Complexity: it’s not complicated.

Driving back from Hampshire the other day I caught an amazing discussion on Radio 4 between Melvin Bragg and a hatful of Boffs.  They were talking about The Science of Complexity and it was captivating.

fractalThere’s a lot of data around these days; big data, and we’re all adding to it every second.  But Complexity exists in places much more inaccessible than your hard drive.  Birds flocking, diseases spreading, crowds moving across a station concourse, stock market activities, why the Arab Spring happened.  Fortunately there are a lot of computers and boffs around these days too and they’re getting better at analysing it.  That’s Complexity Science.

There is a difference between something complex and something complicated.  If something is Complicated it can be made and controlled and more importantly, its behaviour can be predicted – a jet engine for example.  If something is Complex however it is largely a (seemingly) chaotic mass of randomness.

Too much complexity is unfathomable; what will society be like in 300 years? No idea.  But understanding complexity through mathematical modelling can produce policy that will, if observed, narrow the risk of disaster.

Most excitingly of all, the human brain, a raggle taggle of individual neurons is a super-complex thing.  When a complex system does things on a system scale which you could never have predicted from the components and the way they interact then you achieve what’s called “Emergence”.  The arrival of something more than the sum of the parts.  In the example of the human brain you get, for example, “consciousness”.  Wow.  The same applies when multiple brains interact; this leads to the emergence of collective social intelligence.   Man, those nuerons just don’t have a clue what they’re up to.

Complexity scientists are on the road towards getting a handle on why large collections of people and things behave the way they do in big numbers.  It’s really fascinating stuff; number-crunching on a massive scale.  It’s a really useful way for governments and policy makers and companies to deliver a better framework for giving us plebs what we really, really want and it might go some way towards explaining otherwise inexplicable stuff like why Carol Vordermann is on TV and who shot JR.


2 thoughts on “Complexity: it’s not complicated.

  1. While I agree with most of what is said on the program, I fundamentally disagree with the notion that complex systems are not predictable. If one can model the relationships between agents in the system, then one can simulate the system (using a computer for instance) and therefore predict its behaviour. Modeling requires an understanding of these relationships, and that may be complicated 🙂 Our analytic tools are not at a point where it is easy to unravel such relationships. As a result, and because we cannot model them, we may believe that the system is unpredictable, For instance, if we understood how people interact with one another, with news, with their economy, with religion, etc., we could – at some point – understand and simulate a complex system like the Arab Spring. Fifty years ago, we may have thought that weather was unpredictable. Yet today, we leverage thousands of probes, data points and complex fluid dynamics techniques to get increasingly better predictions.

  2. Its a fair point Bernard, and I think they did mention towards the end that with such modelling they were able to cluster outcomes according to liklihood and in doing so effectively enable better prediction. But thanks for contributing to the debate – I really enjoyed the very digestible way in which the subject was dissected.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s