digital economy

Simulating complexity

There will be no dirt, no sweat and no temples in this blog post. But there will be loads of awesomeness because I (as well as a few other folks like me) do research at the very edge of observable science. It is actually so close to the edge you can see the vast expanses of ‘we really do not know yet’ emptiness that needs to be filled in with more projects, more funding and more people.

For those of you who have not heard of complexity science, let me give you a quick update. Until recently scientists thought that the better you know the elements of any given system, the better your understanding of how and why it works. This notion still holds but it gained a bit of a condition to it. You may know every stock trader on a first name basis and have intimate knowledge of their intentions (which, to be honest, are pretty simple: “buy low, sell high”), but is this going to help you predict the stock exchange crashes? No. It is not that difficult to investigate a neuron and learn all about the chemistry that happens in our brain, but will it help you explain human emotions? Unlikely. There are certain systems in the world (and a lot of them) of which behaviour is difficult to predict; systems in which a combination of individual actions of a large number of elements (stock brokers, neurons, cars, ants, you name it) produce unexpected global patterns; systems that undergo ‘critical transitions’ (like stock exchange crashes or ecosystem collapses) even though they were resilient to changes for a long time. In maths, they call them ‘non-linear systems’, in popular science the word ‘chaos’ is often invoked. I call it ‘freaking magic’ whenever I see a simulation of one of these systems unravelling in front of my eyes. It is astounding how often results of such simulations seem to put into question everything we believed in.

There is only one way to deal with such systems. You need to build them in your computer from the bottom up, starting from the smallest elements, code in the relationships between the elements and see what happens when they interact in their virtual worlds. And that is exactly what I do – I build archaeological simulations.

This means that I don’t get out of the office that much any more but I do get many opportunities to work on super-exciting projects with people of very diverse backgrounds and from around the world. Today I’m travelling to Cambridge for a meeting to discuss one of those. A group of people ranging from archaeologists to economists to physicist is developing a simulation testing how different subsistence strategies during the Late Pleistocene (ca 126,000 – 11,700 years ago) deal with abrupt climate change (rings a bell?). By investigating the resilience of different strategies in times of environmental stress we should be able to evaluate which cultural features are the most key to the survivability of a group. As the realisation that we need to tackle global challenges such as climate change, uneven resource distribution, overpopulation etc. using simulations, more and more modellers turn towards archaeology for its unique record, spanning millennia rather than decades, of how humans coped with similar issues in the past. We are no longer just pure science, useful to the general public for boosting tourism and providing topics for BBC documentaries – we are slowly moving towards applied science and, hopefully, this will mean that good times lay ahead in terms of funding, publicity and opportunities.

If you want to learn more about complexity science in archaeological context, read on some of the recent applications of simulation techniques or perhaps even learn how to code your first model (it is not as difficult as you think, promise!) check out our blog simulating complexity.