Violence in the Near East at the dawn of agriculture: conflict vs co-operation

by Edward Pegler on 22 December, 2016

Small population size and little need for competition is an excellent explanation for the lack of conflict during the transition to agriculture. However, there’s also another possible explanation – co-operation.

Jericho, 9000 BC

Jericho, 9000 BC

In my previous post I tried to gather together the evidence for violence in the transition from foraging to farming across the Near East. Much of it is known by academics but it’s rarely put together.

What the evidence appears to show is that there is a fall in violence during the transition to agriculture in the Near East and that this lasts for perhaps 2,000 years. The fall seems to start early, even with the first experiments in agriculture. However, violent conflict increases with the increase in inequality and more complex agricultural methods.

Such a pattern is recognisable in other instances of agriculture, for example in Eastern North America, with conflict reducing gradually around 1,000 BC and rising again around 2,000 years later around 1,000 AD.

The simple explanation

There is a perfectly simple explanation for the fall in violence as agriculture started. As Jonathan Haas argues for Eastern North America, the initial state of violence before agriculture was due to the land reaching its holding capacity. There was no spare land to feed more foragers. As with Robert Carneiro’s model, people turned inward to battle each other.

But when agriculture, or at least some sort of horticulture, started to be practised then the strain caused by food shortage started to ease. Individual groups no longer needed to be at each other’s throats. As this period of plenty started the fear of war lifted as everyone could feed themselves.

However, once agriculture became established then agricultural populations gradually rose. Eventually, there was a return to the old situation where the holding capacity of the land was reached, but now, because of agriculture, this was at a much higher population density. Tribes fought for space and food in the new, fully farmed landscape. Thus, conflict returned.

There is nothing wrong with this model. It’s explanatory power is excellent. Furthermore, Ian Kuijt has made a good case for rising populations all through this period in the Levant, something which only slowed or reversed at the time of increasing signs of violence, in the PPNC.

So why am I writing this post?

The problems

There are two things which are nagging at me:

  • Why is it that in some places in the world the transition to agriculture led to a fall in violence, as seen above, but in other places (in fact more often than not) it led to a rise? If the story above were true, shouldn’t a pattern of falling conflict be seen everywhere?
  • Why is it that during such a period of low food stress we see a rise in the uniformity of religion and evidence for increased trade between groups. Why would they need to bother?

For the second point this circumstance is particularly noticeable in the Near East. Trade in projectile points, obsidian and other minerals and stones (e.g. greenstone) as well as copper noticeably increases during the transition to agriculture, so much so that the Near Eastern Levant region was dubbed the ‘PPNB Interaction Sphere’ by Ofer Bar-Yosef and Anna Belfer-Cohen.

Additionally, a uniformity of culture and, in all likelihood, religion settled across the region, especially from Middle PPNB times, at the time when agriculture became reality.

This is quite extraordinary when it’s realised that many of the agricultural settlements were located around very localised, well-watered alluvial fans but were separated by large stretches of agriculture-free land. Why should these isolated farmers need to have much to do with each other?

Co-operation – the more complex explanation

Foraging was a relatively small-scale system, with groups of less than 100 operating in a way that they believed to be independent of other groups. Interactions with other groups were often for acquiring sexual partners. Groups were relatively mobile and tended to move to where the supply of food was most promising. Groups had little reason to be friendly with other groups. This pattern of behaviour changed when people started to settle in one place, therefore needed to rely more on the foods available locally.

Unreliable agriculture

Early agriculture was highly unreliable. If each group of people were simply to work with the cereals and, perhaps, animals that they started to cultivate or raise then the chances of catastrophic failure would be high for most settlements.

Most early farmers still relied to a large extent on hunting and the collection of grains. However, these were also of limited supply and, with the rising populations seen in early agriculture, local sources could have been exhausted quite quickly during a run of bad harvests for a settlement.

Pooling resources

However, an alternative to self-reliance would be co-operation with other groups. Like being a member of the Masons, a church or any other similar collective establishment, buying into a larger scale grouping than the local tribe allowed people a chance not to fail in their ‘start-up business’.

Religion provided a commonality between groups across the Near East. While they may have spoken different languages, they shared ‘PPN’ism, the religious rituals which led to particular burial rites, etc. These were important in letting people from different settlements know how to behave and interact with each other when they met.

Perhaps more importantly, in any settlements’ time of need due to crop failure, loans from other settlements could be made in grain or other surplus. These loans might be said to be gifts but were in effect to be paid back in due course, even if at low or zero rates of interest. Payment could be again made in grain or other surplus in better future years.

All of this may have been ‘accounted’ by the exchange of less useful gifts, whether stone or metal beads or some other items, these gifts being in recognition of the debt owed. In effect these would act as a form of money. Each settlement would have such items in their repertoire as insurance, a kind of ‘bank of last resort’ during a crisis.

Through such a process of mutual support, agricultural failure in any one settlement would not lead to catastrophe for its inhabitants. Better, higher yield grains and more productive animals could be brought in and, if successfully grown even passed on from community to community. Given enough time (say 2,000 years), individual settlements could become increasingly self-reliant.

The rise of independence

And this, of course, would be where the trouble started. As settlements became more independent, the need to be part of the larger system would fade. Groups would have less reason to deal with each other and more reason to return to the normal state of humanity – suspicion and hostility toward strangers.

Interestingly, the other consequence of this, apart from isolationism, would be the loss of importance of the gift items which were exchanged between groups. And, sure enough, one of the features of the Late PPNB is the burial of individuals with copper beads, stone pendants and other ‘luxuries’, something which had never occurred previously. These items were no longer essential to the greater community – they had become items of individual status and wealth.


What I’m arguing for here only makes sense if agriculture developed locally. If agriculture was imported, fully formed, from another area then there would be no reason to see this need for co-operation between groups, at least not for long. This may explain why the Eastern North American experiment with farming also showed the same pattern of falling violence, whereas many other locations did not.

What I’m further arguing is that warfare may be a result of there being only low levels of risk for the initiator. If you knew the risk was great that you’d lose you probably wouldn’t start a war. Early, experimental farming communities had everything to lose by picking a fight with their neighbours. Co-operation was the only sensible strategy. Those groups of farmers that didn’t co-operate with others, or even chose to be hostile to them, would probably die out quickly without the support of their neighbours.

Conversely, the increasing independance of communities as time went on, whether because of better grains or animals or manuring or improved cookery techniques (e.g. using pots to cook), meant that these groups were much less at risk if they didn’t co-operate. They had much less to lose by choosing to be aggressive and, perhaps, quite a lot to gain.

I’m not saying that communities were ever completely independent; such a state is extremely unlikely. Communities often need something, for example salt, from outside. It’s just that such trade could be done by middlemen, and groups need never worry about their acts of aggression preventing that trade.


Taking a view of co-operation like the one I’ve suggested above sounds a bit like those who said Europe was too interlinked for war in 1914. It’s true that this didn’t stop war breaking out. However, it’s also true that, from an economic point of view, WWI benefitted almost no-one in Europe. After half a century and the loss of millions of lives the concensus in Europe was for a return to co-operation. Half a century later and the mood appears to be turning daft again.

So maybe early farmers were just as daft, not taking potential risks into account either. They may simply have ‘learned’ not to fight through a process of artificial selection – choose to be the aggressor and you will die out, due to your supplies being cut off. We haven’t had that level of selection recently – if we choose to be the aggressor now we might lose the war but the effects are generally not terminal. Our method of learning has been through education, but education is not as harsh a tutor as being wiped out. This is perhaps why we keep making the same mistakes.

At other times in history the relative independance of groups or states has meant that the advantage was with the most aggressive aggressor (and, to a lesser extent, with the most docile and biddable). A way to keep yourself supplied was everything when launching a successful war. Either way, there is no simple rule for whether to be a ‘hawk’ or a ‘dove’ in your action – it all depends.


Bar-Yosef, O. & Belfer-Cohen, A. 1989 The Levantine ‘‘PPNB’’ interaction sphere. In: I. Hershkovitz (Ed.), ‘People and culture in change: Proceedings of the second symposium on Upper Paleolithic, Mesolithic and Neolithic populations of Europe and the Mediterranean Basin’. BAR International Series 508(i) (pp. 59–72). Oxford: Archaeopress.

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

Comments on fall and rise in violence at the transition to agriculture, p122?

Haas, J. 1999 The Origins of War And Ethnic Violence, In: ‘Ancient Warfare: Archaeological Perspectives’ (Carman, J. & A. Harding, eds.), Sutton, 11-24.

Kuijt, I. 2008 Demography and Storage Systems During the Southern Levantine Neolithic Demographic Transition, In: The Neolithic Demographic Transition and its Consequences (Bocquet-Appel, J.P. & O. Bar-Yosef, eds.), 287-313.



{ 17 comments… read them below or add one }

Matt December 30, 2016 at 3:55 am

I think cooperation for the rational reasons described runs into a few problems in the context of early agriculture. Most of the problems that I can think of, weather, locusts would hit regions rather than villages. This would reduce the rational value of close cooperation.

Let me suggest an alternative cooperation mechanism to explain the pattern of violence…. consanguinity. Agriculture was adopted preferentially by families whose numbers and influence then expanded during phases like PPNB when “group feeling” (in an Ibn Khaldun sense) was high. After a few centuries or millennia of sedentary life in different villages, group feeling, and probably inter-village consanguinity itself, declines and violence is more likely to spiral into feud. Other places like the LBK-Atlantic frontier on the Rhine don’t see a detectable drop in violence due to the character of the local hgs. On a practical level, in the fertile crescent and the Eastern US, do you think this is testable?


Edward Pegler January 1, 2017 at 2:38 pm

I think that if it were mainly a weather phenomenon then what you say about all villages suffering equally and no reason for co-operation would be fair. This was a bit of a problem with another post I wrote before. However, my thought was that if everybody was experimenting with different cultivars or animals, or experimenting with the same ones in different ways, then differences between neighbouring villages or settlements’ harvests could have been large, even if the weather was the same. Hence the need for co-operation as advantages would be for different settlements in different years.

As for consanguinity there must always be some blood-links between neighbouring settlements, even if it’s just because men pinched other settlements’ women, Sabine-style, so I presume you’re talking about dynastic links and ancestors and all that jazz.

I don’t know what the evidence is for this, but I would have thought that any loyalty would last little more than a couple of generations between different settements with the same ancestors (it’s hard enough getting me and my sisters to co-operate without losing it to be honest). But a thousand or more years suggests a different mechanism, where blood-links may play a vital part but as an aspect of something which binds all the different settlements together.

Any further thoughts on this?



Matt January 3, 2017 at 4:50 am

Perhaps sharing information itself about cultivars was itself a mechanism that selected for cooperation and reinforced it when already established. People who manage to have good relationships with everyone are poised to thrive when information is valuable. Maintaining group feeling seems difficult in today’s complicated world, but I haven’t had my ancestors’ skulls plastered either. I hate to use religion as a deus ex machina, but I guess that is where I would go next.


Edward Pegler January 4, 2017 at 6:06 pm

Yes, I think that’s reasonable, and I think I’m thinking much the same about today’s world where groups are not local and probably have limited or no understanding of the consequences of their actions.

As for religion I prefer it to be an emergent property of other factors. However, if enough people buy into religions, as happens later in history, then they can have a momentum of their own. I wonder how united the religious view of the early farmers was?

Gawd, this sounds a bit cryptic. It usually means I don’t know what I’m talking about.



Matt January 4, 2017 at 7:06 pm

Yes, after thinking about it for a while we seem to be on the same page. I believe that some positive value for cooperation best explains the data and that the carrying capacity type of explanations not really sufficient.

The content of religion pre-history will always be at least a little difficult to agree on, so looking at it as an “emergent property of other factors” may be a good practical way of dealing with it.

GT indicates to me that, at least locally, there was sufficient unity to build a communal site. Some of the unique images seem to have replicated to Catalhuyuk according to some. But I probably bring my post monotheistic assumptions of religion which may expect and rely on more doctrinal unity than the PPNB may have ever had.

Edward Pegler January 1, 2017 at 2:50 pm

“Other places like the LBK-Atlantic frontier on the Rhine don’t see a detectable drop in violence due to the character of the local hgs. On a practical level, in the fertile crescent and the Eastern US, do you think this is testable?”

Yes, I think that the idea presented above it testable up to a point. It would make the prediction that violence should fall for a significant period only in places where pristine agriculture is practised. This would include the Near East, China, Mesoamerica, South America, NE US, parts of West-Central Africa and, perhaps, New Guinea. However, it wouldn’t include Europe, SW US, India, Central Asia, North Africa.

How easy it would be to test such an idea would depend on the ability to recover bones or other evidence for or against violence. One of the major problems with some tropical parts of the world is the very poor recovery of bones or, for that matter, anything but phytoliths, sherds and stone tools. This tends to limit the survey’s extent and perhaps makes it statistically not useful. However, examples of any contrary situation in the remaining locations (e.g. South America, China) should be enough to prove the argument wrong.

Does that answer the question?


Matt January 3, 2017 at 5:06 am

Violence is pretty hard to consistently measure with osteoarcheology because weapons are cultural and some cultures prefer spears, etc. Consanguinity might be more easily observed, but I am no expert on what can be done with DNA.



Matt December 30, 2016 at 3:01 am

If figs were the first crop then the return was delayed indeed.


Edward Pegler January 1, 2017 at 2:22 pm

Dear Matt

That is brilliant! I will, I hope, enjoy reading that.


Robert Kerr December 25, 2016 at 7:07 pm

Sorry Ned, Xmas intervened.
Surely the spread of neolithic DNA shows that the settled agricultural communities did in fact ‘move’, in the sense that new settlements must have sprung up to cover Anatolia and thence west and north. Whether the new settlements were pacific ‘calvings’ or due to the overpopulation of a given region and resulting stress would no doubt vary. I don’t see how we can be prescriptive and other human factors no doubt played a part. The wheats and barleys of Anatolia adapted well to northern and to southern europe; thus armed, the new immigrants clearly had the wherewithal to move into existing hunter/gatherer territory and continue to spawn new settlements for four thousand years, until cultivable land became scarcer. Most conflict would surely be between the farmers and the hunters, though there was no doubt plenty of scope for trade.


Edward Pegler December 28, 2016 at 7:35 pm

I’m even later with my reply.

Yep. I agree with everything you’ve said. I also very much take your point about conflict with the hg’s.

What I’m curious about is the bit before agriculture, the Late Epipalaeolithic, when Natufian settlements were appearing and the early phases of sites like Abu Hureyra. So, for example, did the settlers have more of a reason to stay there than some not particularly appetising grass, e.g. they were on a gazelle migration route between, say, africa and the middle east or saudi and the middle east?

As for conflict with hgs, I almost get the impression that the people who settled were they only people in the Levant… that they didn’t get attacked by the hg’s there because the hg’s were them. I don’t know. I need to read more on this. Any thoughts?



Robert Kerr December 29, 2016 at 8:28 am

From what I’ve read, Ned, there is an environmental argument that the Natufians learned to select the best ground and cultivate in a more technical sense as the cold, dry (?) conditions of the Younger Dryas made their ‘looser’ proto-farming methods inadequate. It possibly meant they were indeed the only population group to continue living there in any numbers, so had a ‘head start’ when conditions ameliorated.
I would assume that the gazelle migration routes would have shifted south during the Dryas episode and that must suggest that their simple farming methods provided enough food. It does seem hard to believe.
I know the spread of farming settlements out of the Anatolian region took thousands of years and must obviously have generated a mix of reactions from hunter-gatherers. From what I’ve read they co-existed and traded their different products for upwards of a thousand years in northern Europe; whether that happened at all in Anatolia and the Levant I know not.
Gubekli Tepe remains supremely enigmatic, whatever happened!


Edward Pegler January 1, 2017 at 2:56 pm

Happy New Year, Robert!

Thanks for the further helpful comments about the post and sorry I didn’t get back to you before.

My major regret when visiting Eastern Turkey was not to go to Gobekli. Mind you I was ill enough that I could scarcely tie my shoelaces so maybe it’s not surprising.


Robert Kerr December 22, 2016 at 9:17 pm

This analysis seems to take no cognizance of the findings from Gubekli Tepe, where there is no evidence of agriculture yet an enormous amount of cooperation drawing upon a wide geographical area was involved in constructing megalithic structures. The design and layout would appear to have been according to a commonly accepted symbolism by the hunters and gatherers, who may not even have lived for more than part of the year in large social groups. Clearly agriculture, in all but the most self-sufficient of settlements, would make the possibility of trade between established communities more likely but unless it can be shown that in fact settled agriculture had developed there is every possibility Gubekli Tepe is a clear indicator of very advanced social cohesion before the existence of permanent agricultural settlements.


Edward Pegler December 23, 2016 at 2:05 pm

Dear Robert

Yes. You’re right about Gobekli Tepe and, probably, about the co-operative nature of its construction. It dates from PPNA times, before clear signs of domestication of cereals (which occurs in the MPPNB). However, there has been speculation over the years about whether limited cultivation (horticulture or agriculture) was practised back in PPNA times, if for no other reason than that the sedentary communities in the Levant were showing signs of population expansion (community size, density, burials etc) and that this might have been difficult to manage without some cultivation.

Gobekli itself, not being a settlement, has no signs of anything to do with cultivation or agriculture. However, there were plenty of settlements around at this time (including Hallan Cemi & Cayonu, quite close to Gobekli). The change from foragers (and hunters) to pure farmers was a long process, taking perhaps 3000 years to complete. People at the beginning of the process, in the Natufian, would look just like foragers, those of the PPNA largely like foragers and perhaps a little like farmers, as would be the case for the builders of Gobekli, the people of MPPNB times…

As a slightly provocative idea, you could argue that the fact that Gobekli is covered in pictures of wild animals is not because that’s all people lived on. An analogy is people of the industrial revolution taking a keen interest in landscape. You only notice things when you start to lose them.

What I personally find difficult to understand is why anybody would bother to settle the region in the first place. It limits your options so much more than moving on.

best wishes Ned


Robert Kerr December 23, 2016 at 10:21 pm

Thanks, Ned,
You wonder why people would want to settle the area. Not sure what you mean: was the area not rich in harvestable wild grasses, i.e. the cereals that later became the mainstay of agriculture?
What puzzles me is how such highly developed stone working skills could have developed in a region that had relatively recently been in the grip of the Younger Dryas and presumably supported few people. Could the creators of GT not have moved in as conditions ameliorated, having developed their skills further south into the ‘fertile crescent’?


Edward Pegler December 24, 2016 at 8:00 am

Dear Robert

There are some useful comments on the limestone slabs at I slightly take issue with the idea in the article that limestone is always ‘so soft’. Try a good dolomite. However, carving and shaping the stone is a skill that could be learned relatively quickly as long as you had the inspiration: where that came from is fascinating – maybe we’re still missing something. Either way, it’s perfectly possible that the people originally came from the southern Levant.

My issue with the argument about the special grasses of the region is that, firstly, they are spread across a larger region than just the Levant, many occurring across southern Turkey. Either way, I wonder if this is not just an argument from hindsight. If the first farmers had appeared in India and domesticated a particular suite of grains or seeds there, wouldn’t we naturally look at the distributions of those wild grains and seeds and think that there was something special in them. This is true of Mesoamerica, and that bizarre cultivation of teosinte.

Harlan’s argument is that people could easily feed themselves by harvesting the wild grasses of the Levant. This is true of many forager locations and might cause people to temporarily settle. The problem comes once the population rises to a point where the grasses can’t feed them all. Then populations (at least some of them) normally return to moving again. Why did the forager settlers of the southern and eventually northern Levant insist on staying put? Maybe it was just luck that someone came up with the idea of cultivation, but the delayed return would tend to put them off.

Sorry about that rant, but I still don’t see the whole picture.

Apart from that I forgot to say in the last reply that the problem of co-operation before agriculture that you rightly mentioned in your previous comment applies to both of the explanations given in the post. Perhaps it’s worth pointing out Keith Otterbein’s work, where he says that co-operation is an essential prerequisite for farming as you need the stability to get a harvest that someone doesn’t just steal. However, that doesn’t explain why a similar phenomenon happens in Eastern North America.

Sorry. That was a bit long.

Happy Christmas



Leave a Comment

Previous post:

Next post: