Skip main navigation
We use cookies to give you a better experience, if that’s ok you can close this message and carry on browsing. For more info read our cookies policy.
We use cookies to give you a better experience. Carry on browsing if you're happy with this, or read our cookies policy for more information.
A pendulum

Initial conditions and extreme events

Sensitivity to Initial Conditions

The pendulum mentioned in Step 2.2 is not sensitive to initial conditions. If there is a small error in the length of 994 mm, say one mm, it will still swing back and forth in very close to two seconds. Similarly, if it is released a bit more to the left or a bit more to the right, the difference will be negligible.

By comparison, social systems can be highly sensitive to initial conditions. For example, say you need an expert’s advice to write an urgent report. The expert leaves their office at 12:00, and you call at 12:01. The timing of your phone call will affect the quality of your report.

Most social interactions are not replicable. Generally you can’t try a policy, reset the system, and try another policy. Even if you could, if the system were sensitive to initial conditions it would evolve differently because you could never make it exactly the same as it was. In general, point-predictions cannot be made for social systems.

There are different kinds of prediction, e.g. it is impossible to say when and where lightening will strike, but it is possible to predict that there will be lightening and put in place policies to manage it such as installing lightening conductors. You can’t know in advance that lightening will strike a particular building, but this policy mitigates the effect if it does.

Similarly it is impossible to predict if any particular house will catch fire, but empirically it is certain that some will. Civic policies to manage fires include maintaining a fire brigade. Although unusual, house fires are sufficiently common that insurance companies can predict how many there will be of what type, and set premiums accordingly. Of course those probabilistic predictions are estimates based on past experience, and if some kinds of fire become more common the premiums will go up next year.

Extreme Events

Some events are so individual, rare, extreme, or unprecedented that it is not possible to make predictions based on their previous distributions. For example, no country has ever left the European Union but following the possible Grexit, which has not happened until now, will there be a Brexit? And if Britain adopts a policy of leaving the EU, what will be the consequences?

Extreme events can disrupt patterns that are so established that they are taken for empirical laws. For example, holiday companies project future holiday demand based on previous demand, but a terrorist event can change this. More commonly, extreme events such as flooding can disrupt predicted social and economic activity.

What do you think?

Can you think of examples in your life of unpredictability due to sensitivity to initial conditions? Do you have experience of extreme events? Share you experiences in the comments below.

Share this article:

This article is from the free online course:

Global Systems Science and Policy: an Introduction

UNESCO UNITWIN Complex Systems Digital Campus

Get a taste of this course

Find out what this course is like by previewing some of the course steps before you join:

  • Policy Makers including presidents, directors of NGOs, and citizens
    Policy makers
    article

    In Global Systems Science this article defines policy makers are defined to be politicians, their officers, citizens and other stakeholders.

  • Prediction and the policy dilemma
    Prediction and the policy dilemma
    video

    The Policy Dilemma involves policy makers trying to predict if their policies will work. This article explains why prediction is so hard.

  • Policy design
    Policy design
    video

    In this video Jeffrey Johnson explains that policy, like design, is a coevolution between problem and solution involving compromise and satisficing.

  • Conclusion to the course
    Conclusion to the course
    article

    This article concludes the Global System Science and Policy course by noting that Complex Systems Science is young and cannot solve all problems.

Contact FutureLearn for Support