Warfare and the transition to agriculture


Jason Collins


October 25, 2013

As I pointed out in my last post, Jared Diamond called the transition to agriculture the worst mistake in the history of the human race. Yet despite evidence that the early adopters of agriculture were shorter and had worse nutrition and health than the hunter-gatherers that preceded them, the agricultural way of life came to dominate human society. Health may have rebounded as humans adapted to their new lifestyles, and the long-term benefits are now clear, but why would the early adopters of agriculture shift to what would appear to be a much a poorer way of life?

One answer to that question is offered by Paul Seabright and Robert Rowthorn in a 2010 working paper. Seabright and Rowthorn’s idea is that early farming communities needed to develop defences to protect their stationary stores of food. With increased defensive ability comes increased offensive potential and ability to steal from neighbours. As a result, it is optimal for an individual group to adopt agriculture where the benefit from theft exceeds the cost of moving to agriculture.

As neighbouring groups face the same incentives (and hunter-gatherers next to agriculturalists are now experiencing increased theft), all may adopt agriculture. However, once all groups adopt agriculture and the associated defences, there is no longer opportunity to steal to make up for the costs of agriculture. This is effectively a prisoner’s dilemma where it is optimal for each party to adopt agriculture regardless of what the other party does but when all parties adopt, they are all worse off than they would be if no-one adopted.

Seabright and Rowthorn develop a model that demonstrates this point and they considered that the result held for a variety of parameter values (including potential benefits from theft, costs of defences and the difference in productivity between agriculture and the hunter-gatherer lifestyle). They noted that the model could also be applied to other scenarios involving investment in defence which may also have offensive capability.

I have some sympathy for this argument. Archaeological evidence from some of the earliest agricultural settlements indicate the development of defences. Agriculture allows for specialisation, including specialisation in developing the tools of theft and war.

However, I am still to be convinced that this argument is necessary to explain the adoption of agriculture. If, even with their lower health, agricultural communities have higher numbers of children, they will come to form a larger part of the population. Given that many hunter-gatherers spaced births due to their nomadic lifestyle, an agricultural existence may allow for more births, outweighing the health costs of the agricultural existence. Hunter-gatherer populations could then be squeezed out.

One way of framing this argument could be to consider if the agricultural adopters had different preferences for quality or quantity of children to those who remained hunter-gatherers. If the farmers had preferences that gave more weight to number of offspring than health or leisure time, it may be a rational lifestyle to adopt. The higher number of offspring then leads it to become the dominant lifestyle in the total population.