Violence in the Near East at the dawn of agriculture: the evidence

by Edward Pegler on 10 December, 2016

Does farming lead to violence, violence lead to farming, or was the period during and after the invention of farming in the Near East as peaceful as it gets?

unfunny cartoonYou’ll find many opinions on the role of farming on the origins of violence and of violence on the origins of farming in human history. I’m not sure how many of the current opinion holders have ever looked at the available data for the Near East, where the first transition to farming occurred. They really should.

As far as I can tell it was Ofer Bar-Yosef, talking about the first walls of Jericho back in 1986, who first drew attention to this data:

‘There is no archaeological evidence, in the form of burned settlements, the remains of mass massacres, or fortifications, for the existence of social aggression in the Levant before the 6th millennium B.C.

Prof. Bar-Yosef was looking purely at cultures which straddled the change from foraging to farming in his own neighbourhood, along the Levantine margin of the Mediterranean. He wasn’t looking further north and west at contemporary events in the foothills of the Taurus and beyond. I want here to extend that view.

Spotting conflict in the archaeological record

The aim of this post is to discuss the available evidence for violence at the foraging-farming transition in the whole of the Near East. To do this I need to discuss the evidence of fortifications or extensive destruction of settlements, usually by fire, and the clues from burials. Where other evidence is available I will use it. Sadly, the evidence of weapons is, as far as I can tell, of little use for this period, as these are too generic to indicate whether they are used for hunting or conflict.

For signs of violence, burials reveal either sharp blades and points apparently within body cavities or in bone, and other evidence of trauma (some kind of injury shown in the bone). Particular traumas that are often seen and may be associated with violence are ‘cranial depressed fracture’ (due to head bashes with a blunt instrument) and ‘parry fractures’ (ulna breaks, due to holding up arms in self-defence). However, any fracture could be the result of accident.

As much as anything else though, trauma studies reveal the variability of detail given in reports. The examples of trauma or weapon damage sited below are undoubtedly not all that have been recovered. This is why I have mentioned a lack of evidence only where studies have been done.

Furthermore, bones do not reveal soft-tissue damage, whether by violence or not. It is also difficult to analyse bones which have been subject to corrosion after burial. This means that any estimates of violence will be low compared to reality, although this will be true of other archaeological ages too.

Lastly, all dated quoted are approximate, calibrated dates.

The Late Epipalaeolithic (~12,750 to 9,750 BC)

Possible signs of conflict in the Late Epipalaeolithic (Natufian).

Possible signs of conflict in the Late Epipalaeolithic (Natufian).

In the Near East, the Late Epipalaeolithic is the time of the first evidence for settlement. Many of the excavations for this period are concentrated in the southern Levant, although other excavations in Syria and Iraq indicate that there are more to be found elsewhere.


Tell Qaramel, Syria (10,900-9500 BC) – Excavations by Ryszard Mazarovski’s team recorded five round towers within the settlement. They date from the end of the Epipalaeolithic to about the middle of the PPNA, around the time that initial farming experiments were taking place. The reasons for their construction are unknown.

Skeletal Evidence

Hayonim Cave (13,200-11,800 BC) – Of 48 individuals, two young men and a young girl showed evidence of cranial trauma (both of the men from the earlier half of the period, the girl from the later half). Another child showed a fractured jawbone. Other injuries include a fractured upper leg, fractured lower leg, and two cases of a broken foot.

El Wad (13,000-?11,000 BC) – A young man showed cranial depressed trauma. Other injuries include a broken jaw, a fractured hand and a fractured foot.

Kebara Cave (12,500-11,000 BC) – From Fanny Bocquentin’s study of of six adult males and twelve children, three adult males, all from the earlier half of the period, show serious injuries. One had a projectile buried in his spine. A slightly younger man had a cranial fracture on the top of his head, and the last had a skull impact that embedded a portion of skull in his brain.

Ain Mallaha (12,100-10,400 BC) – Of 102 individuals, an old man and woman and a younger man showed evidence of cranial trauma. Additionally, an adult female, of earlier date, had a projectile point buried in the front of her face. Other injuries included a possible parry fracture (ulna) toward the end of the period and a fractured hand and foot.

Rakefet (11,300-10,400 BC) – Of 4 individuals, one child, around 10 years, showed evidence of cranial trauma.

Zawi Chemi-Shanidar (10,600 – 10,300 BC) – Arkadiusz Sołtysiak sites evidence for healed cranial fractures in 4 out of 9 buried individuals (two men, one young woman and a child), described by the original investigators (Ralph Solecki and others) as ‘of malicious intent’.

Et Tin (?10,300-9,500 BC) – Of 5 individuals none was found to have serious injury, although this was based on only fragmentary jawbones.

Nahal Oren (10,300-9,500 BC) – Of 41 individuals, an older man and woman (from the second half of the period) had evidence of cranial trauma. If this is the same woman reported by Denise Ferembach in 1959, then it was fatal. Additional injuries include a rib fracture.

(By way of comparison, at cemetery 117,  Jebel Sahaba, Sudan, dated to at least 12000 BC and probably older, 40% of the 59 or so individuals showed signs of violent, possibly fatal, trauma.)

Much of the above data comes from the doctorates of Vered Eshed and Fanny Bocquentin, both based on material from the southern Levant. Eshed, in his analysis of the data, found that fractures in general were seen on about 3% of 200 skeletons analysed. However, skull fractures were much more common, seen in 9 out of 54 skulls analysed (about 17%). Bocquentin’s data shows that of 8 people showing signs of trauma suggesting violence 6 or 7 were from the first half of the period, including the two with projectile wounds. Whether this difference is significant would need further study.

PPNA to Middle PPNB (~9,750 to 7,400 BC)

Possible signs of conflict during the PPNA and Middle PPNB

Possible signs of conflict during the PPNA and Middle PPNB, with the extent of agriculture shown in green.

The Pre-Pottery Neolithic A (or PPNA) is the time when initial experiments in agriculture were tried. However, it was from early in the Pre-Pottery Neolithic B (or PPNB) that agriculture appears to have become central to the survival of many Near Eastern communities. Even so, it was of a primitive form. I have lumped these two phases together as the skeletal evidence for them is often not distinguished in the literature.


Almost all of the Pre-pottery Neolithic settlements are built in areas without natural protection given by the landscape. Of all the settlements of this period so far found in the Near East only one has defensive structures. However, sites in the Anatolian highlands such as Aşıklı Höyük have  houses which are densely clustered with small windows and roof entrances. These have been taken by some, following James Mellaart’s lead, to indicate a defensive aspect.

Jericho (9,750-9,300 BC) – Kathleen Kenyon’s team found evidence for massive walls and a central tower here in the 1950s. The walls and tower were first constructed around 9,750 BC, around the beginning of the the PPNA. They went out of use by around 9,300BC.

The walls were interpreted by Kenyon as defensive structures, and have been taken as such by many theorists since. However, Ofer Bar-Yosef made a good case that the walls were in fact to defend the settlement against mudflows coming from upslope during flash-floods. This was based on the thickness of the walls being much greater on the upslope side, the rapid accumulation of mud against each phase of walls (rebuilt at about 100 to 200 year intervals), and the abandonment of wall building once the settlement had risen noticeably above the floodplain.

The reason for the building of the tower, as with that of Tell Qaramel, is unknown, but its central location in the settlement suggested to Bar-Yosef that it was not defensive.


Beidha (~7,900 BC) – There is evidence of fire destruction in Level IV, thought to be roof fires, that appears to have resulted in changes in house design after this time.

Skeletal Evidence

Nemrik 9  (9,600-?7,300 BC) – Of 28 individuals studied, Arkadiusz Sołtysiak found no evidence for cranial fracture or damage. Minor injuries to individuals included hands, a lower arm, a rib and a knee.

Körtik Tepe (~9,600 to 9,300 BC) – of 10 individuals identified healed depressed cranial fractures occur on 2 adult young men.

Jerf el Ahmar (9,500-8,700 BC) – 1 young adult female was found ‘sprawled’ on the cleared and cleaned floor of a burnt building with evidence of peri-mortem decapitation (the head is missing).

Ganj Dareh (8,300-7,500BC) – Of 49 individuals studied, 1 middle aged man showed depressed cranial fracture and one young woman a healed fracture on her upper leg.

Jericho (8,350-7,380 BC) – Of 477 individuals studied the injures seen, all minor and healed, included a rib and arm fracture, two upper leg fractures and a lower leg fracture. Steven A. LeBlanc asserts (in the unseen ‘Early Neolithic Warfare in the Near East and its broader implications’) that there were ‘a few… healed skull fractures’ but these are not mentioned by Glencross & Boz in their literature search.

Abu Madi, Ujrat el-Mehed and Wadi Tbeik (8,500 and 7,300 BC) – Of three early agricultural sites listed here, all from South Sinai, a rather confusing report by Glencross & Boz states 20 individuals studied and only one individual showing evidence of a healed rib fracture.

Yiftahel (?8,100-7,000 BC)- Of 3 plastered skulls dated around 7,900 BC one, probably an adult woman, showed evidence of a small, healed depressed cranial trauma. However, Vered Eshed found only minor skeletal fractures on 8 other individuals studied.

Abu Hureyra – (?8,600-?6,000 BC) Of 39 individuals, Sołtysiak found no evidence for cranial fracture or wounds.

Çayönü (8,200-7,400 BC) – of 626 individuals identified, 6 (5 including young adults) show evidence of healed cranial depressed fracture, 1 adult has evidence of a depressed fracture near the eye, there is evidence of serious bruising on a jaw, and there is also a parry fracture on a adult.

Aşıklı Höyük (8,200-7,400 BC) – of ?65 individuals identified, depressed cranial fracture is found on 1 mature man and, possibly, on a mature woman.

Musular (?7,500 BC) – of 8 individuals studied from this site, associated with Aşıklı Höyük, 1 showed depressed cranial fracture.

Ghwair 1 (?7,600-7,500 BC) – Probably from the final phase of the settlement (phase III), an incomplete skull (with forearm fragments) of an elderly female had a projectile point embedded in her jaw. This appears to be the most severe incident during this period. However, the burial is badly disturbed so the date could be later.

(Again, by way of comparison, the Mesolithic hunter-gatherer cemetry of Vasil’evka III, from the Middle Dnieper Rapids area in the Ukraine, dated to 9600 BC, contains two bodies with arrowheads imbedded in bones, one in the spine, one in a rib.)

Vered Eshed’s doctorate also highlights a further 137 individuals from this period, coming from 7 sites in the southern Levant: Abu Gosh (8,300-?7,500 BC), Gilgal (9,600-9100 BC), Hatoula (?10,200-9,300 BC), Horvat Galil (?8,700-7,800), Kfar Hahoresh (?8500-7000 BC), Nahal Oren (?9,100-8,600) & Netiv Hagdud (9,200-?8,000 BC). Of these, there was only one case of cranial trauma identified during this period (at the earliest site, Hatoula) and occasional minor trauma only in other cases.

Notably, Eshed concludes that overall trauma levels are unchanged between the Epipalaeolithic and the Pre-Pottery Neolithic. However, the evidence of cranial depressed fracture in the Pre-Pottery Neolithic is much lower, with just the 1 individual from Hatoula identified out of the 30 skulls studied (compare with the results for the Epipalaeolithic above).

Other indicators

Traces of human blood have been found at Çayönü on the floor of the ‘Terrazzo Building’ and the ‘Skull Building’. The ‘Terrazzo Building’ has a smooth floor with drainage channels, whereas the ‘Skull Building’ was where many skulls were placed. As pointed out by Tatiana Kornienko, similar buildings have been found along the Upper Euphrates River at Nevalı Çori & Tell ‘Abr 3.

Klaus Schmidt, the excavator of Göbekli Tepe, also drew attention to the apparent carving of a headless man with erection near the base of one of the upright stones there (the ‘Vulture Stone’ of structure D), interpreting it as a sign of a death-erection due to execution.

This itself raises the larger issue of decapitation in the Late Epipalaeolithic and Pre-pottery Neolithic. Evidence of skulls separated from the rest of the skeleton after decomposition is frequent in the Near East during this time. However, some of the evidence, especially the presence of some neck vertebrae showing cut marks, or the body from Jerf el Ahmar, suggests that some heads were removed from bodies at the time of death or shortly afterward.

Late PPNB to PPNC (~7,400 to 6,250 BC)

Possible signs of conflict in the Late PPNB and PPNC in the Near East.

Possible signs of conflict in the Late PPNB and PPNC in the Near East.

This is the time where more recognisable styles of agriculture, including the use of animals, became widespread and pottery started to be used. There is an increased tendency during this period for the development of ‘mega-sites’, large settlements, quite densely packed.


During the PPNC only one site has currently been shown to have walls. However, house-clustering becomes more common throughout the region, from Anatolia down to the Levant.

Tell Maghzaliyah (6,800-6,600 BC) Excavated by a Nikolai Bader’s Russian team in the late 1970s and early 1980s, a ‘massive enclosure’ was found in the final levels (14 and 15) of the settlement.

Ras Shamrah (later ‘Ugarit’, 7,500-7,000BC) – Excavated by Claude Schaeffer in the 1930s, this had a 5 metre high, earth cored, gravel and stone rampart which could well have been for sea defence.

From about 6,500BC more sites have walls or similar defences, including Hacılar, Kuruçay, Bademağacı, Tell Es Sawwan and Khirokitia, as well as Nea Nikomedia in northern Greece.


From around 6500 BC there are frequent signs of extensive fire, such as at Tell Sabi Abyad, Can Hasan, Yümüktepe, Hacılar, Bademağacı (mentioned below), Höyüçek and Çatal Höyük.

James Mellaart also mentioned possible ‘traces of disturbance from Çukurkent (~6700 BC), although what these are I couldn’t tell, so will not consider it further for the moment.


Lee Clare and his co-writers point out that this period, especially from around 6,500 BC, is associated with the rise in the occurrence of ‘sling missiles’ (rugby or american football shaped clay pieces) as the number of arrowheads falls, and that these missiles appear to occur frequently in areas with other indicators of conflict (e.g. Tell Sabi Abyad, Höyüçek and Çatal Höyük).

However, the quality of such missiles needs to be considered to be sure that they could always effectively be used as weapons against others. This is highlighted by Catherine Perlès, who argues for the use of poorly baked sling missiles to control livestock in a region of Greece with no other indicators of conflict in the centuries after 6,500 BC.

Skeletal Evidence

Basta (Jordan) (7,500 – 7,000 BC) – Of 29 adults, five had healed cranial fractures. Two fractures, one healed and one fatal, were also found on a boy of about 8-12 years.

Ain Ghazal (7,400-6,700 BC) – Of 7 individuals studied 1 man had a flint blade fatally embedded in his head, as well as broken bones, strongly suggestive of violence.

Çatal Höyük (7,400-6,300 BC) – of 670 individuals identified, 8, including two women, show evidence of what’s described as ‘head wounds’. One old female also shows a healed depressed cranial fracture. Other potential conflict related injuries include 3 parry fractures and a healed ulna fracture (a parry fracture too?). Other injuries include five fractured upper leg bones, one healed, a fractured rib, a shoulder fracture, a healed foot bone, ‘splinters of bone’ embedded in the elbow, and one individual with two fractured but healed lower leg bones and damage to the base of their spine.

Atlit Yam (?7,000-6,500 BC) – Of 60 individuals studied, Vered Eshed only found evidence for only minor trauma.

Nemrik 9 (7,000 – 6,500 BC) – from the final phase of site occupation there are arrow heads associated with the skeletal remains of several individuals which are thought to have been embedded in the bodies at the time of death. According to Brian Ferguson (perhaps quoting Gary Rollefson’s ‘Violence in Eden‘ comment), ‘One male skull contained two points, a second skeleton had a point in the pelvic area, and a third had a broken point next to a broken arm. These points are of a type that is unusual locally, suggesting that attackers had come from a distance.’

Bademağacı (7,100-6,200 BC) – of 48 individuals identified, 7 show evidence of depressed cranial fracture, mostly healed, but one, in the case of an older woman, probably a fatal blow. Other individuals show fractures around their hands and fingers. As well as this 11 people, including women and children, also appear to have been killed in a house fire late in the sequence (around 6300 BC cal).

Abu Hureyra (~6,900 -6,500 BC?) – the original excavators found the skeleton of one man with an arrow head in the chest region, suggesting a fatal shot (trench E, phase 6).

Tell Ain el-Kerkh (6,500-6,200BC) – Of 237 individuals at the site Akira Tsuneki & Sean Dougherty identified 127 adults, 10 men and 3 women showed skeletal trauma, 7 of them showing multiple injuries, possibly indicative of violence. 1 man had fatal fractures on his head and jaw, associated with signs of ‘boring’ (a weapon?). Three other individuals showed signs of skull trauma. Another man had healed fractures on his hand, wrist, shoulder, possibly with ‘other wounds’. Two women had forearm fractures, one with a broken shoulder, the other with several broken ribs and a broken foot. Most injuries were to the hands, feet and upper body. As well as these, a young man seems to have been killed by what in all likelihood was an accident, his thigh bone being badly broken.

Khirokitia (6,900-5,400 BC) – out of an unknown number of individuals identified, an adult man showed cranial depressed fracture, another showed a healed cheekbone fracture, and a third man several healed fractures, including the jaw and front and back of the head.


General patterns

All of the information above is, frankly, just a mess of fragments. While there is clearly evidence of some violence it certainly does not suggest that the period in question was a time of great conflict. Subsequent millennia (notably from around 5,000 BC) appear to have presented evidence of far more violence. Having said that, absence of evidence…

Whatever, there are subtle hints of patterns, as can be seen from the maps. Here are some that may exist:

  1. Violent conflict decreased during the Epipalaeolithic in the core regions of future agriculture, from about 13,000 BC down to 9,500 BC.
  2. During the period of transition to full agriculture levels of conflict were at theire lowest. However, adult human sacrifice, involving beheading, may have occurred in settlements along the Upper Euphrates River and into the Taurus mountains to the north and west.
  3. After 2000 years of relative peace, a rise in conflict starts from around 7,400 BC across the entire region, increasing in intensity gradually to about 6,400 BC.


Glencross & Boz have highlighted the presence of cranial depressed fracture (mostly healed) on a number of skulls during the transition to farming. From the data above it appears to be most common in young men, less common in both older men and women, and rare in young women and children…

(it should be pointed out, as Patricia Lambert has for California, that healed cranial fracture is a result of surviving the cause, so happened earlier in one’s life – therefore, for the sample above, fracture is likely to have happened generally when men were young, but when women were older).

It affects anywhere between about 1% and 20% of individuals studied, depending on the site. Due to variations in numbers studied, I’m not sure that it’s useful to compare numbers between periods. There may or may not be significance in its distribution along the northern margin during PPNA-MPPNB times.

Cranial depressed fracture is suggestive of 1) a forceful kick or blow to the head with a blunt instrument, 2) something heavy falling on one’s head from a height, or 3) of a fall onto one’s head from a height. If there is either a serious conflict or ritual (sacrificial) element to these fractures then it’s quite strange that most people survived the impact.

It’s perhaps most reminiscent of knocking someone out temporarily without fatal intent, like any good action adventure moment. One reasonable suggestion is that this is young men from within a settlement fighting amongst each other (as pointed out by Patricia Lambert – see comments below – solving an argument by clubbing someone can be a socially acceptable way of preventing someone getting killed, as seen in the Yanomamo of the Amazon).

Alternatively, it could even be the result of coming of age games, perhaps involving ‘fighting’ large, hoofed animals, within or between settlements . We still see the same idea in Pamplona and bullfights (this really depends on the shape of the depressions and could be easily disproved). Lastly, it could be a form of religious self-immolation. People do the funniest things, after all.

Lastly, the fact that some older women also experienced head injuries is, as Bonnie Glencross has pointed out, a possible sign that some of men’s less desirable habits are nothing new.

House Clustering

The pattern of house clustering, it’s initial occurrence at the northern margin of the agricultural zone (e.g. Aşıklı Höyük), and its spread south during the more rumbustious Late PPNB may well suggest a defensive element to the clustering.

With their small, high windows, roof entry instead of doors, and lack of large gaps between neighbouring walls, house clustering certainly seems like a reasonable form of defense instead of a perimeter wall. As James Mellaart said of Çatal Höyük ‘if an enemy succeeded in breaching the wall he found himself in a closed room from which the ladder had no doubt been removed with the defenders waiting for him on the roof.’ However, what is less discussed is how high the walls of the outer layer of houses were. If they weren’t that high I can’t help wondering what a quiet invader carrying a ladder could achieve.

Alternatively, or additionally, walls may have been designed to prevent the movement of animals, whether domestic or wild. This is possibly backed up by the use of ‘long walls’ within settlements of the southern Levant (e.g. Ain Ghazal & Atlit Yam) in the Late PPNB to PPNC.

Currently, the evidence can be read either way.


That’s a different kind of post.


Akkermans, P.M.M.G. & Verhoeven, M. 1995 An image of complexity: the burnt village at Late Neolithic Sabi Abyad, Syria, Am. J. Archaeology 99, 5-32.

Dating to around 6100 BC cal.

Banning, E. 2003 Aceramic Neolithic: Pre-pottery Neolithic, In: ‘Encylopedia of Prehistory, Volume 8: South and Southwest Asia’ (eds. P.N. Peregrine & M. Ember), Springer, p3-4

Bar-Yosef, O. 1986 The Walls of Jericho: An Alternative Interpretation, Current Anthropology 27, 157-162

The argument for flood-defence for the walls of PPNA Jericho.

‘There is no archaeological evidence, in the form of burned settlements, the remains of mass massacres, or fortifications, for the existence of social aggression in the Levant before the 6th millennium B.C. This is not to say that there were no rivalries that ended in some fighting, but the archaeological record as shown in the deep stratigraphies of many tell sites (such as Mureybet, Abu Hureyra, Bouqras, Ramad, Asswad, Jericho, Ain Ghazal, and Netiv Hagdud) was mainly formed by the natural collapse and subsequent rebuilding of adobe or mud-brick houses. The noncontemporary erection of mud houses (sometimes on stone foundations) created what is known in archaeological jargon as “spiral stratigraphy”; new building layers are not recognizable over the entire area of a given tell, each excavation area having its own sequence of building events. On rare occasions general planning can be seen (as in the case of Tell Bouqras), but it is in no way interpretable as rebuilding after total destruction caused by warfare.’

Bocquentin, F. 2003 Pratiques funéraires, paramètres biologiques et identités culturelles au Natoufien : une analyse archéo-anthropologique. PhD these. University of Bordeaux.

Page 456 presents a summary of findings on trauma in Natufian skeletons of Israel.

Bocquentin, F. & Bar-Yosef, O 2004 Early Natufian remains: evidence for physical conflict from Mt. Carmel, Israel, J. Human Evolution 47, 19-23.

Arrow stuck in spine of Epipalaeolithic man.

Clare, L. et al. 2008 Warfare in Late Neolithic/Early Chalcolithic Pisidia, southwestern Turkey. Climate induced social unrest in the late 7th millennium cal BC, Documenta Praehistorica 35, 65-92.

Discussion of the PPNC period sites of western Turkey.

Dougherty, S.P. & Tsuneki, A. 2016 Injury-Related Morbidity and Mortality in Neolithic Syria, 85th Annual Meeting of the Am. Assoc. Physical Anthropologists Abstracts.

More info on Tell Ain El-Kerkh.

Erdal, Y.S. & Erdal, Ö.D. 2012 Organized violence in Anatolia: A retrospective research on the injuries from the Neolithic to Early Bronze Age, Int. J. Palaeopathology 2, 78-92.

A very informative paper on Turkish trauma.

Eshed, V. et al. 2010 Paleopathology and the Origin of Agriculture in the Levant, Wiley Interscience.

Evidence of trauma (as well as disease and general health) in Levantine Natufian and Neolithic pops. I couldn’t find more detail on the one cranial trauma of the Neolithic, but it could be anywhere from PPNB to PPNC.

Eshed, V. 2001 From Foraging to Farming in the Holocene (The Pre Pottery Neolithic Period 8,300 – 5,600 B.C) in the Southern Levant: The Skeletal Evidence. PhD, Tel-Aviv, pp215.

Tables on p131-132 and 139-140 give individuals and sites for cranial trauma in Eshed et al 2010 (but in Hebrew).

Ferguson, R. B. 2015 The Prehistory of War and Peace in Europe and the Near East. In: ‘War, Peace, and Human Nature: The Convergence of Evolutionary and Cultural Views’ (ed. D.P. Fry.), p191-

A good review with interesting detail, including the Ghwair skeleton. However, I suspect that some of the dates are mis-applied between PPNA and PPNB.

Glencross, B. & Boz, B. 2013 Representing Violence in Anatolia and the Near East During the Transition to Agriculture: Readings from contextualized human skeletal remains. In: (Knüsel, C. & Smith, M. eds.) ‘The Routledge Handbook of the Bioarchaeology of Human Conflict’, Routledge, p90-108.

Excellent source of info. The table is terribly formatted, however, with inconsistent dating and information on Çatal Höyük lost due to wrapping text causing information lines to go out sync.

Hammer, E.L. & Arbuckle, B.S. 10,000 years of pastoralism in Anatolia: a review of evidence for variability in pastoral lifeways, In: Nomadic Peoples (web verson), pp43.

Review of early attempts at pastoralism.

Kansa, S.W. et al. 2009 Whose Bones are those? Preliminary Comparative Analysis of Fragmented Human and Animal Bones in the “Death Pit” at Domuztepe, a Late Neolithic Settlement in Southeastern Turkey, Anthropozoologica 44, 159-172.

Cannibalism in Late Neolithic (5800 to 5400BC) Turkey.

Knapp, A.B. 2013 The Archaeology of Cyprus: From Earliest Prehistory Through the Bronze Age, Cambridge, pp640.

See p127 for discussion of walls and possibly conflict-related injury at Khirokitia.

Kornienko, T.V. ? 2015 On the Problem of Human Sacrifice in Northern Mesopotamia in the Pre-Pottery Neolithic, Archaeology, Ethnology and Anthropology of Eurasia 43, 42-49.

Human sacrifice evidence at Jerf-el-Ahmar, Cayonu.

Kozlowski S. K. 1989 Nemrik 9, a PPN Neolithic site in Northern Iraq, Paléorient 15, 25-31.

Inhabited between 9000BC and around 6500 BC. Evidence of violence from arrow wounds happens in the latest phase (at the end of occupation).

Kuijt, I. & Goring-Morris, A.N. 2002 Foraging, Farming, and Social Complexity in the Pre-Pottery Neolithic of the Southern Levant: A Review and Synthesis, J. World Prehistory 16, 361-440.

Reference to architectural trends (e.g. house clustering and ‘long walls’).

Lahr, M.M. & Haydenblit, R. 1995 The Human Remains from the site of Et-Tin, Israel. Paléorient 21, 97-111.

Maisels, C.K. 1999 Early Civilisations of the Old World, Routledge.

p126 – Tell Maghaliya levels 13-14, final levels, show evidence of circuit wall. These date perhaps 6800-6500BC.

Maisels, C.K. 2005 The Near East: Archaeology in the ‘Cradle of Civilisation’, Routledge, pp256.

Of Tell Maghzaliya: ‘A defensive wall, whose footing consists of edgewise limestone blocks up to 1.5m thick, and which incorporates at least one tower, runs along the western slope of the tell, closing off the settlement…’

Mazurowski, R.F. et al. 2009 Chronology of the Early Pre-pottery Neolithic Settlement Tell Qaramel, Northern Syria, in the Light of Radiocarbon Dating, Radiocarbon 51, 771-781.

The towers of Tell Qaramel (lovely name).

Mellaart, J. 1967 Çatal Hüyük: A Neolithic Town in Anatolia, McGraw-Hill, pp232.

p68-69 for quote.

Moore, A.T.M., Hillman & Legge 2000 Village on the Euphrates: from foraging to farming at Abu Hureyra, Oxford, pp585.

See p268 for description of individual with arrow in Abu Hureyra, p255 and 257 for estimated chronology.

Peterson, J. 2002 Sexual Revolutions: Gender and Labor at the Dawn of Agriculture, Rowman Altamira, pp192.

See p82 to 84 for reference to Natufian (unseen), Basta and Ain Ghazal trauma.

Peterson, J. 2010 Domesticating gender: Neolithic patterns from the southern Levant, J. Anthropological Archaeology 29, p249-264.

Perlès, C. 2001 The Early Neolithic in Greece: The first farming communities in Europe, Cambridge, pp356.

Roper, M.K. 1975 Evidence of Warfare in the Near East from 10,000-4,300 BC. In: (Nettleship, M. A. & Givens, D. eds.) ‘War, its Causes and Correlates’, De Gruyter, p299-344.

Schmidt K. 2006 Animals and headless man at Göbekli Tepe. Neo-Lithics 2, 38–40.

Simmons, A.H. & Najjar, M. 2006 Ghwair I: A small, complex Neolithic community in Southern Jordan, J. Field Archaeology 31, 77-95.

Simmons, A.H. & Najjar, M. 2003 Ghuwayr I: A Pre-Pottery Neolithic B Settlement in Southern Jordan: Report of the 1996-2000 campaigns, Annual Report of the Department of Antiquities of Jordan 47, 407-430.

P90 of the 1st and p422 of the 2nd contain details of a projectile point in a mature female skull. The report makes clear the uncertain context of the burial implied in the paper.

Slon, V. et al. 2014 The Plastered Skulls from the Pre-Pottery Neolithic B Site of Yiftahel (Israel) – A Computed Tomography-Based Analysis, PLOS One.

Solecki, R.S. et al. 2004 The Proto-Neolithic Cemetery in Shanidar Cave, Texas A&M, pp234.

Sołtysiak, A. et al. 2015 Human remains from Nemrik, Iraq. An insight into living conditions and burial customs in a Pre-Pottery Neolithic village, Paléorient 41, 101-114.

Doesn’t mention the arrow heads but does discuss trauma.

Sołtysiak, A. 2015 Antemortem Cranial Trauma in Ancient Mesopotamia, Int. J. Osteoarchaeology

Sites evidence for trauma in one of the three documented Preceramic sites of Mesopotamia. Zawi Chemi (Zagros PPNA) has ACT in 4 of 9 buried individuals (two men, one young woman and a child) (42%). Of the 39 in Abu Hureyra (PPNA) and 28 in Nemrik 9 (PPNA & PPNB), no-one has ACT (0%). Chalcolithic sites 12% (74 individuals). Normal Bronze age rates in Anatolia and Jordan quotes as between 9 and 30%. His Early Bronze age rates are about 6%.

Tsuneki, A. 2011 A glimpse of human life from the Neolithic cemetery at Tell el-Kerkh, Northwest Syria, Documenta Praehistorica 38,83-95.

References for other dates and places

Bachechi, L. et al. 1997 An Arrow-Caused Lesion in a Late Upper Palaeolithic Human Pelvis, Current Anthropology 38, p135-140.

A skeleton from Sicily, dated to some little time before 14,000 BC, so Palaeolithic.

Lahr, M. M. et al. 2016 Inter-group violence among early Holocene hunter-gatherers of West Turkana, Kenya, Nature 529,394–398.

Meiklejohn, C. et al. 2009 From single graves to cemeteries: an initial look at chronology in Mesolithic burial practice, In: Mesolithic Horizons (McCartan, S. et al. eds.), Oxbow, 639-649.

Dates of Vasil’evka III site as 9310-9990 BC cal.

Shaw, I. & Jameson, R. 2002 A Dictionary of Archaeology, Blackwell, pp640.

Information of Vasil’evka III burials. ~p607

Stock, J.T. et al. 2005 F-81 skeleton from Wadi Mataha, Jordan, and its bearing on human variability in the Epipaleolithic of the Levant, Am. J. Physical Anthropology 128, p453-65.

Evidence of man tied up with hands and feet at the back, then buried face down, dated roughly 15,000 BC. Drawn attention to by Glencross and Boz.

Additional references

Lambert, P.M. 1997 Patterns of violence in prehistoric hunter-gatherer societies of coastal southern California, In: ‘Troubled Times: Violence and Warfare in the Past’
(eds. Debra L. Martin, David W. Frayer), Psychology, 77-109.

Very interesting comparative study of evidence for healed cranial fracture, which shows many similar features in terms of age and sex distribution. There is also a talk by her on this on youtube.

References not using evidence

Elliot, S. 2012 The Walls that Did Not Come Tumbling Down: Are the Early Neolithic Walls of Jericho the First Evidence of Warfare? RUSI J. 157, 72-79.

Probably not.

Rowthorne, R. & Seabright, P. 2010 Property Rights, Warfare and the Neolithic Transition, TSE working paper

Prisoner’s dilemma theory of agriculture adoption individual pro’s vs defence cons, which assumes defence was important.



{ 8 comments… read them below or add one }

elicia david October 23, 2018 at 6:40 pm

Whenever I find sites crowing about ancient violence showing head injuries, with pictures, the pictures look suspiciously like the people have been trepanned. A very common practice for millennia. It was incredibly common in the Neolithic. This is the danger of male archaeologists, foaming at the mouth to prove that humans have always been as violent as we all are, but ignorant of the very different mindset in prehistory, of telling us what happened.


Edward Pegler November 1, 2018 at 5:47 pm

Thanks for the comment. They’re as few and far between as my posts these days. Trepanning seems to be putting a hole in one’s head for surgical or spiritual purposes. I guess that might look like a spear wound or some such violent encounter. However, the marks given as cranial trauma for the Early Neolithic are almost invariably crushed areas of the skull with cracks and ring structures. These couldn’t be the result of any kind of surgery. Either way, the point of the post was to suggest that violence levels were much lower in the Early Neolithic than they are later, contrary to the expectations of hawkish archaeologists. I occasionally foam, but mostly from my nose.


Matt December 14, 2016 at 4:11 am

The dating and the extreme violence at Domuztepe seems to fit in well with your sense of increasing violence after about 7000. You referenced it, but didn’t discuss, I am curious.

I don’t find the number of healed cranial fractures relative to the perimortem ones particularly strange. Firstly, if slings were in widespread use, the two sides might not have ever gotten close enough to finish off the wounded. I wonder where on the skulls the depressions are? A study of Chumash clubbing injuries suggests that most will be frontal anyway. Secondly, as the victorious skirmishing line advanced they would tend to be in possession of the dead and wounded of both sides. I would assume that the in-group wounded would get assistance and the in-group dead would get a prompt burial.

25:00 minute mark-



Edward Pegler December 14, 2016 at 6:15 pm

Dear Matt

Thanks for the useful comments. Just started watching the video, which weighs in at an hour, so I’ll be watching that a little later.
Domuztepe was left in the reference list for me as I thought it was interesting, but it fell just after the period of interest, around 5800BC or so, I think. However, the settlement did exist during the PPNC so got on the map.

I entirely take your point about the head injuries. They are variously located on the head, some frontal, some parietal on no particular side. Largely the ones I’ve seen photos of are not severe (they’re often quite difficult to make out). I presume that what you’re arguing is that the bodies of enemies won’t be buried so are lost to us, which is quite possible as long as the enemy takes posession of the territory.

Where I think that this is unlikely is that settlement continuity is very high in the PPNA PPNB, so there’s little evidence of sites changing hands. As for hunter-gatherer invasions hg’s would, I guess, retreat after an attack, so bodies of hg attack victims would tend to be left in the hands of the settlers. Hence you should see plenty of buried, unhealed fractures. They exist but aren’t common. On the other hand, if these were just skirmishes, with little killing, then the evidence would fit fine.

As for slings they may well have been used, as they were a preferred weapon in the Near East, but their acme appears to be later, in the PPNC and after. What I didn’t discuss is the use of maceheads, and it would probably be worth updating the info on these and slings in the post.

thanks once again and do correct me if I’ve misunderstood.



Matt December 15, 2016 at 12:51 am

A few comments in reply. The video addresses headwounds caused by “war clubs”, particularly location thereof at the 25 minute mark… I figured that a lot of this material was New World and perhaps outside of your interest. I think you got my point about the victors of a hand to hand battle, the side that advances (wins?), can prevent the enemy from finishing off their wounded, giving them a chance at recovery. What I was trying to say is that for the defeated dead and wounded it was, perhaps, sky funerals all round. Or burial in a place that archaeologists aren’t expecting. Many people buried in houses in that place and time, no? Perhaps you had to be related to the living family to get buried there.

I accept what you say about continuity of settlement meaning that attackers of whatever description were rarely able or wanted to dislodge whole villages, so the defenders must have tended to win and/or the fighting was over the periphery of the croplands? A more vulnerable spot in my mental model would be that I have assumed the fractures were due to violence that was mainly organized, periodic, and in between groups.

Lastly, I would love to read a blogpost of yours about neolithic maceheads and slings of the region. If they hadn’t been dug up almost everywhere, I would have thought the stone macehead to be rather improbable object.


Edward Pegler December 15, 2016 at 6:37 pm

Dear Matt

Thanks once again. I’ve just watched the video you recommended by Patricia Lambert (this one is available on its own, about 21 mins long, at This video is so useful and really helped me understand. It made clear just what the significance of death assemblages were (i.e. people who survive don’t get buried of course, so healed wounds are not necessarily representative of the age at death). Also I found the rather brutal Yanomano technique for ‘resolving conflict’ within groups or with individual strangers a real eye opener.

In terms of the data, I think I need to go back to the sources again (sigh) to check on positions of injury. Bonnie Glencross’ analysis of injury position is what I’ve used so far, but maybe it’s not applicable to everything. And as for the macehead and sling stuff, I will get on to that. However, whereas maceheads are relatively clear people have a wide range of views about the uses of clay sling bullets.

thanks once again



Matt December 16, 2016 at 3:41 am


Again, you are very welcome. I do have some tangential questions and ideas about sticks and stones… Perhaps you will enjoy addressing them. I’ll start with some questions. Do objects identified in the literature as maceheads ever predate nearby agriculture by very much? There are suspiciously sized “bored stones” back as far as the paleolithic which are identified as “net weights” and the like. If all these toroidal objects have been correctly identified, what could make early farmers so much more enthusiastic about stone maces as weapons compared to hgs? Could “maceheads” be used for grain processing or labor based currency in the neolithic?

If one had to fight with meso/neo lithic technology, I would not want to slow down my offensive swing or my defensive parry by putting a rock on my stick. If I was willing to carry more weight, I would rather have more reach to my club like a Yanomano quarterstaff. A piece of wood swiftly applied to an unprotected head will almost always hit hard enough to end effective resistance, I think. Speed and reach are much more important (than overkill) to preserve one’s own life. This is why I said that I thought stone maces somewhat improbable. Does the presence of a bunch of actual stone maceheads used for combat imply the presence of defensive headgear, possibly fibrous, on one’s likely enemies?

Edward Pegler December 16, 2016 at 5:33 pm

Well, at the moment I have no idea, but I’ll start looking. They all seem like good questions.

However from a pure engineering point of view the mace appears to be a compromise, allowing the carrying of a shorter weapon (a bit like a club-hammer I suppose) but trying to maintain a large momentum. In that respect it might be useful for close combat. The obvious disadvantage is its durability, which wouldn’t be high with a stone head. It seems to me they’d just keep falling apart. I’d rather use a blade (well, not me personally).

Whatever, I’ll get back to you on this.



Leave a Comment

Previous post:

Next post: