentiation
between these based largely on the length of any given weapon. As will be discussed
below, these are light and thin weapons with a very wide variety in metric traits.

91
the cutting edge

They are characteristic of the Middle Bronze Age in both Ireland and Britain and
have broad similarities with other continental pedigrees (Burgess and Gerloff ).
As may be clear to some readers, the terms we use for these early weapons – dirks and
rapiers in particular – are directly borrowed from later Renaissance nomenclature.
These terms are well known to prehistorians and are often acknowledged
as being problematic, but it has to be stressed emphatically at this point that
they confuse further an already difficult period of human history, especially for
scholars of weapons from other periods. This is because they inappropriately
imply very specific modes and contexts of use which are not based on a critical
understanding of their forms and functions but on the recycling of antiquarian
thinking of the nineteenth century – a cumulative sequence of errors that this
paper will seek to redress. As difficult a path as it may be to follow for the
prehistorian, it will be argued that we should abandon these borrowed robes
in favour of the less functionally-loaded terms of Middle or Late Bronze Age
swords and daggers, as will be discussed presently.

mechanical structure of the weapons

It has been widely argued that these weapons were comparatively unserviceable
and inadequate for interpersonal combat, most especially in the context of a
multi-opponent melee. They are considered to have been used for stabbing
actions alone and not to have been suitable for making attacks using the blade’s
edges to cut (Ramsey : ; Waddell, : , Burgess and Gerloff ,
Barber : , Harding : , Osgood : ). However, it will be
argued that the design of the weapons themselves, the results of experimental
test cutting with accurate replicas along with the actual damage evident on
the surviving weapons all combine to give a very different picture. It will be
demonstrated that they were versatile and functional weapons for use on the
field of battle, and that they were most certainly not objects of purely ritual
significance or limited to a single specific context of martial use.
One primary factor behind the understanding that Middle Bronze Age swords
were used primarily for thrusting is the apparently weak hilting mechanism utilised
(Osgood : ).The hilt plate at the base of the blade tapers from the shoulders
back towards the end of the butt with a very noticeable angle (typically moving
from c.-mm thickness at the shoulders to just .-.mm at the end), and rivet
holes were pierced through this area (usually numbering two, but there are some
variations).This system of hilting the blade meant that the rivets passed through the
thinnest part of the blade – a factor which has sometimes been considered to be
a design flaw. While a simple thickening of this area would have made it stronger

92
 what’s the bloody point?: bronze age swordsmanship in ireland and britain

 Middle Bronze Age sword with organic hilt from Shower, Co. Tipperary. Courtesy of the National
Museum of Ireland

and would have been perfectly possible technologically speaking, this alteration
was only undertaken in the final years of the many centuries these weapons were
in use. We must consider then why this taper and thin section were retained for so
many centuries when they were less robust than was technologically possible.
The hilts on the vast majority of these weapons were manufactured from organic
materials. These fully enclosed the butt portion of the blade – that is they covered
the base (-mm of it), part of the two faces, and the two sides of the butt.
Organic materials such as horn, antler, hardwood or bone are naturally springy and
provide excellent grip when in compression. Given the wedge shape to the butt of
these weapons (terminating in that very thin section noted above), we can begin to
see a purpose to this design when it comes to the practicalities of establishing a firm
hilting mechanism. A cut slightly narrower than the shoulder thickness of the butt
of the blade would have been cut into base of the handle. This would have been
wide enough that the thin base could fit in easily, thus the graduating expansion of
the butt would have allowed the bronze blade to be wedged firmly into this slit in
the handle.Thus a firm friction grip is formed which may have been augmented by
glues or resins.This, it would appear, is the primary mechanical bond between blade

93
the cutting edge

and hilt, and the rivets are in fact a secondary or auxiliary element.The function of
the rivets is to pinch the sides of the handle tight around the blade and to maintain
this tightness, hence their frequently large diameter of mm+. They would also
serve to minimise any lateral movement which would weaken the friction bond.
Further supporting evidence for this might be seen in the metal hilted dagger from
Belleek (Burgess and Gerloff , No. ), which clearly had an organic element
wedged between the outer metal hilt and the butt of the blade contained within.
There are examples where a sliver of metal survives in place after partially breaking
free when being drilled (Molloy ), indicating that some (if not all) of these
weapons had the rivet holes drilled when the hilt was already in place. We should
also note that only about one in six ancient pieces displays significant damage to
the hilting system, a percentage not that far removed from the swords of the Late
Bronze Age in Ireland (see Eogan  and Molloy ).

mechanics of use

These weapons tended to be slightly blade heavy, but given that the vast majority
would have weighed below g even when hilted (see Molloy ), this does
not create significant problems either for controlling the weapons or risking
injury to the wrists. As noted above, the blades frequently have distinct cutting
edges present as well as a sharp point, and edges only occur on a sword for two
reasons:

• To inflict a wound on an opponent through a cutting or slicing motion using

the blade’s edge only.
• To assist in the penetration and removal of the point in a stabbing action as
they cut as the blade enters (thus increasing the size of the wound) making
this attack more effective, and they also allow easier withdrawal of the blade
from the flesh.

For this latter action the area near to the point need only be sharp, although
on weapons examined by the author the sharpness most frequently extended
throughout the entire blade. For a sword, there are a number of basic elements
common to most forms, and Amberger (: ) has detailed three ways in
which an edged-weapon can be used to cut flesh and/ or bone:

 Penetration by incision and percussion

 Penetration by incision and laceration
 Percussion

94
 what’s the bloody point?: bronze age swordsmanship in ireland and britain

Swords of the Middle Bronze Age are characterised by the second of these cutting
process, incising and lacerating, or more colloquially ‘slashing’ – i.e. the edge
slices the flesh as it passes along it. The blade needs to be drawn along the target
in a linear action allowing the sharpness of the blade to cut; the longer the draw,
the deeper the cut. While percussion is not the primary force brought to bear,
the blade is ‘whipped’ at the target to slash-impact and in the process the blade
is drawn along the target. The slicing and impact are therefore not performed
as two separate actions initially, but the draw of the blade with pressure from
the arm may continue briefly after the initial impact force has dissipated. This
is generally referred to as a ‘draw-cut’ or a ‘pull-cut’. More shallow cuts can be
made by the edges in this manner as the blade is thrust forward or retracted on a
thrust or cut.The role of thrusting with these, as with most swords (see Clements,
this volume), would have been integral to their intended mode of use, and indeed
some like the aforementioned ‘Lissane rapier’ may have been designed with
this as the primary mode of intended attack. However, most weapons were not
restricted to either one or the other of these modes of attack. The shorter swords
and daggers in this broad class would have been less effective at cutting than the
examples of around mm+ in length (when hilted), given that the blade would
not have adequate draw length along the target. This is not to say that they were
completely ineffective in this regard, because the edges could still inflict severe
cuts to areas of exposed flesh when used for short, swift cutting attacks.
Many of these strikes are not likely to have been fatal immediately, the main
goal would be to slice muscles, tendons or arteries seeking to debilitate rather than
sever a limb clean off (see below).This is important because with the weight of the
swords of the Middle Bronze Age, it would be not be possible to engage in more
heavy percussive cuts for two main reasons. Firstly, these weapons are extremely fast,
being capable of easily making two cuts in a single second when power is applied
from the elbow (and shoulder to a limited degree). If the blade was to be drawn
further back to make a blow incorporating shoulder and upper body strength, this
would dramatically slow down the speed of attack, thus compromising one’s own
guard to too high a degree. This leads into the second reason for the use of light
percussive force, and that is that these weapons were too light to actually provide
a heavy force of impact. If the blade is used in a more open arc using the upper-
body, the effective force of impact is only marginally increased as the light weight
will not particularly aid in increasing power and velocity.
The relatively short length of these weapons would mean that a user would have
to be relatively close to an opponent to land such a strike. Therefore, a heavy blow
would not provide a significant strategic advantage, would not inflict a much more
effective strike even if landed and would significantly increase risk of injury to the
user. Furthermore, while the hilting device was perfectly efficient for the manner

95
the cutting edge

of use discussed and determined through experimentation (see below), clumsy

overbearing strikes dramatically increase the risk of damage.The comparatively light
force applied in the typical use of these weapons would not have represented a major
risk to the structural integrity of the weapon. Of course some weapons did fail and
no doubt the user suffered the consequences, but this is simply the price of warfare,
a price still paid today when hi-tech weapons fail to perform or malfunction.
For these early weapons, sword-on-sword clash would not have been in the
form of the grand theatrical blows of Hollywood, but would have involved far
more subtle control of the weapon, and importantly of the opponent’s weapon.
Instead of heavy strikes and dead-blocks being made in attacks, one would rather
envisage draw-cuts being made alongside lethal thrusts. Defences would have
been of a more deflective nature through parries, stepping in to stifle blows or
simply avoidance of strikes. Blades are redirected through manipulation rather
than brute force and would more usually be slid along each other in contact
situations rather than struck squarely against one another. The extent of edge
damage on the Irish weapons examined by the author corroborates this – while
edge damage is present on a great many pieces, it is typically quite light and
infrequently distributed. When the author tested replica bronze swords against
each other, even light edge-on-edge impact resulted in clear damage to the
weapons on each and every strike (Molloy ).

testing middle bronze age swords

Two replica Middle Bronze Age swords were used to conduct a series of tests to
determine the efficacy and likely modes of use.These tests involved using the swords
against straw test-cutting pieces, against replica body armours (colour plate ) and
the body of a recently slaughtered pig at the Ashtown Food Centre in Dublin.
The pigs for test-cutting were two nine-month old females.They were suspended
by the hind legs and the forelegs were held in tension by bungee cords in a horizontal
position roughly at my elbow height (thus intended to represent cuts to the forearm
of a human opponent).The cutting took place within half an hour of the slaughter to
ensure that rigour mortis did not unduly influence the texture of the flesh and that
the bones were also in suitable condition (they begin to harden soon after death).
The reason for inducing tension in the limbs was to best simulate the resistance
to movement one would expect with a living subject. The cuts were thus made
to the inside portion of the leg, primarily in the area of the lower leg and
upper trotter. Due to the necessities of the facility being used, the pigs had been
eviscerated prior to testing, and this would have had an effect on the thrusting
test aspect of this project. However, the rib cage is the most difficult part to

96
 what’s the bloody point?: bronze age swordsmanship in ireland and britain

penetrate with a bladed weapon and while penetration of the soft internal organs
would have affected the depth of penetration by the swords, it would certainly
not have increased the resistance to a degree which would influence the validity
of the exercise. All test cuts were performed by the author and the proceedings
were recorded by video camera and digital still cameras.
The test-cutting mats used are a traditional Japanese medium for practicing
with swords, and are prepared by soaking them in water for three days, drying for
a further day and then they are rolled tightly together and bound. This broadly
represents the resistance of flesh to cutting with a sword, as corroborated when
compared to efficacy against the pig flesh.
The armours tested against were a piece of baked and waxed leather, a piece
consisting of ten layers of linen soaked in animal hide glue and a piece of .mm
gilding metal (a copper alloy similar to bronze sheet). These were attached to a
test dummy which was fixed on top of a work bench, thus allowing a degree of
movement so that the blades were not striking a completely dead target.
A replica Group IV dagger/short sword (after Burgess and Gerloff ) supplied
by Andrew Walpole and a Group II sword provided by Neil Burridge of Bronze
Age Craft were used for the testing.The former was hilted in deer antler, and when
hilted measured mm (mm unhilted) and weighed g.The edges had a thin
depth before swelling out to form the central thickening or midrib, thus although
sharpened, they only had a thin maximum cut-depth before cutting was impeded
by the expanding angle of this central area. The latter was initially hafted using
pine and subsequently with antler by the author, and was mm long (mm
unhilted) and weighed g. The edges were hardened on this latter weapon by
hammering along a line from mm from the edge down to the edge itself, and the
weapon was then sharpened (the hardening technique is discussed further below).
When using the Group IV sword for cutting against the test mats, it was only
capable of cutting one or two layers, which would constitute the outer skin on
living tissue.This assessment was borne out when test cutting on the pig cadavers,
with the blade easily cutting the skin, but having no significant impact on the
underlying muscle tissue. When the blade was used for thrusting at the chest of
the animal, it easily passed through the rib cage at right angles to the ribs (hence
having to cut the bone in order to pass between ribs). Likewise, when tested
against armour, this weapon could easily penetrate both the hardened leather and
linen pieces when thrusting, but was incapable of penetrating the copper alloy. It
was clear that this short robust piece was most suited to stabbing, either by linear
or arced thrusting. Cutting, while possible, would have only created superficial
wounds to areas such as the face, neck and forearms.
The Group II weapon, despite being very thin in its cross-section (mm), was
considerably more versatile. Initial test-cutting on the mats resulted in the pine

97
the cutting edge

hilt breaking along the wood-grain, but when hafted using deer antler, there were
no further problems. Indeed, neither weapon tested had any visible mechanical
problems with the hilting mechanism despite being used very roughly during
testing in order to push the weapons to their functional limits.
When tested against the mats, this blade was capable of cutting seven layers of
mat when struck with force commensurate to its weight and drawn along the
target simultaneously. Cutting on a forward thrust, it was capable of inflicting
superficial damage of two or three layers. As with the above weapon, these levels
of damage were born out when testing against the pig. Cuts to the middle area
of the ulna, where the flesh was in the region of -mm thickness, proved
very effective. The blade was able to slice to the bone, cutting cleanly and easily
through the flesh, and on one test cut it sliced shallowly into the bone itself. If
this was compared to attack against a human target such as the forearm, the cut
would serve to slice through muscle, sinew and tendons potentially disabling
a limb or at the very least dramatically impeding its efficient use. This testing
evidence clearly shows that these weapons were, mechanically, able to stand up
to a cutting attack when correctly executed in accordance with their mechanical
properties, and that such strikes could cause serious injury.
As with all of the swords tested during this program of research, the Group II
sword was incapable of cutting through any of the armours set around the test
dummy’s chest area. For thrusting attacks, the weapon was very efficient, capable
of penetrating the pig’s rib cage with ease, and passed through the armour with
little problem. It was noted however, that when it passed through the armour and
hit the tougher material behind, the weapon sustained a slight bend, indicating
that thrusting attacks could potentially also damage these weapons. This is borne
out by repaired bends and surviving torsional warps on a number of ancient
pieces examined in the National Museum of Ireland, Dublin.

evidence of use on ancient pieces

An examination of the Middle Bronze Age swords from Ireland has proved
that many weapons from all four classes (see Burgess and Gerloff ), have
damage along their edges that were formed by force of impact with another
sharp object of similar hardness and edge angle – other weapons being the most
likely suspects (Molloy ).York’s examination of a sample of swords from the
Thames basin has proved that similar damage is to be found on the majority
of British Middle Bronze Age swords from this area and chronological setting
(York : ). For the Irish material examined by the author, it was rare to
find weapons exhibiting very deep cuts or impacts to the blade edges, although

98
 what’s the bloody point?: bronze age swordsmanship in ireland and britain

there was a wide variety in the forms of damage which were found. The most
frequent form of damage, found on over half of the weapons examined, was light,
sharp-angled nicks caused by impact off another blade. The depth of these nicks
typically indicated that the impact had not been of great force as the damage was
in essence superficial. Some blades were completely cut through and broken, but
the general lack of severe damage to other areas of the blade may indicate that
this was either deliberate destruction of the weapon or else due to casting flaws
and/or stress induced weaknesses caused during use. Significantly, there was no
damage apparent to the flats of any of the blades, an area of the blade advocated
for parrying and blocking in later literature (see Clements  and this volume).
What emerged as the general pattern, however, is that most weapons had some
evidence of combat use, but that this evidence was rarely dramatic and was
indicative of a mode of use whereby the weapons did not receive repeated
forceful blade-on-blade impact.
In the case of Middle Bronze Age swords and daggers, we have weapons that
have unambiguous evidence of use in combat against similar weapons, attested
by the surviving damage on the weaponry. The swords had a firm and functional
hilting arrangement (given their typically light weight and manner of use), two
cutting edges (in many cases hardened and sharpened), a cross-section that allows
the blade’s edge to slice but not cleave and a point with a broad enough angle
to resist breakage. The case for their effective use in combat is truly compelling.

the use of middle bronze age swords

We should therefore look to how these weapons could be used in combat, in

order to see their potential effectiveness. If they were used for cutting in the
manner described above we can see that hard blade-on-blade impact trauma was
not to be expected, and heavy, lateral jarring of the blade-hilt bond would not
be likely. Indeed, a heavy impact manner of fighting with bladed weapons had
entirely no precedent in this period – the robust swords of the Later Bronze Age
were still many, many generations away.
The recent dating of the Kilmahamogue shield former (Hedges et al., )
indicates the high probability that leather shields () were used in conjunction
with these weapons (as one might suspect), and consequently, defensive moves
would not always require blade-on-blade contact, as this could be performed by
the shields. It is certain that the warriors that used these swords in life or death
contexts would have been well aware of the strengths and weaknesses of their
weapons, and would have been loathe to risk using them in a manner which
could leave them unarmed and at the mercy of a foe.

99
the cutting edge

 The leather shield from Clonbrin, Co. Longford. Courtesy of the National Museum of Ireland

Middle Bronze Age swords need not necessarily have been the primary weapon
of choice on the battlefield, as there were a wide variety of spears in contemporary
use, some of which are better considered as pole-arms (requiring two-handed use).
While it is possible that swords were auxiliary weapons which were mainly used
after the loss of a spear in combat, it is nonetheless clear that these weapons were
suitable for use on the field of battle.The edges of the blades were capable of cutting,
but in the vast majority of cases they would have been useless for cleaving – their
weight and cross-section (raised central area/midribs) would severely hinder deep
cutting of the edge and the repeated stress such blows would place on the hilt could
prove detrimental to their integrity. The points were always sharp, typically -°,
and capable of penetrating flesh. If the weapon has these qualities and limitations,
the manner in which it was used can be broadly interpreted as follows:

• Strikes come from the elbow and shoulder, with the wrist controlling finer
movement, especially slicing cuts. The strength and reach of the blade were
not sufficient for strikes involving the whole upper body and/or waist.
• Draw cuts would be the intended form of edge-attack, these are executed by
slicing the blade along the target either on delivery or more likely on recovery

100
 what’s the bloody point?: bronze age swordsmanship in ireland and britain

of a stroke (i.e. pulling as opposed to pushing the blade along the flesh). This
would allow the blade to ‘bite’ and cut without excessively shocking the
structure through force of impact.
• Thrusts would also have to take into account the mechanical properties of
the weapon – attacks against hard targets such as the cranium or sternum may
have been avoided. The cutting edges would facilitate the withdrawal of the
blade from areas of soft tissue such as the abdomen or neck as they assist in
cutting the point free on withdrawal.

It may appear quite disturbing to discuss inflicting such grievous and agonising
harm on another human being, but this is what these artefacts were designed for,
and how they must be understood. Lacerations to the soft tissue of the arms and
legs are not merely an inconvenience to a person or the cause of nasty scarring
for future memory – it is possible to cut tendons and muscles, effectively disabling
a limb without physically cleaving it from the body. A stabbing attack to the chest
would have to penetrate the bone, sinew and cartilage of the breast palate, and this
could either break or trap a light weapon, without killing an opponent. For group
combat, the most suitable stabbing attacks are to the abdomen, groin, foot, forearm,
face (around the sinuses and eyes where the skull is weakest) and the neck.
A duelling stance such as that found in modern sports fencing would be
unlikely when using these weapons; while fighters in lethal combat must
minimise their potential target area, they still have to maximise their own striking
ability. A more frontal stance is typically required to enable the use of both hands
and the leading foot for offence and defence. Footwork in the Bronze Age must
have been as important as it was in any other period (Wagner & Hand, ),
as control of the distance between a fighter and his opponent is paramount in
trying to control a fight to one’s own advantage. The leading foot therefore
should be on the same side as the leading arm, whether this is moving forward or
backwards, shield or sword to the fore. With the light shields and comparatively
short reach of the swords of the Middle Bronze Age, it is likely that footwork
was swift and the pace of combat must have been equally quick.
Our knowledge of the demographics and regularity of Bronze Age warfare are
negligible, but based on the settlement evidence, in the Middle Bronze Age we
should not envisage large bodies of men numbering in the thousands regularly
congregating for combat. On these grounds, one could argue that warfare was not
undertaken by highly drilled infantry fighting in close formations. However, this
need not be taken to imply fighting was in the form of unruly or disorganised
skirmishing. We should envisage perhaps a loose system of organisation broadly
analogous to Viking or Anglo-Saxon warfare. While the independent prowess of
the elite warrior was no doubt of significant importance on the field of battle,

101
the cutting edge

the cohesion of a battle line and mutual co-operation of warriors is fundamental

to combat using hand-to-hand or shock weaponry. The wooden training swords
() from Ireland can be seen as evidence that prehistoric warriors were very
much concerned with training and development of their weapon skills.
On the battlefields of the Middle Bronze Age one may have expected to find
warriors wielding lighter throwing-spears, single-handed spears (probably used
in conjunction with a shield), heavy two-handed spears, light swords and daggers,
short-hafted large spearheads (), bows and arrows and perhaps even slings. Far
from representing a massively random assortment of potential weapons, this
variety indicates a rich martial tradition with a wide skills base in the effective
use of combat weaponry. One only has to think of the later mythological hero
Cuchulainn in the epic Tain Bo Cuailgne, making use of countless different
forms of spear, shield and sword as he defends Ulster against the many champions
of Medb and Ailill.

the late bronze age

A common belief about warfare in the Late Bronze Age is that the arrival of the
leaf-shaped sword brought widespread changes in the conduct of warfare. The
setting for its arrival therefore needs brief clarification before we can enter into
a discussion about the impact and the changes in combat techniques that can be
identified in relation to this.
A problem we face is that there has frequently been a simple belief in the
literature that because the new form of sword was heavier and more robust than
its predecessors, it was inherently superior. Such a pseudo-Darwinian evolution
of edged-weaponry is, however, most difficult to maintain (See Clements, this
volume and also Hamilakis : ). A skilled swordsman in any period, with a
sword of inferior quality, is going to have a distinct strategic advantage over an
unskilled swordsman wielding a superbly crafted one. It is plainly not realistic
to simplify the transition from Middle to Late Bronze Age swords as a simple
replacement of a poor design with a ‘better’ one, as will be discussed presently.
If we are to discard these unhappy base dualities of ‘better’ and ‘worse’ in the
transition from one artefact form to another, then we will need to investigate
how functional attributes relate to broader traditions of use (Dobres , de
Marrais ). This is especially the case when we speak of objects with such
a straightforward nature as swords. The transitional period from Middle to Late
Bronze Age swords is likely to have lasted decades, when we take in the entirety
of the British Isles, and it would certainly have been a socially dynamic process
rather than some form of ‘pseudo-event’ based change.

102
 what’s the bloody point?: bronze age swordsmanship in ireland and britain

Above:  The Late Bronze Age wooden sword

from Cappagh, Co. Kerry. Courtesy of the
National Museum of Ireland

Right:  Middle Bronze Age Irish warrior

with V-notch leather shield and Basal-Looped
spearhead set on a short shaft. Original illustration
by Jorrit Kelder

103
the cutting edge

If we consider that blade-on-blade contact was not frequently practiced (see

Molloy , and below) and that shields and spears were also fundamentally
important aspects of the warrior’s panoply, then we need to see the changes in
artefact form as part of a socio-martial as well as technological shift. The new
sword’s net effect on the overall conduct of warfare and related balances of power
in this transitional period would be likely to have been piece-meal and lacked
any immediacy. Warriors who had grown up with the martial art traditions of
the older weapons need not have openly embraced the new technology and the
changes it implied in modes of swordsmanship.This may have been a contributory
factor to the differential pace of change in different regions of the islands. In
saying this, we should also be very much aware that the existing combat traditions
conversely represented a ready environment for the reception of the new tradition
of leaf-shaped swords, when other social factors were favourable.
What is clear in any case is that there is no evidence for upheavals or broader
social changes associated with the adoption of this new design. While Burgess
and Gerloff state boldly that the change from Middle to Late Bronze Age swords
‘points inexorably in one direction: the collapse of a long established social order’
(Burgess and Gerloff : ), this is clearly placing too much emphasis on the
artefacts as significant agents inculcating social change. The changes which we
detect in the material record may appear to have a linear inevitability moving
from one form to another, but it is essential to see these changes in terms of
evolving traditions of martial arts praxis and not simply the casting away of the
‘old’ in favour of the ‘new’. This involved existing martial traditions developing
responses to the new style of artefact, and modifying their combat techniques
accordingly.

so what was so new about the late bronze age swords?

Having set forth some of the problems associated with the nature of this transition,
we can begin to analyse the character of these new weapons and discuss aspects
of the changes which they represented in swordsmanship and indeed combat
as a whole. A major difference with these new swords was that the tang of the
weapon was integral to the handle – that is, it passed the entire way through the
grip of the fist and emerged at the base of the hand.This meant that the structural
integrity of the joint between blade and handle was more secure than in the
previous tradition discussed above. This in turn allowed for the manufacture of
heavier weapons which affected the balance and control of the weapon. This was
because the full-tang not only altered the harmonics and geometry of the whole
weapon but also changed the actual grip used on the hilt itself.

104
 what’s the bloody point?: bronze age swordsmanship in ireland and britain

The most notable change, however, with these new swords was the distinct
leaf-shaped blade which swelled gently out from the hilt reaching its maximum
width approximately two-thirds the way from hilt to tip with a gently convex
shape. There was typically a slight but notable distal taper on these weapons,
with the shoulder being the thickest part and the area of the maximum width
also being the thinnest area of the cross-section (save towards the point itself).
The blades were typically heavier than the average Middle Bronze Age weapon,
ranging roughly from -g for the most part, but with occasional pieces
weighing up to and over a kilogram (Eogan , Molloy ). While many
Middle Bronze Age weapons were above g in weight, the majority fell below
this mark.
The Ballintober sword was one the earliest of the truly native forms of
Late Bronze Age sword, being manufactured solely in Ireland and (primarily
southern) Britain. These swords were hilted in the Middle Bronze Age tradition
of fitting a tang into a one-piece handle, possibly relating also to the Rosnoen
swords being imported from the Continent to Southern Britain. After the
initial transitional phase of introduction, the new form of weapon saw insular
adaptation in the form of the Wilburton and Ewart Park (Eogan Class ) swords.
These weapons constitute an eclectic mix of sizes and weights, bound together by
basic morphological, more than purely functional, traits. Within this class we find
weapons ranging from mm long weighing in the region of g to weapons
over mm long weighing three times as much. Given their long and quite heavy
proportions, the replica weapons tested in my research represented functionality
at the higher end of the mechanical scale, as will be discussed below.

testing

A replica Irish Class  sword from Bronze Age Foundry and a replica Ewart Park
sword from Bronze Age Craft were tested in the same manner as the Middle
Bronze Age swords discussed above. The former sword weighed g and was
mm in length, it had a wide edge angle and was sharpened but not work
hardened. The second sword weighed g and measured mm, and was work
hardened with Neil Burridge of Bronze Age Craft using a specifically designed
device manufactured from mild steel. The reason for testing these larger blades is
that they were representative of blade weights and forms found on most Classes
of Late Bronze Age sword (excluding the intrusive Gundlingen sword tradition),
thus being the most representative for these limited experiments.
When the Irish Class  sword was tested, it was capable of cutting with ease
to a depth of  layers on the test-cutting mats. This weapon was used to cut

105
the cutting edge

near to the shoulder joint of the pig where there was little thickness of flesh and
bone was reached almost immediately after breaking the skin. It was capable of
cutting into the ulna (c.mm) but not cleaving it, although the broad edge angle
mitigated against this weapon cutting deep into bone as it had to separate this
hard substance as it cut. When tested on the flesh of the foreleg, the weapon cut
c.mm of flesh (down to the bone and scoring the surface of it).
This weapon could easily stab through the organic armours tested, but was
incapable of cutting through any of them. Its high length to weight ratio would have
meant that percussive blows, although not cutting armour, would have concentrated
significant force in a small area potentially damaging underlying tissue.
In test cutting with the replica Ewart Park sword from Bronze Age Craft,
striking with the blade ‘dead’ in an axe like manner dented the test mats but did
not cut them, so it may have been possible to injure bones and tissue against bare
flesh if used in this manner. However, when the sword was used with the elbow
leading the cut and drawing the blade along the mats it cut between  and 
layers, clearly indicating the superior efficacy of this mode of use. Similarly, when
tested on the pig cadaver, it cut cleanly through the flesh and into the surface of
the bone in the lower foreleg section with thicker flesh and cut through many
of the smaller foot bones/metacarpals.
A cut to the upper foreleg cut the flesh and severed through the ulna
(c.xmm thickness) cutting further flesh and making a shallow cut to the
radius. The edge of the blade which hit the bone (on initial impact) was buckled
on two occasions, providing clear evidence that attacks to even an unarmoured
body can result in damage to the weapon. This may have been influenced by my
own inexperience at cutting flesh and bone, especially the rate of drawing the
blade, as part of the cut may be altered so as to not cause this damage.
For thrusts, the steady increase in the width of the blade mitigated against
its penetrative power, but it was still capable of penetrating over half-way across
the pig’s chest, a depth of over mm, just short of the maximum blade width,
which would of course be fatal. When tested against the body armours, it was
capable of scoring the leather and linen deeply, but with several attempts on
each, it could not cut through them. In the case of the leather armour, it ‘dented’
on impact resulting in the transferral of energy from the weapon to a smaller
area of the target, potentially injuring bones and muscle on a living opponent.
It was possible to thrust through the leather and linen armours relatively easily,
indicating that the point angle was well suited to penetrating both flesh and
armour. As with all of the other weapons tested, this made no appreciable impact
at all on the copper alloy armour. It should be noted that the body armour was
simulating that of the torso, and if cuts were made to a smaller target such as the
forearm, it is more likely that the swords would cut through at least partially.

106
 what’s the bloody point?: bronze age swordsmanship in ireland and britain

When used with the elbow leading an attack, as opposed to wider arc cuts
made from the shoulder, the blade strikes at a slightly acute angle to the target
(i.e. not perpendicularly along its length), and is consequently drawn across
the surface cutting though incision and percussion. This raises the point as to
the purpose of the swelling or leaf-shape of the blade, which is a characteristic
feature of the vast majority of Late Bronze Age swords in Ireland and Britain (the
Carp’s tongue swords of Britain being a notable exception). It has been argued
that this swelling was to provide extra weight for an ‘axe-like swing’ (Osgood
: ). However, the majority of blades are at their minimum thickness here,
getting progressively thicker as they progress back towards the hilt, thus creating
a distinct distal taper. It is highly unlikely therefore that the increase in width was
simply to incur an increase in weight.
I have argued elsewhere (Molloy , ) that it is the actual curve of the
blade which was the important feature, as it allows the blade to bite deeper as it is
pulled along a surface on impact, operating in much the same fashion as a curved
sabre blade. This broadening of the blade also serves the function of providing
better structural integrity at the centre of percussion than a straight edged blade
of similar length, by reducing flexibility while retaining a thin blade cross-
section. This allows a low cross-sectional profile ideal for cutting living flesh; this
is a highly beneficial feature as it produces less friction through separation when
cutting thus increasing the efficacy of a cut. The slight difference in edge angle
between the two replicas tested had a significant impact on their efficacy, with
the thinner edged (and bevelled) Ewart Park sword proving better at cutting than
the Irish Class  sword.

evidence for use

The author has conducted extensive analysis of all classes of Bronze Age
weaponry from Ireland, and it is clear that the majority of weapon types in that
country saw active use in combat. The edge damage on the swords is useful
initially for indicating if weapons had been used in combat, but it can also be
helpful in telling us the manner in which they were used (colour plate ).
Experimental combat using the above replicas and another range of swords
(further Irish examples and Aegean Bronze Age forms) was conducted to
compare the damage to that on ancient pieces (Molloy ). The vast majority
of damage on ancient swords which had not been deliberately ‘killed’ (see
Bridgford ) was in the form of superficial nicking. It was immediately
evident that this damage was rarely from blows struck with significant force,
and indeed represented forms of damage inflicted on the replica swords when

107
the cutting edge

well executed parries were undertaken. Damage from dead or static blocks
was extremely rare, indicating that these were actively avoided in the style of
swordsmanship practiced. As with the Middle Bronze Age swords there was little
or no evidence surviving for blocking or parrying using the flat of the blade as
opposed to the blade edges. Some of the damage found on the ancient swords,
interestingly, had correlates on the replica swords from damage inflicted when
striking through the pig flesh and impacting on the bone, and also from striking
the unrolled edge of a metal shield.
It is important to note that the frequently superficial nature of the damage
indicates an acute awareness of the material limits of the weaponry on the part
of the ancient warriors who used them in combat. Striking the edge of one
sword on another rapidly deteriorates the edges and risks causing failure of the
sword itself. The number of swords with broken tangs but little edge damage
may further indicate that these were being struck off shields more frequently
than other bladed weapons, as in this context much of the energy from the blow
is transferred back (with leverage) to the area in the centre of the hand, exactly
where a great many swords have failed (see Eogan ; Colquhoun and Burgess
; O’Faolain and Northover ).
The damage therefore tells us that many of the extant pieces saw active
combat, but perhaps more importantly, that they were used in a manner which
indicates a very clear knowledge of effective swordsmanship. While the term
‘martial art’ is often taken to indicate Asian fighting styles, in its more direct
meaning as a definable system of combat passed on by masters to students, we
can see that at least a loose system of martial arts was in effect in the Bronze Age
of north-western Europe.

mode of use

The grip used when handling these Late Bronze Age swords can either be
four fingers on the handle, or else the index finger can be wrapped around the
ricasso, aligning the trajectory of the blade’s edge with the natural trajectory
of the arm. The U-shaped depression in the hilt plates at the shoulders created
a comfortable place to locate the thumb, as it fits into the curvature here and
enhances control of the sword. When using this weapon, primary control and
power comes from the shoulder, with the elbow slightly bent and generally
‘leading’ the blow with the point of the elbow moving on roughly the same
trajectory as the blade edge. This allows the blade to be deployed in quite rapid
tight-arced cuts, comfortably cutting as many as three times in two seconds with
a sword of over g weight.

108
 what’s the bloody point?: bronze age swordsmanship in ireland and britain

The wrist could also be ‘broken’ to allow for rapid changes of direction of the
blade. However, if a full handle grip was used, the shoulder, upper body and waist
could all be employed in making a strike, thus increasing power and range but
decreasing the rapidity of directional changes. The trade-off here clearly being
an increase in combat space and force of attack (thus distance from the enemy)
against control and speed of recovery (increasingly important with the proximity
of an enemy’s weapon).
It should be noted that while we group these weapons according to typological
classes, the manner of use of weapons within each class varies dramatically in
accordance with the varying weights and lengths. A g weapon will clearly
have different combat considerations to an g weapon with a blade of nearly
double its length. We must therefore remain cautiously aware of the role of the
personal preferences of ancient warriors in the design of the various weapons
we analyse today.
As with the Middle Bronze Age swords, the evidence strongly indicates that
edge-on-edge contact was avoided in the style of swordsmanship practiced in
the Late Bronze Age. If they were used in group combat, this indicates that
shields were used in conjunction with swords as a necessary defensive weapon,
as one might indeed expect given the simplicity of the concept. Unfortunately
comparatively few shields survive from the Late Bronze Age in the British Isles,
but the few we do have indicate a marked variety in both size and materials used.
Examples of wood (Alder) have been found which were manufactured from logs
split lengthways from a mature tree (i.e. not a slice across the trunk) and carved
to shape. The example from Annadale in Ireland suggests that examples for
practical use may have been in the region of -cm diameter with a (present)
weight of just over .kg. There were also lighter shields of baked and waxed
leather in use in the Late Bronze Age, measuring around cm across in the case
of the shield from Clonbrin, Co. Longford.
Visually striking shields of bronze sheet would have accompanied these organic
shields on the battlefield, and these ranged in size from small bucklers in the
region of cm in diameter to larger ornate pieces of over cm in diameter. Early
investigative experiments in the functionality of metal shields used an example
which was .mm thick (Coles ) and the swords used against this cut through
it with ease. Consequently, it was argued by Coles that metal shields were purely for
display rather than combat purposes, an often cited belief in archaeological literature.
However, more recent experiments conducted by the author (Molloy ) show
that many of the shields of bronze were very much functional and serviceable
weapons. The replicas tested were manufactured from sheet copper (bronze sheet
being unavailable) and were based on Irish examples, ranging from c.. to as much
as .mm in thickness. In the context of this paper it must suffice to say that they

109
the cutting edge

withstood a variety of weapons attack, including swords, effectively (see Molloy

 for full discussion), and this will be the subject of a future publication.
In light of this broad range of choices in defensive weaponry, it is therefore
not entirely surprising that the swords of the Late Bronze Age rarely exhibit
evidence of repeated blade-on-blade contact. While re-sharpening was very
common and would have removed much of the damage caused during combat
on many swords, the pieces which do exhibit damage rarely do so to extreme
levels. Far from being bouts of senseless bashing and brawling, one can see in the
subtleties of design of the weaponry of the Late Bronze Age, that there was a
complex martial tradition in existence with its roots firmly located in the Middle
Bronze Age combat system.

conclusion

It is important in artefact studies to acknowledge that each and every individual

artefact came into being in the form that it is today because of conscious decisions
by past people, decisions voiced to skilled craftsmen and smiths who made the
vision a reality (See Kristiansen :  - ). In the case of weaponry, it is
impossible to divorce this decision making process from the field of violent intent.
Weapons fulfil many social roles – as status symbols with regard to wealth and
scope of influence for example. These are, however, secondary purposes which
could never have been conceived of if the artefact form never came to social
prominence through other means. If a given weapon (or even a particular form
of weapon) is to have a symbolic value, it must be recognised as being a weapon, it
must represent a tool that has visibly committed violence in the past and therefore
be imbued with a certain potential or embodied power. The functional analyses of
weaponry which this paper has explored, reveal far more than simply the ‘hows’ of
an artefact’s use, purpose and intrinsic significance, but are also extremely beneficial
in examining the ‘whys’ of its creation and social prestige.
Combat must be interpreted as action. It is the event(s) that take place
during warfare, it is human beings utilising their physical mechanical properties,
augmented by manufactured tools, to engage in activities with closely definable
objectives – to kill, maim, disable or otherwise defeat other human beings by
force and finesse. This world of action is partly fossilised in the forms and shapes
of the tools which were used to undertake these activities, and working from
the specific to the general we can begin to show how these artefacts reflect the
martial systems of prehistory. While the weapons may be less informative than
we might wish about the macro-environment and tactics of the battlefield, they
each tell a story of experiences of life and death in the Bronze Age.

110
 what’s the bloody point?: bronze age swordsmanship in ireland and britain

acknowledgements

There are many people to thank for invaluable help with this research which
is drawn from my PhD thesis. Firstly, I am most grateful to Alan Peatfield, my
thesis supervisor, who has offered me great guidance at every stage of my work.
I would also like to thank Mary Cahill and Margaret Lanin of the National
Museum in Dublin and Richard Warner and Sinead MacCartan of the Ulster
Museum for access and much assistance in examining the material covered in
this paper. Thanks are also due to Tony Kenny of the Ashtown food centre for
invaluable assistance with the program of testing, and to Ciaran Flanagan for
help with the testing process. Thanks to Charles Molloy for comments on early
drafts of this chapter and to Kristian Kristiansen, Joanna Bruck, Philip De Souza
and George Eogan for advice on various aspects of this work (all errors and
mistakes, as they say, are my very own!). I am also very grateful to Neil Burridge
of www.bronze-age-craft.com for his constant advice and enthusiasm for this
work, including the provision of several replica weapons. Thanks also to all on
the Ancient Weapons forum of www.swordforum.com for frequent, informed
and lively discussion on this topic. This research was carried out under funding
from the Irish Research Council for Humanities and Social Sciences.

111