This is a draft of an article that was published in MHQ: The Quarterly Journal of
Military History (in a slightly different version). All citations should be to the
published version and not to this draft. Arther Ferrill
How did prehistoric man wage war? Did he fight in organized formations or were his
conflicts merely skirmishes of the sort that occur among some modern primitive societies?
Was prehistoric man aggressive at all, or did he live in an idyllic, peaceful environment,
as some believe? Was organized warfare the creation of civilized man, a fiendish
by-product of the emergence of civilization in the Ancient Near East? These and many other
questions have often been raised, and some authorities still regard them as open and
unresolved, yet archaeological discoveries in the twentieth century have provided many
reasonably definitive answers.
In my book, The Origins of War: From the Stone Age to Alexander the Great
(Thames and Hudson: London and New York, 1985), I dealt briefly with some aspects of
prehistoric warfare, but since then, partly because of published responses to my work, one
of them an article in this journal by Robert O'Connell, I have modified several of my
views. Also new research by me and by others has made it possible to elaborate on and
sharply focus several controversies.
There seems little reason to mince words or pull punches about man's aggressive
instincts. In prehistoric times man was a hunter and a killer of other men. The killer
instinct in the prehistoric male is clearly attested by archaeology in fortifications,
weapons, cave paintings, and skeletal remains. Whether these "instincts" are
biologically or culturally induced remains a matter of controversy, but by the end of
prehistoric times man was a fighter, capable of waging organized warfare of the sort seen
in later historical societies. The earliest civilizations along the Nile and in the
Mesopotamian valley witnessed a burst of warfare, intensified by the increased power of
the new states to marshal troops and pay the high costs of fighting. But organized warfare
was not new; it had been practiced for millennia in prehistoric times. When man first
learned how to write, he already had wars to write about.
Unfortunately, until quite recently anthropologists and prehistorians usually ignored
the importance of war in human culture. Because they tended to be pacifists or because
they were interested in other aspects of human culture, they often even denied that early
man and modern primitive man were warlike. Within the last generation there has been a
dramatic change- -now at least some anthropologists are beginning to realize that war is a
nearly universal social activity and that patterns of military organization within
prehistoric and primitive societies are as important as the political, economic and
religious systems they developed. A recent, good book on the subject is by R. Brian
Ferguson, ed., Warfare, Culture, and Environment (New York, 1984). Ferguson
offers a useful definition of war: "organized, purposeful group action, directed
against another group that may or may not be organized for similar action, involving the
actual or potential application of lethal force."
As a military historian, I cannot resist adding that the emphasis in any definition of
war must be on organization. When General Sherman said that war is hell, he was not
offering a definition. War is teamwork. It requires learning and can be practiced
efficiently only after intensive training, usually accompanied by firm, sometimes savage,
discipline. It is potentially dangerous, more so than hunting and much more so than
political, religious, and economic activities (except when they lead to civil war and
rebellion). Fortunately war may occur only occasionally and need not be a constant social
condition. Even so, in some historic societies (ancient Sparta and Rome, for example) the
need for defence (or aggression) was so great that most males were required to stand at
constant readiness for war. Although Sparta and Rome are extreme examples, most societies,
undoubtedly including prehistoric ones, had some institutionalized patterns of preparing
for military action even during periods of relative peace.
Perhaps the main difference between prehistoric and historic war is that in many cases
prehistoric populations did not share a common frontier. There was a no man's land between
settlements, and most military confrontations at least began there. Certainly that was
true of most Paleolithic conflicts. In the Neolithic with the development of
fortifications it became more common to wage war directly on one's own or enemy territory.
One further point of definition or clarification is in order. Prehistory did not end
every place on the face of the earth at the same time. In the eastern Mediterranean it
came to a close around 3500 BC with the appearance of civilization and writing along the
Nile and the Tigris- Euphrates Rivers. On the other hand in Northern Europe and elsewhere
in the world prehistoric conditions prevailed sometimes for thousands of years after
ancient Egyptian and Sumerian civilizations swept through the Fertile Crescent. Because it
is possible that Neolithic conditions after 3500 BC in areas outside of the Eastern
Mediterranean were influenced by cultural diffusion from Egypt or Mesopotamia, I shall
concentrate mainly on evidence earlier than that date. For similar reasons the experience
of modern primitives is not completely valid evidence for conditions in prehistoric times.
Primitive warfare, because it has often been conditioned in one way or another by
civilized societies (for example, the introduction of the horse into North America by the
Spanish) should not be confused with prehistoric warfare.
In recent anthropological studies of warfare the emphasis has been on the causes of
prehistoric and primitive war and on the relationship of war to the formation of early
states. Although some of this work has been outstanding, very little has been written
about how prehistoric man actually organized for war, how his weapons determined the
tactics of battle, what kind of training and discipline must have been involved, to what
extent prehistoric warfare was offensive or defensive, what were the details of logistical
support, the probable size of armies, and finally the development of strategy and tactics
in organized warfare against man.
One popular misconception about prehistoric warfare is that populations were so small
that warfare on a modern, historic scale is simply out of the question. In fact, that is
entirely wrong. Too many writers today tend to think of war as involving armies of
millions of men, but only in the twentieth century has this been the general rule. At
Waterloo both Wellington and Napoleon had armies of less than 100,000 men, and a half
century later at the Battle of Gettysburg neither army had that many. At the Battle of New
Orleans there were 9000 British and 4000 Americans on the field. Actually throughout much
of modern history armies have been far smaller than most people realize. In 1567 the Duke
of Alba marched to suppress a revolt in the Netherlands with only about 10,000 men. In the
French Huguenot wars armies numbered about 10,000 to 15,000 strong. In 1643 at Rocroi a
French army of 22,000 defeated Imperial Spain. Suffice it to say that armies of 5,000 to
15,000 men are large enough to represent major military striking forces in most periods of
Population figures for prehistoric times in the Mediterranean region are notoriously
difficult to determine, but there are some reasonably reliable estimates, as we shall see.
Also estimates of New World native populations before contact with the Europeans are
impressive for such places as the Hawaiian Islands where prehistoric armies were large.
Even some of the Northwest Coast Indian tribes, such as the Tlingit and the Kwakiutl, had
populations of about 10,000. In the Eastern Mediterranean as early as the seventh
millennium BC 5,000 to 6000 people may have lived at Çatal Höyük in modern Turkey, and
the population of Jericho at about 8,000 BC has been estimated at 2,000 with a possible
defending force of 500 to 600 men. At the beginning of the Neolithic period in the Near
East some armies may have numbered up to 1000 or so, and by the end of the period
somewhere between 5,000 to 10,000 men. Armies of that size compare with full scale
historic armies of a much later period. If size alone is a consideration, prehistoric
armies were capable of practicing warfare in a highly sophisticated fashion. In fact men
can be organized effectively for war in groups of less than 500.
There is no evidence for the practice of war before the late Paleolithic Age (35,000 to
12,000 BC). A few weapons are known to have been used much earlier. Stones and clubs,
man-made pebble choppers, and the spear were available hundreds of thousands of years ago,
perhaps millions. They were definitely used in hunting game and probably in attacks by man
on man, but there is no clear evidence. The famous Paleolithic cave paintings of France
and Spain, dating from the period of 30,000 to 20,000 years ago, show no certain scenes of
man killed by man. Mainly they depict animals, several thousand of them. Only about 130 of
the figures have been identified as possibly men, and many of them are dubious, simply as
men. Even so, the vast majority of the 130 are shown in peaceful scenes. A tiny number
appear to be pictures of men dying from wounds inflicted by spears or arrows, but they are
so badly drawn that not a single one can be certainly identified as a wounded or dead man.
It is possible that the bow and arrow and the sling go back into the Paleolithic Age,
perhaps as far back as 50,000 years ago, but again there is no definite proof of their use
that early. Stone darts, sometimes called "arrow heads," were made during the
Paleolithic Age, but they were not necessarily attached to arrows fired from a bow. They
may simply have been points inserted in spearheads or throwing darts. No one knows where
the bow and arrow were invented, but it appears most likely that they first came into use
at the end of the Paleolithic Age (12,000 to 10,000 BC), after the period of the cave
The new weapon spread quickly around the Mediterranean, more slowly perhaps even around
the world, from one prehistoric culture to another. It is uncertain whether the New World
bow reached America as a result of cultural diffusion from Africa, Europe and Asia, or
whether the bow was spontaneously invented in several different locations. Most forms of
the bow appeared in the Neolithic period, even the composite bow. One authority on early
bows has speculated that the flat short bow had a northern and north eastern origin, that
the simple long bow was western and that the composite bow came from the east.
It would be difficult to exaggerate the significance of the bow for prehistoric
warfare. It provided a revolutionary increase in range and volume of firepower. Before the
introduction of the bow long range firepower was provided by the thrown spear (sometimes
with the help of an atlatl, a spearthrower that extended man's forearm and gave the spear
more range, accuracy, and power). But the bow more than doubled the range of the spear,
and since the arrow was so much smaller and easier to carry, it was possible to deliver a
much greater volume of fire against the enemy. In some cases it could have been done from
concealment. When Neolithic man took position in a line and fired on command, he unleashed
a powerful barrage of arrows.
Almost simultaneously other new and important weapons appeared in the late Paleolithic
or Neolithic periods. The dagger, the sling, and the mace were found at Çatal Höyük in
Anatolia dating from about 7000 BC. The sling is an especially important weapon, deadlier
and with greater range and accuracy than the early simple bow. Everyone knows the famous
biblical story of David and Goliath, but few people realize how widespread the sling was
throughout the world and how devastating a weapon it could be. The ancient Greek writer,
Xenophon, tells us that as he led a group of Greeks out of Persia back to the Aegean his
slingers from the island of Rhodes fired slingstones farther than the Persian arrows and
that their accuracy was greater. Projectiles for slings can vary dramatically in size from
pebbles to lead shot to fist-sized stones. The larger missiles can smash skulls and break
bones, even against armor.
Some authorities believe that the sling was not often used because slingers took up too
much space in line. That is wrong on both scores. The sling was a common weapon in ancient
war, and slingers could fight in relatively close formation. Slings need not be slung
overhead, nor need they be long. Short slings slung underhanded like a softball pitcher
with only one swing of the hand became the standard, at least by Roman times, and probably
much earlier. Another misconception is that it takes nearly a lifetime to learn the use of
the sling and that only men who used it as boys in their native land could be recruited as
slingers. Nothing could be further from the truth, even though there were some regions in
antiquity, such as Rhodes and the Balearic Islands, famous for their slingers. Still, one
of my own students in military history became reasonably adept with the sling during
Spring vacation a few years ago, and all Roman legionnaires received regular training in
it. The weapon was important and widely used, particularly in a siege.
At least by the Neolithic Age man also learned how to put bashing weapons made of stone
onto wooden handles. The Indian tomahawk is the classic example, but around the ancient
Mediterranean the mace was more common although battle-axes were known too. To modern
readers the panoply of prehistoric weapons seems quaint and antiquated--there are no
intercontinental ballistic missiles, no hydrogen bombs, no tanks or aircraft carriers, no
gunpowder in any form. But prehistoric warfare was savage. There were also no Geneva
Conventions, and a captive who gave his captor name, rank, and serial number would have
had his skull broken (assuming captives were taken at all) or even more likely would
simply have been reduced to permanent slavery. Captive women were taken as slaves and
concubines, and modern distinctions between the treatment of the civilian and military
population were often nonexistent.
Perhaps the most impressive evidence of prehistoric preoccupation with warfare are the
archaeological ruins of massive fortifications constructed by early man. In Paleolithic
times men used natural shelters for protection against animals and enemies. Caves,
forests, rivers, and deserts all serve as defensive barriers, but with the introduction of
long range missiles in the form of arrows and slingstones, accompanied by the need to
produce food in an agricultural settlement, man had to build artificial barriers, usually
high walls, for defense. Neolithic fortifications were sometimes massive. The walls of
Jericho were ten feet thick and thirteen or more feet high. A twenty-eight foot tower that
was thirty-three feet in diameter with a central stairway and an entrance at the bottom
was attached to the wall. Although the entire wall remains unexcavated, it probably
extended about 765 yards and enclosed an area of approximately ten acres.
Although Jericho eventually became a settled agricultural community, it first attracted
residents as a hunting site. To protect themselves against invaders the inhabitants built
the wall. Evidence now indicates that the wall went up before the cultivation of plants.
Elsewhere I have suggested that the military need for fortified defense against the new
projectile weapons forced man to settle down and led to the discovery of agriculture.
Behind his new walls Neolithic man could store surpluses of food, and because he could
fall back behind the walls for protection, he could work the land outside them with some
sense of security.
A quite different form of military defense is seen in the architecture of Catal Huyu?k.
There were no massive outside walls, but the houses were all interconnected, sharing
contiguous inner walls. Entry into the rooms was through holes in the roofs reached by
ladder. As a result the line of the outside walls of the rooms around the settlement
formed a kind of fortification. When attackers approached, the inhabitants could simply
scamper up their ladders, retrieve them, and if an invader broke through a wall, he simply
found himself in a single room. Many other Neolithic settlements in the Near East were
protected by fortifications of one kind or another.
It should be obvious that war was a very important part of the life of prehistoric man
in the Neolithic Age. What remains is to determine whether it was true organized warfare
comparable to that practiced in civilized societies. That requires some assessment of
Neolithic strategy and tactics, which may sound somewhat highblown, but if Neolithic man
did not apply strategy and tactics to his fighting, then it was not organized warfare. The
elementary, basic requirement for true war is the ability to form troops in column and
line. If a body of warriors cannot march in column and fight in line, it is not an army.
Forming a column and holding a line requires teamwork, training and discipline. The
natural instinct in a clash of arms is to run, an act that jeopardizes everyone, but there
is safety in the line. If the enemy cannot break through your line, or come around behind
it, you will win. If your line is penetrated, you are finished, and your life is in grave
In The Face of Battle John Keegan wrote:
Inside every army is a crowd struggling to get out, and the strongest fear with which
every commander lives--stronger than his fear of defeat or even of mutiny--is that of his
army reverting to a crowd through some error of his making...Many armies, beginning as
crowds, remain crowdlike throughout their existence.... Tactically quite un- articulated,
they were vulnerable to the attack of any drilled, determined, homogeneous force...The
replacement of crowd armies by nuclear professional armies was one of the most important,
if complex, processes in European history.
There is every reason to believe that even in Neolithic times man learned to fight in
an organized fashion. The fortifications themselves suggest teamwork and leadership,
discipline and order. Fortunately, however, we are not limited simply to fortifications.
Neolithic cave paintings show warriors forming a line, firing on command, and marching in
column behind a leader who was wearing a distinctive uniform that distinguished him from
the rest of his troops. One painting may possibly even show Neolithic warriors executing a
double envelopment. Because there are no written documents for prehistoric times we do not
know about the great wars that must occasionally have broken out, and obviously we cannot
know for certain about the tactics of individual battles. There are not even paintings or
drawings that are as detailed as the ones from Bronze Age Egypt or the Minoan Aegean.
Anthropologists have identified some common strategies in prehistoric warfare. One of
them is to interdict use of unoccupied territory to prevent exploitation of its resources
by others. Associated with that is the maintenance of the no man's lands between
prehistoric communities. Tactics in such a strategy often did not involve full scale
battles and consisted mainly of raids and terrorism. Another strategy was to plunder the
settlement or territory of a neighbor much as the Galls did against Rome when they sacked
the city in 390 BC. For a relatively large and mobile force with a rich but weak neighbor
a raid was a better way of acquiring resources than working for them. Finally there was
the strategy of unconditional surrender: the defeat of the enemy and seizure of his
territory. This often involved larger battles and considerable violence. Unfortunately we
cannot often reconstruct the details of prehistoric battles.
But there are some tantalizing hints about the nature of Neolithic warfare. One of them
comes from the very earliest stages of the Neolithic, probably even from the end of the
Paleolithic Age. There is an ancient Egyptian cemetery, actually at the northern edge of
modern Sudan, discovered during the intensive excavations that were sponsored as the Aswan
damn was under construction, when everyone knew that much archaeologically rich land would
soon be under water. The excavators called it "Cemetery 117" and identified it
as Epipaleolithic (12,000-4500 BC) from the so-called Qadan culture. This particular
cemetery is of special interest because nearly half of the fifty-nine skeletons show signs
of violent deaths inflicted by small flake points (microliths), probably arrowheads. Some
of the dead suffered from multiple wounds, and points were discovered in the sphenoid
bones in two skulls, suggesting that the victims were shot under the lower jaw, probably
as they writhed in pain on their backs. A young adult female had twenty-one stone
artifacts in her body. Another, an adult male, had nineteen wounds. It is possible that
some of the others, whose skeletons now show no sign of injury, were killed too, since not
all deadly wounds leave a mark on the skeleton.
This cemetery on the Egyptian-Sudanese border is not the only prehistoric burial site
that contains evidence of human violence. Neolithic cemeteries near the Dnieper rapids in
the Soviet Union and at Schela Cladovei in Rumania also reveal the signs of warfare. They
too date to sometime before 4000 BC. All three settlements have one thing in common. They
were on rivers where the fishing was no doubt good, and where there may have been
reasonably rich agricultural land. CSR, Competition for Scarce Resources, undoubtedly had
much to do with prehistoric warfare. Some of the examples mentioned above are quite
famous--Jericho, Çatal Höyük, and Cemetery 117, but they are not unique. Just last year
a two- volume book was published entitled Enclosures and Defences in the Neolithic of
Western Europe (Oxford, 1988). In fact Neolithic sites all over the world reveal the
signs of planning and building against outside attackers. Warfare in prehistoric times was
the rule--not the exception.
Quite recent research has shown that the Neolithic world was dotted with
fortifications. As early as the fourth millennium they appear in settlements all over
Northern Europe. A good example is the Neolithic enclosure at Compiagne on the Oise in
modern France. As one authority has recently said, "Such sites proliferated in
Western Europe during the fourth-third millennia BC and are the oldest monumental
structures found so far in the central Paris Basin." Normally a timber palisade
surrounded by shallow ditches formed a perimeter around the settlement. The one discovered
in 1978 at Compiagne, during the extension of an industrial park, was only partially
excavated, but aerial photographs reveal that it was about 750 meters long in the form of
a rectangular bow with a straight palisade trench as the string of the bow. Altogether the
enclosure included about 14 or 15 hectares, and the circumference of the timber palisade
was 1800 meters. This is roughly three times larger than the enclosure at Jericho. The
quantity of earth removal for the trench is staggering. As the illustration shows, the
ditches were wide and deep. The palisade was made of posts, an average of 14 posts for
every 10 meters, lined with clay for protection against moisture, and the gaps filled with
wickerwork. Oyster shells and pottery sherds were placed in the palisade trench as a
foundation for the posts. Comparison with several similar sites in La BassCAe and in
Picardy (six separate sites altogether) shows that they were defended with structures
built in about the same fashion.
At Crickley Hill in county Gloucester, four miles south of Cheltenham, there is a small
(4 acre) prehistoric fortification. It had a causewayed enclosure of two rings of ditches
with banks made of horizontal layers of stone and earth on the inner rings. There were at
least three outer and inner ring entrances with gates into the settlement. Fenced roads
led from the outer gates into the interior camp. When that camp was abandoned, the inner
ditch silted up and, later, new occupants rebuilt the fortifications with a single outer
ring There was a post palisade somewhat less than two meters high directly behind a stone
bank on the inner edge of the ring. The three entrances corresponded roughly to the
previous ones. What is striking about this phase of the settlement is that there is clear
evidence that it was attacked and destroyed. Flint arrowheads cover the roadways from the
eastern entrances into the inner camp, and over 400 were found in the eastern entrances
themselves. Obviously the defenders were overwhelmed, because the site was destroyed by
fire and abandoned. Clearly the ditches were designed to slow the invaders while the
defenders fired at them from the palisades. Actually the earlier double walls would have
been a more effective defense in depth. An author writing recently for Scientific American
has suggested that the increase in warfare and fortifications in the Middle Neolithic in
Europe may have been caused by a colder and wetter climate and the competition for scarce
resources. Much of the best land had already been claimed and there was no longer an
outlet for population pressure. But there was also an intensification of warfare in North
Africa and the Middle East, and there a deteriorating climate cannot be the explanation.
As prehistory came to an end with the emergence of civilization in Mesopotamia and
along the Nile, warfare played a critical role in the formation of the new states.
Actually all over the world when primitive societies became emergent states, military
institutions were critical and sometimes even determinative. Early Egyptian history
testifies to the importance of armies. Even before the unification of the two kingdoms, in
the predynastic period, the Palette of the Fortresses shows fortified towns under siege
and the booty of war on the other. The famous Palette of Narmer, first of the Egyptian
pharaohs, depicts the new ruler on one side in the act of slaying an opponent and on the
other he is reviewing the headless bodies of his enemies under the standards of his army.
At the bottom the bull of Narmer is destroying a fortified site. In the upper right one
can see Narmer's ship, and it is possible that his invasion force moved down the Nile into
Lower Egypt by flotilla.
In Mesopotamia the war chariot was used as early as 3000-2500 BC, and in both Egypt and
Mesopotamia the weapons arsenal was highly developed with new arms and armor made of
bronze. Whereas the bow and arrow were used extensively in Egypt, in Mesopotamia Sumerian
infantrymen were armed with javelins, spears, daggers and swords. In both regions other
prehistoric weapons, such as maces and battle-axes, were widely used. In Mesopotamia,
especially, siege warfare and fortification were highly developed. By the second
millennium, in the Battles of Megiddo and Kadesh, armies of 20,000 men marched distances
of hundreds of miles with the logistical support system that entailed. Warfare had emerged
Bogucki, Peter, Forest Farmers and Stockherders (Cambridge,
Burgess, Colin, et al., Enclosures and Defences in the Neolithic of Western Europe,
2 parts (Oxford, 1988).
Divale, William T., Warfare in Primitive Societies: A Bibliography (Santa
Ferguson, R. Brian, Warfare, Culture, and Environment (Orlando, 1984).
Ferrill, Arther, The Origins of War (New York, 1985).
Keegan, John, The Mask of Command (New York, 1987).
Manzanilla, Linda, ed., Studies in the Neolithic and Urban Revolutions, 2 parts
Mellaart, James, The Neolithic of the Near East (London, 1975).
Rausing, Gad, The Bow (Lund, 1967).
Telegin, D. Ya., and Potekhina, I.D., Neolithic Cemeteries and Populations in the
Dnieper Basin (Oxford, 1987).
Turney-High, H.H., Primitive War: Its Practice and Concepts, 2nd ed. (Columbia,
Whittle, Alasdair, Neolithic Europe: A Survey (Cambridge, 1985).