Caesar’s Army

By MSW Add a Comment 45 Min Read
Caesars Army

The Republican period ran from a traditional start date
established after the last King of Rome in 509BC and ended after Caesar was
assassinated and Augustus became supreme ruler in 31BC. Detailed archaeological
and historical evidence for this period of Roman military history is somewhat
elusive, but we do have three authors who provide us with some evidence of the
structure and behaviour of the early Roman army: Livy, Polybius and, of course,
Caesar himself. Their descriptions are good as far as they go, but provide us
with only a narrow view of the army of the late Republic. The authors,
excepting Caesar, do not provide a contemporary view of Caesar’s army and tend
to focus only on an idealized image of the legions, with little or no attention
given to other units such as allies or mercenaries. The case is not much better
with regard to the archaeological evidence. Here we can find some suitable
contemporary evidence for the Roman army of the late Republic, but it is
extremely limited, focusing on a few camps in Spain and naturally Alesia.
Unfortunately, as most of this data was unearthed over a century ago, even this
evidence is patchy. Our understanding of Caesar’s army, therefore, is built
from a mix of contemporary and not so contemporary sources, scholarly research
and debate.

Clearly Caesar’s legions did not simply spring to life fully
formed. While they were raised as complete units, this arrangement came about
as part of an inherited system of behaviour. To understand the character of the
Caesarean legion it is worth taking a brief look at how the legions developed
from their foundation. In reviewing the development of the legion as a whole,
some of the relationships between Gaul and Rome will become apparent. The
legion, as we understand it, was the consequence of a long period of changes
absorbed in the wake of military defeat. Although the Roman legion is now seen
as the epitome of military genius, the reality was somewhat different; in fact,
Rome’s success as a military power seems to have been due more to its ability
to learn from its mistakes. Rome’s brilliance was that it could accept its
weaknesses and adopt the successful elements of its enemies, whilst all the
time placing these within a structured military system. In due course, the
legion developed from the propertied man’s annual obligation to fight to
protect his land during the summer months, to the requirement for all males to
fight lengthy warfare for a campaigning season (usually March to October) and
finally to professional soldiers paid to fight constantly over prolonged
periods of many years. This transformation was concurrent with the expansion of
Roman territory and the change from the protection of the local community to
the requirements for a standing army to protect a vast and diverse Empire.

The armies of early Italy were little more than war-bands of
infantry and cavalry, raised by local tribes or princes when required, and
fighting when the seasons allowed. Charismatic and powerful men led these
bands, the best of which could provide protection for the individual and an
opportunity for advancement. It could be argued that these traits were
enshrined in the system from this early date, as the patronage of wealthy and
powerful leaders can be seen in the armies of Rome throughout its history.
Caesar was certainly a general who instilled in his legions personal devotion
to him. Finally, the early system of ad hoc recruitment became more
standardized, in order to provide a more regular and predictable turnout of
men. The first Roman armies consisted of around 3,000 men and were organized on
the basis of tribal groups, called a legio or a ‘levying’. Each tribe
contributed 100 men towards the total force – the origin of the ‘century’ of
Caesar’s legion. The wealthy elites in the tribe were relied upon to supply the
cavalry for the army, as only they could afford expensive equipment and horses.
This group, the equites or knights, survived as a social group long after their
original function had vanished, serving as officers in Caesar’s army. In
essence these early armies fought in similar ways to most early European armies
of the period. Warrior bands would fight under the command of their local
leader in unstructured groupings, while commanders would range around the
battle, urging them on to fight and selecting suitable enemy leaders for
combat. It is interesting to consider that the Gallic armies Caesar faced were
very much on a par with these early Roman armies.

After the Roman Republic began in 509BC, the Roman army
became far more formalized, along the hierarchical lines society was taking.
Greekstyle Hoplite warfare was the most advanced military tradition at the time
and represented a significant advancement on earlier military tactics. In
Hoplite warfare, troops fought as a densely packed line of heavily armed
spearmen. Cohesion was the key to this style of fighting and so the
hierarchical nature of Roman society was replicated in the legion. Only those
men with property were allowed to fight, giving the men a sense of duty and an
interest in preserving the State. It is worth noting here that Roman citizens
were providing the entire range of troop types in the army – a situation that
was not to last. As Rome expanded the areas coming under its domination, there
also came a necessity for an expanded army. The increased wealth of Rome meant
that the number of infantry and cavalry centuries could also be increased.

By the fourth century BC, Rome’s experiences of fighting
with the Latin and Gallic tribes had revealed serious weaknesses in the Hoplite
form of warfare. The more flexible fighting styles of the Gauls and Samnites
often highlighted the sluggish and formulaic nature of Hoplite warfare. To
counteract this problem the Romans began to adopt less dense formations, which
in turn required a revision of their tactics and equipment. The Roman phalanx
was reorganized into fighting formations called maniples – literally
‘handfuls’. The overall result was slowly changing the legions from the rigid
defensive formation of the phalanx into a cohesive collection of flexible
offensive fighting units. The fourth century BC saw the entire Italian
peninsula come under the control of Rome and as the sphere of influence
increased, so did the embracing of allied Italian peoples into the Roman army.
On campaign it became common for allied soldiers to comprise about the same
number of troops as the Roman legionaries. Armed like the legionaries, these
allied troops formed up either side of the legions in alae or wings. From this
time on, allied or mercenary units came to be a significant feature of the
Roman army.

‘On the expedition he
[Marius] carefully disciplined and trained his army whilst on their way, giving
them practice in long marches, and running of every sort, and compelling every
man to carry his own baggage and prepare his own victuals; insomuch that
thenceforward laborious soldiers, who did their work silently without
grumbling, had the name of Marius’ mules.’

[Plutarch, Lives of
the Noble Greeks and Romans, 13]

The turn of the first century BC saw a significant change in
the character of the Roman army. This has often been attributed to just one
man, Marius, but this may be somewhat overstated and simplistic.
Transformations had been taking place over the course of the previous three
centuries, therefore Marius’ role is most likely to have been as a catalyst for
the changes already developing. The army was now spending long periods away
from home and it was becoming a genuine career choice with the possibility of gaining
wealth and land from the proceeds of victory. Previously, the Roman army had
been an organized militia, galvanized with discipline and training.
Increasingly, the army was becoming more specialized and developing a core of
professional soldiers as Roman society became geared for on-going war.
Pragmatism was required in the face of the demands put on the army by the
ever-expanding areas of land to be controlled. This also resulted in a
relaxation of the prescribed requirements for entry to the army, which had
gradually been eroded almost as soon as the legion was developed. The
acceptance of elements of the non-land owning populace into the army had only
been done in extremis before, but now Marius attempted to increase the strength
of the army by changing the property requirements for service. He allowed the
recruitment of proletarii, the landless citizens of Rome, making the
class-based system redundant. The state had for some time been supplying
equipment to the army and so standardization was already occurring. Hence by
Marius’ time, the army was equipped fairly uniformly and the complex strata
exhibited in the previous system was removed. Thus Marius was unwittingly
responsible for homogenizing both the equipment and structure of the legions,
by allowing the process to become formalized.

In accepting the landless into the army, the State also had
to accept the responsibility of arming the legions, which, up until then, had
been paid for by the soldiers themselves. After their service these landless soldiers
were left to fend for themselves, so while they were in the legions they were
willing to follow any charismatic leader that promised them land afterwards.
Gradually, a subtle difference in the mind-set of the legions emerged, from a
landed militia protecting their own Roman lands, to a professional army
dominating annexed foreign provinces. Soldiers began to develop greater loyalty
to their generals, identifying with their leaders who, like them, sought
personal enrichment rather than seeing the maintenance of the structure of the
State as the purpose of their fighting. Individual grants of land and money,
distributed after military successes and service, were now becoming
commonplace. The politics of the late Republic were becoming ever more competitive,
hence in the struggle for advancement senators succumbed to more and more
aggressive policies and expansionist pressure. This was the world that Caesar
made his own.

Caesar leaves out much of the detail of his army because his
audience was well informed about the character of the Roman army of the period.
Caesar does tell us that he was using a system of cohorts, a simplified version
of the previous maniple system. Marius is thought to have been responsible for
the change from maniples to cohorts but it is likely that both systems could
have been in place together for a period. When Caesar set out on his Gallic
adventure in 58BC the cohort system was well established. Manipular legions
were made up of three distinct lines of formation, each one equipped
differently and socially differentiated. The cohort legion did away with these
equipment complexities, replacing them with a flexible body of similarly armed
men. It is likely that the system was introduced to deal with the difficult
nature of warfare in Spain. Tactics had to be developed to combat both the
Spanish guerrilla warfare and the mountainous character of the landscape. The
system retained the three-line formation, with a strengthened rear rank as a
reserve.

Caesar formed his legions for battle into what is called a
triplex acies formation, being four cohorts in front, with two lines of three
cohorts behind. This formation was something of a compromise between being wide
enough to form a broad frontage and deep enough to have reserves. The middle
cohorts provided the reserve for the front cohorts, while the rear cohorts
could be used for outflanking the enemy or if the legion was attacked in the
rear. The basic building blocks of the legion were the eighty-man centuries,
each one commanded by a centurion. Caesar depicts the centurions as the
fundamental glue of his legion, providing harsh discipline and motivational
inspiration in equal measures. Six of these centuries would provide a 400-man
cohort: this played the role of an individual tactical unit on the battlefield,
ten cohorts together forming the legion.

In the last century of the Roman Republic the command of the
army was removed from publicly elected consuls. Only after their period of
office could they command – a circumstance that had the effect of further
breaking the army’s connection to a citizenry. Caesar tells us that most of the
legions for his Gallic campaigns were raised during the winter months, when
campaigning had ceased. As governor of three provinces, he inherited the four legions
that were stationed in them and these could be augmented by further recruitment
from the provinces. Initially it seems that Caesar would raise the legion out
of his own money, then, after gaining acknowledgment of the unit from Rome, the
legion would become the responsibility of the State to maintain. Caesar
enlisted troops into the legions of both citizen and part-citizen ‘Latin’
status, thus continuing the process of blurring the qualifications for entry to
the legions. By collecting these diverse groups into unified legions, and by
exerting direct personal control over them, Caesar was able to place himself as
the main focus of the legion’s loyalties, before that of any other – even the
State. This meant that his forces were extremely loyal to him: a factor that
was to play its role in the subsequent Civil War (49–45BC).

The officers of the legions were not trained. In fact, most
received their posts as part of their political career. Six tribunes were
placed as middle-ranking officers of the legion and these were generally young
and untested men of aristocratic birth, often lacking in initiative or bravery.
Caesar chose the tribunes personally, many purely on the basis of political
expediency and patronage. Above the tribunes was a quaestor, a junior senator
who oversaw an entire province and provided Caesar with finances. Overall
control of a legion was placed in the hands of a legatus. Caesar chose a number
of legati; usually former tribunes, they were senators from a variety of
backgrounds and experiences, but were usually politically motivated choices.

Over the course of the Gallic Campaign, Caesar increased his
legions from the four he inherited, to twelve at the height of the fighting for
the Alesia Campaign. Thus, Caesar could have had up to 57,600 men at his
disposal. However, this number was only the paper strength of the legion, based
on around 4,800 men per legion. The actual number would, more likely, be half
this amount by the campaigning of late summer 52BC. Battle casualties,
infirmity, exhaustion and general wastage from the previous months’ campaigning
would all have played a part in reducing this tally to less than 30,000 men at
Alesia.

The 30 miles of fortifications – making up the
circumvallation and contravallation at Alesia – may seem very large, but when
one considers that its construction was divided among almost 30,000 available
men, each soldier would only have had to dig around 5 feet of trench and
rampart in the six weeks it took to prepare. In this light, the defences do not
seem such a superhuman task. In fact, when one considers that it was usual for
Roman legionaries to build a camp at the end of the day’s march, the building
of these fortifications seems well within the capabilities of the average
soldier. In truth, the greater task would have been the logistics of the
operation, including the planning, design, organization, supply and
implementation of the construction. However, this is where the Roman military
machine came into its own. If the defences at Alesia are exceptional, this is
only because the management of the army was exceptional and equal to the
Herculean task. Motivation to create such an engineering feat was an important
factor in Caesar’s army. Infantry training tended to focus on physical ability,
including running, jumping, marching and building – clearly necessary for the
construction work.

Increasingly, espirt de corps was also encouraged. The
legions began to be individualized by number and by name. Added to this,
individual legions were picked out by nicknames, often recognizing their
exploits, and by use of awards and honours on their banners and standards. This
is not to say that the legions were uncritically loyal to their leaders. While
the soldiers were made to give oaths of allegiance that imposed legal and
religious constraints on them, the Roman army was still prone to revolts and
indiscipline was common, mainly over pay or ill treatment. Often rewards would
be granted to keep the legionaries content. Caesar would often bestow on his
soldiers a promotion as part of the system of rewards, especially for
centurions. At Alesia Caesar’s handing over of slaves and booty to his soldiers
after the defeat of Vercingetorix is a clear example of this. To this was added
the spoils of war, along with donativa, one-off payments made by the general in
gratitude of service. After the Alesia Campaign, Caesar decided to double the
pay to ensure the loyalty of his soldiers for the coming civil wars.

The remaining archaeological evidence from Alesia is
confined to the siegeworks themselves. Roman weapons of the late Republic are
scarce throughout Europe, most coming from siege sites in Spain. At Alesia
there is surprisingly little in the way of Roman military equipment and this is
likely to be due in part to biases in the excavation of material. In the main,
archaeologists have focused their attentions on understanding the form of the
Roman defences. This means the results have concentrated upon the character of
ditches that bordered the Roman circumvallation and contravallation. These, it
has been discovered, were filled with javelins and arrowheads, which make up
the predominant proportion of the total weapons discovered at Alesia. While
some Roman pila (the legionary’s offensive missile) have been found, most of
the other weapons are likely to be Gallic in origin. If one considers that the
ramparts were mainly the focus of incoming missiles from the Gallic army, then
outgoing Roman missiles would have been fired into areas further from the
ramparts, where little or no excavation has been undertaken. Similarly, all the
swords so far discovered seem to be of Gallic origin, and again this may not be
unusual, as Roman weapons would have been retrieved after the battle had
finished to be used in future campaigns, whereas Gallic ones would only have
been recovered if they were considered valuable.

‘Gaius Sulpicius …
commanded those who were in the front line to discharge their javelins, and
immediately crouch low; then the second, third, and fourth lines to discharge
theirs, each crouching in turn so that they should not be struck by the spears
thrown from the rear; then when the last line had hurled their javelins, all
were to rush forward suddenly with a shout and join battle at close quarters.
The hurling of so many missiles, followed by an immediate charge, would throw
the enemy into confusion.’

[Appian, History of
Rome: Gallic Wars, 1]

Archaeological evidence shows that the panoply of equipment
of the Roman soldier differed only in detail from that of the better-equipped
Gallic warriors – sword, long shield, helmet, mail armour and spear.
Nevertheless, their appearance was different enough for one soldier to comment
that soldiers in the distance were not Roman because of their ‘Gallic weapons
and crests’. The Romans’ equipment wasn’t the only thing that was different
from the Gauls; their fighting techniques varied too. After throwing their
pila, the legionaries would engage the enemy adopting a crouching stance and
using a juxtaposition of punches with the shield boss and stabs with a sword.
It is evident from sculptural and archaeological evidence that the majority of
Roman legionaries were equipped to fight in this style, with heavy armour and
equipment. A large number of weapons have been found at Alesia, almost 400 in total,
140 of which are either javelins or spears. Unfortunately, these could be
either of Roman, Gallic or German origin. Some of the spearheads must be Roman
but it is unclear which, as leaf-shaped spearheads were a common form used
across Europe at the time. However, using other sources of evidence, we can
piece together a picture of Caesar’s legionaries. In the late Republic the
average legionary would have a decorative tall bronze helmet with a short neck
guard and large cheek pieces, a derivative of Gallic styles. The form was
called ‘Montefortino type’ and was sometimes enhanced with horsehair or feather
trappings. These decorative helmets were beginning to be replaced by a more
easily produced, plain style of helmet called a ‘Coolus’ or ‘Buggenum-type’ helmet.
It is likely that both types were in use at Alesia. In general, legionaries
would wear a mail coat that reached nearly to the knees, which was hitched up
with a military belt. Some sculptures suggest that these coats could have
further mail reinforcing on the shoulders. Pieces of chest fastenings from
these mail coats have been found at Alesia. However, as the Gauls were the
inventors of this form of defensive equipment, defining whether these fittings
are Gallic or Roman is very difficult. All legionaries would also carry a long
curved wooden shield, strengthened by a vertical central spine and which had a
small bronze boss covering the handgrip. On a legionary’s belt there would have
been a stabbing sword on the right hip and a long dagger on the left. Both
these weapons seem to be copied from types common in Spain.

‘[Caesar’s] …
soldiers, hurling their javelins from the higher ground, easily broke the
enemy’s phalanx. That being dispersed, they made a charge on them with drawn
swords. It was a great hindrance to the Gauls in fighting, that, when several
of their shields had been by one stroke of the pila pierced through and pinned
fast together, as the point of the iron had bent itself, they could neither
pluck it out, nor, with their left hand entangled, fight with sufficient ease;
so that many, after having long tossed their arm about, chose rather to cast
away the shield from their hand, and to fight with their person unprotected.’

[Caesar, The Gallic
War, I. 25]

The one piece of the legionary’s equipment that can be
directly attributed to Roman invention was his primary offensive weapon, the
pilum. Pila were a form of spear with a long thin metal shaft and a small
pointed head. Although pila are a Roman invention, examples of similar types of
weapons were also developed in Gallic and German contexts. Among the huge
number of missile weapons found at Alesia, pila are the predominant Roman
weapons evident. Up until recently, only fragments of pila were discovered,
mainly coming from the foot of Mont Réa. But in 1991 a complete example of a
Roman pilum was discovered in Fort Eleven. The examples from Alesia are
characterized by long thin shanks and points that come in pyramidal,
leaf-shaped or lance-shaped forms. There are different ways in which pila were
connected to the wooden shaft: some were connected by a socket that fitted over
the shaft and was riveted to it; other pila shafts had a wide tongue that was
sandwiched between the shaft and riveted in place. A final type had a pointed
tang that was driven into the shaft and secured by a collar. Usually, a round
shaft was used for the socketed and collared pila, whereas a square shaft was
used for the tongued versions. It is likely that the socketed pilum was lighter
than the tanged pilum, this is because the tanged pilum was often weighted to
provide extra penetration on impact. After his excavations at Alesia, Napoleon
III had replica pila made and tested. The results showed that the reconstructed
examples could be launched 30m and penetrate wood 3cm thick. Heavier pila were
weighted with lead and could have been thrown up to 70m. It is likely,
therefore, that pila were thrown at the Gauls before they entered the defences
of the circumvallation and contravallation at Alesia.

The equipment available for use by the legionary was
extensive and the legionary himself carried much of it. The requirement that
legionaries carry all their equipment is attributed to Marius: hence the term
‘Marius’ mules’, although the likelihood is that this is a misattribution. The
soldiers had always been required to carry their equipment, a regulation that
was regularly flouted. It is likely that Marius simply reinforced a standing
regulation in an attempt to make the soldiers more self-reliant and less
reliant on a baggage train – part of creating a professional army. Vegetius
suggests up to 60 pounds of equipment should be carried during training and
Josephus claims each soldier carried a saw, a basket, an axe, a pick, a strap,
a billhook, a length of chain and three days’ rations. The rest of a soldier’s
equipment, his tents and so on, were carried with the baggage. Along with this
equipment, the Roman army took with it large numbers of servants and slaves,
who were usually not armed but had sufficient knowledge of tactics to be of
use. Sometimes these groups were added to the regular army to give the
impression of a larger force than was actually available, such as at Gergovia,
where they were mounted on horses to look like cavalry. Merchants and camp
followers would also be part of the train, providing services that would
otherwise be unattainable in foreign countries. Large numbers of followers and
baggage had the tendency to slow the column on marches and so attempts were
made to reduce these numbers. One way of reducing the reliance on camp
followers was the local requisition of supplies and the creation of storage
bases along the route of march. In Gaul, where Vercingetorix had instigated a
scorched earth policy, this was impossible to maintain and so it is likely
Caesar had more followers than he would have wished to have.

The legions also took on campaign with them numbers of artillery pieces. These could be used aggressively, either in open battles and sieges to provide preparatory fire, or defensively to protect camps and siegeworks. Roman artillery comprised various sizes of ballista (sometimes called a catapulta or catapult). The ballista was a torsion catapult that used two twisted skeins of hair or tendons to provide energy for a string that fired projectiles. The ballista was built in various sizes that fired anything from small crossbow-sized bolts to large cannonball-sized stones. Parts of ballistae occur occasionally on archaeological sites. At Alesia three bolt heads were discovered, betraying the presence of ballista there. Ballista would allow Caesar to cover most of the regions outside his defences with a combination of fire from the hills surrounding the defences and the ramparts of the circumvallation. The larger engines were probably fixed in place once built and reserved for siege work, whereas the smaller artillery was often broken down and moved on pack mules, but could also be erected on carts or wheels (carroballista) to make them mobile on the battlefield.

Caesar´s Germanic Cavalry

Caesar’s Allies

‘It was the practice
of the Romans to make foreign friends of any people for whom they wanted to
intervene on the score of friendship, without being obliged to defend them as
allies.’

[Appian, History of
Rome: Gallic Wars]

Throughout the Republican period, Rome relied heavily on its
allies for additional infantry and cavalry forces to make up for a lack of
manpower. In some cases these were specialist fighters recruited on an ad hoc
basis and in varying strengths from the locality. More important were the
allied light troops who were drawn from Mediterranean regions Rome had long
been in contact with and with whom they had the closest relations. These troops
were customarily raised for the duration of a campaign, which sometimes led the
Romans themselves to questions their allies’ quality and commitment. Allied
formations were usually under the control of an individual unit’s chief, and
were armed, equipped and fought in the particular unit’s traditional style.

Usually the allied contingent of the Roman army was of equal
or larger size than the legionary force and was usually formed along Roman
lines. Units regularly seem to have been about 500 or 1,000 strong and broken
down into either six or ten centuries. These could be arraigned with the main
army in the centre of the Roman line or placed on both wings. Sometimes the
more lightly armoured allied contingents were mixed with the heavier armed
legionaries to prevent the allies from being picked off. Along with the heavily
armed troops, many of the allies provided specialist light infantry troops,
such as archers and slingers. These units are likely to have been dressed
according to their ethnic origin, although there is no definitive evidence of
which units were at Alesia. Over forty arrowheads have been recovered from
Alesia and it is thought that some of the Roman forms of arrows with one and
two barbs are likely to have been used by allied Roman troops. Roman arrows
were manufactured from iron and were up to 7.8cm long and 2.5cm wide. Tests of
reconstructed bows suggest they would have had a maximum range of around 300m,
and so they would be able to fire at the Gauls beyond even the deepest of
Caesar’s defences. Slings were also used and a number of examples of slingshot
come from Alesia, most notably three with inscriptions on them. Reconstructed
slingshots have shown a range of up to 400m, easily enough to provide covering
fire from any of the hilltops around Alesia into the valleys below.

Throughout his campaigns in Gaul Caesar does not define the
constitution of his cavalry and so we cannot be certain about their number or
ethnicity. It is likely that at least some of the cavalry at Alesia were Roman.
At the beginning of the conflict, Caesar was also able to call upon friendly
Gallic tribes to provide him with Gallic cavalry and these were employed as
warriors, as well as scouts, guides, interpreters and messengers. However,
Caesar had always considered them unreliable, a belief which was confirmed once
Vercingetorix rebelled and Caesar lost the majority of his Gallic troops to his
rival. Some must have been retained however, if only in an intelligence role,
but by the beginning of the Alesia Campaign Caesar was forced to employ new
cavalry in the form of German mercenaries.

‘Caesar, as he
perceived that the enemy were superior in cavalry, and he himself could receive
no aid from The Province or Italy, while all communication was cut off, sends
across the Rhine into Germany to those states which he had subdued in the
preceding campaigns, and summons from them cavalry and the light-armed infantry,
who were accustomed to engage among them. On their arrival, as they were
mounted on unserviceable horses, he takes horses from the military tribunes and
the rest, nay, even from the Roman knights and veterans, and distributes them
among the Germans.’

[Caesar, The Gallic
War, VII. 65]

Caesar tells us that the bulk of the German mercenaries were
cavalrymen. Evidence for Germanic cavalry in the archaeological record at
Alesia is slight; this is partially because German weapons are hard to
distinguish due to their similarity with Gallic weapons. One shield boss with a
central projecting stud is almost certainly German, given its similarities to
later German bosses. Bones of horses coming from the ditches discovered at the
foot of Mont Réa have been identified as coming from a male horse of no more
than three years old. These bones may represent the well-bred young stallions
given to the German cavalry. If this is correct, the evidence would conform
well to the German cavalry attack on this region. Recent interpretation
suggests that these horse remains may have been deliberately buried in the
ditches, not simply to cover them up, but in a form of sacrificial burial. It
was the Gallic custom to bury sacrifices in the peripheral ditches of
sanctuaries. In Gallic eyes, the circumvallation ditch may have been associated
with the practice of ritual sacrifice in these periphery ditches after the
defeat.

‘There is not even any
great abundance of iron, as may be inferred from the character of their
weapons. Only a very few use swords or lances. The spears that they carry –
framea is the native word – have short and narrow heads, but are so sharp and
easy to handle, that the same weapon serves at need for close or distant
fighting.’

[Tacitus, Germania,
VI]

Along with the cavalry, the Germans are described as using
spearmen who mingled with the mounted troops. This tactic may explain the
successes of the German cavalry during the Alesia Campaign. Tacitus tells us
that German warriors were lightly armed, the cavalry often having only a spear
and shield. The infantry were armed in a similar manner, with the addition of a
number of small javelins. All the Germans were dressed lightly with few wearing
armour or helmets. Some even fought completely naked. In the main, German
warriors wore breeches and large cloaks that would be draped over their
shoulders in regional style. Their clothing was manufactured in simple colours
and patterns, the only ostentatious part of their dress being elaborate knotted
hairstyles. These varied from tribe to tribe and so identified each warrior as
the member of a particular clan, a practice that had continued for hundreds of
years.

‘[The Germans are] … a
people who excelled all others, even the largest men, in size; savage, the
bravest of the brave, despising death because they believe they shall live
hereafter, bearing heat and cold with equal patience, living on herbs in time
of scarcity, and their horses browsing on trees. It seems that they were
without patient endurance in their battles, and did not fight in a scientific
way or in any regular order, but with a sort of high spirit simply made an
onset like wild beasts, for which reason they were overcome by Roman science
and endurance.’

[Appian, History of
Rome, 3]

Caesar tells us that the River Rhine marks the border
between the Gallic peoples to the south and the Germans to the north. German
warriors were famed for their physique, size and fearlessness – attributes
which in Roman eyes made them appear as savages. Like the Gauls, the Romans had
a stereotype for the Germans, who were seen as brutish to the point of
indifference to death (a recurrent image even up until the present). As ever,
the reality was very different; the German peoples had as complex a society as
any other of the period. There were close affinities between Celtic peoples and
Germans, in art, religion and culture. An innate conservatism meant that German
tribes were slow to change and reduced access to resources seems to have meant
that their technologies were less advanced than in Gaul. What they lacked in
technology the Germans made up for in vigour. It was this trait that so
impressed Caesar in his brief excursion across the Rhine into Germany – a trait
that was to put to great use during the Alesia Campaign.

By MSW
Forschungsmitarbeiter Mitch Williamson is a technical writer with an interest in military and naval affairs. He has published articles in Cross & Cockade International and Wartime magazines. He was research associate for the Bio-history Cross in the Sky, a book about Charles ‘Moth’ Eaton’s career, in collaboration with the flier’s son, Dr Charles S. Eaton. He also assisted in picture research for John Burton’s Fortnight of Infamy. Mitch is now publishing on the WWW various specialist websites combined with custom website design work. He enjoys working and supporting his local C3 Church. “Curate and Compile“
Leave a comment

Leave a Reply Cancel reply

Exit mobile version