GPM First
Chapter 6 of Complex Adaptive Leadership (978-1-4724-4791-3) by Nick Obolensky

Getting to Grips with Chaos and Complexity



This exercise/game has been around for some time. I wish I knew who invented it so I could give credit. It has changed many people’s attitudes. It is used by a variety of places to help develop leadership – for example, it has been used in the past on the first day of officer training for the Royal Marine Commandos to show there is more to leadership than meets the eye. The exercise is also a great way to show how chaos and complexity has an underlying order and simplicity which is a rather counter-intuitive notion. The first time one organises this exercise is terrifying because it seems impossible. But one soon realises that the impossible is very possible. You will need at least eight people and it can be played by over 80. The ideal number is around 25. I am not sure if you have access to 25 people willing to undertake what will seem an impossible task – but this is well worth the experience! It is a working experiment that shows how simple complexity can be, and it leads to a very counter-intuitive and intriguing conclusion about leadership within complex tasks.

Here is a description of how to run the exercise.

The first thing is to find an empty space – 50 people would need the area of a tennis court – 25 people would need at least half that. The space needs to be clearly delineated (using white tape can help here if needed, given the absence of clear boundaries like a fence/wall/verge and so on). This is what to tell the people, after a scene setter appropriate to your context (all verbal instructions to be given are shown in italics – if you follow them verbatim you should not go too far wrong).

‘Please note the boundaries of the exercise area.’ – Point out and describe the boundary. ‘I would like you to position yourselves at complete random within the area and be at least a good arms length from anyone.’

After they are in their starting position:

‘Now that you are positioned in the area, please pick at complete random two other people participating within the exercise area BUT you are not allowed to indicate who you have picked in ANY WAY. Do not pick me – I am not participating.’

‘These two people are now your reference point – you cannot change your reference points, nor can you indicate to them that they are such (even though, as the exercise goes on, you will have a strong desire to!).’

‘Let me explain a few more important and simple rules to help you before I give a specific objective:

Stay within the boundaries.

Use the space when you start and continue – you will want to gravitate towards each other but this will make things harder – so keep your distance from others and use the space.

Move slowly and make the minimum needed minor adjustments to your position. Cover the least possible ground – if you make big and/or fast movements across the area you will be unpopular and also be breaking the rules.

You can only stand still when your objective is achieved.’

‘Now let me explain your objective – it is very simple – what you need to do is to adjust your position slowly and gradually so that you are at an equal distance from each person you have chosen as your reference points – equal distance does NOT necessarily mean in-between.’

Here you can slowly move to show yourself at the top point of an isosceles triangle with two others as an example:

‘For example if this person here and that person there were my two reference points, by moving slowly and gradually and covering the least possible ground to here I would be at equal distance. Remember that equal distance does NOT necessarily mean in-between.’

Having shown what an example of equal distance looks like you can continue:

‘So you will need to adjust your position gradually staying within boundaries, using the space, keeping distance from others, moving slowly, making minimum needed adjustments, and only being still when you are at equal distance (in other words do not wait for others to stop before you re-adjust!!). And just to remind you, you are not allowed to change or indicate who your reference points are even though at some stage you will have a very strong desire to tell them what to do!’

‘Remember too that you can only stand still when you are at equal distance from your reference points –so if they move, you will need to move and adjust as well. Once all have achieved their objective the exercise will be over. How long do you think this will take?’

At this stage some people may well laugh and some may say it is impossible – after all if everyone is trying to adjust at the same time it will seem like they have been set an impossible and never ending on-going loop. It will seem like a recipe for chaos (and in fact it is!). But if the rules are followed they will achieve steady state in the very rough proportion of time of one to one and a half minutes per ten people (so 25 people should take around two and a half to four minutes). Do not tell them this! But you can say

‘Don’t worry, it will go faster and easier than you think – you may be surprised!’

Then say:

‘OK? Ready, steady, go!’

There will be a big urge on your part to step in and help/give directions when it seems to fall apart – do not do so! You will see that it will oscillate between coming together and seemingly falling apart – but let it flow. You will need to practise ‘wu wei’1 [187] – holding back and allowing the natural flow.

Once all are still, just check by asking:

‘Are you all now at equal distance from your reference points?’

Assuming they all say yes, tell them how long it took and congratulate them! After all as a group they self-organised and self-managed a highly complex task. Ask them:

‘Given the number of possible solutions, how complex was this task?’

Many will say it was highly complex, which indeed from a mathematical perspective it is. And then ask (which often gets a laugh):

‘What would have happened if … we had put one of you in charge?’

Why do many people laugh at this question? Because it is apparent that if someone had been ‘put in charge’ the exercise would probably go on for a very long time indeed.

The counter-intuitive and intriguing conclusion is that the more complex the situation and task, the less directive traditional leadership is needed.

To enable a good debrief (or as a reflection for you if you cannot actually do the exercise) say (ask yourself):

‘You have achieved in a very short period of time a task that some of you thought impossible, and most of you thought would take a long time to do. What enabled you to do such a thing? What was in place that enabled you to achieve this?’

Make a list – you will find that it will be quite similar to the list which appears in this chapter. The aim here is to get some underlying principles of how to get complexity to work for you (rather than against you) – you will find it is, in fact, quite simple!

The outcome of the exercise at the beginning of this chapter often surprises people. It seems impossible, yet it works out very quickly. The counter-intuitive (and for some disturbing) conclusion is: the more complex things are, the less traditional leadership one needs. Instead a leader needs to put in place certain key principles so that the organisation being led can become self-leading, and the leader can then practice what is recognised in Chinese philosophy as the highest form of leadership – wu wei.

The implications of this simple axiom are immense, and form the basis of Complex Adaptive Leadership. The exercise is complex in that the number of possible solutions, and the way to get those solutions, is huge. As an example, it would take over 100 years just to count the number of possible solutions with 25 people in the exercise. Most agree that if someone was put in charge, then the exercise would take a very long time if the leader practised a typical oligarchic approach (that is, took charge and took responsibility for coming up with the solution).

But what is needed in place of traditional leadership? The debrief question ‘What enabled you to complete a highly complex task?’ typically results in a list.1 [188] The list below is rank ordered by frequency:

  • clear individual objective;

  • a few simple rules;

  • continuous feedback;

  • discretion and freedom of action;

  • skill/will of participants.


In addition to the five principles above, the following three often came out in subsequent group discussion. They are not perhaps as obvious as the first five:

  • underlying purpose;

  • clear boundary;

  • a tolerance of the players for uncertainty and ambiguity.


Other principles sometimes emerge during the debrief of the exercise but the eight above are the most important ones. One which is not mentioned above, and is often discussed, is the issue of trust. However, if the principles above are applied with integrity then individuals can proceed independently of trust.2 [189]

Let’s look at each in a little detail. We explore how they can be applied and what they look like organisationally in more detail in Chapter 7. At this stage it is just intended to see how they can fit generally:

  • Clear individual objective. Each person has a very clear idea of what they are trying to achieve. Although they cannot say exactly where they will be, how far from their reference point they will end up or even how (that is, the route) they will get there, the objective they have is clear enough to get them moving. Note the objective is specific enough to give a clear indication if it is achieved or not, but not so specific as to tie the individual down (for example the angle and distance, as well as the route, is flexible). In other words, the individual sets the specific objective (by choosing the two specific people) within a broad objective. It would be very counter-productive to even try to specify the detail of the objective! How many times do we fall into the trap of trying to fathom out the objective to the ‘Nth degree’? Although this can work well for relatively simple tasks and projects, when things are complex and dynamic such detailed working out of the objective does not help at all. In fact it gets in the way, taking up time, and demotivating those involved. What is sufficient is that each individual knows what needs to be achieved and, as important, will know immediately when it has been achieved. As the situation changes, and if that achievement then is lost, the individual has enough clarity to adjust and adapt. Another important aspect is that each individual has a sense of ownership of their own objectives, as they could choose which two people were their reference points.

  • A few simple rules.3 [190] The rules are sufficient to enable effective action, and enough to keep the system from descending into too much chaos. However, they are not so many as to cause the system to slow down and become cumbersome. The balance between having enough rules to fulfil obligations, and allowing enough freedom to act, is a fine one. The Disney Corporation found that it had gone too far down the route of having rules and procedures for everything. When they tried to move that concept to Europe and a new Disney-world in France they found that it did not work. This experience then forced a look at how they used rules and procedures in general. This saw a major simplification and the results which flowed enabled the new resort to stay afloat, and the current resorts to improve their operations.4 [191] The move from a heavily measured organisation to a more complex adaptive one is mirrored in the evolution of Semco in Brazil. This is described in the book by their CEO Ricardo Semler called Maverick, mentioned in Chapter 3.5 [192]

  • Clear boundary. The boundary gives a definition where the action is. For the exercise, the boundary is both the area within which it takes place, as well as the boundary of how the game is to be played (that is, by looking at the two reference points and then moving following the rules to adjust one’s position). We typically define boundaries as ‘who-is-in-charge-of-who’, with the ‘organogram’ being an all-too-typical way of showing how the organisation is structured. Rather than focusing on the boundary of the organisation per se, Complex Adaptive Leadership looks at the nature of the relationships across the boundaries outside of the organisation as well as the internal networks. This is a different way of looking at things – typically we look at the objects within a network (the departments, units, individuals and so on). The other way of looking at it is by seeing the relationships as the main thing, with the objects as the background, as illustrated by Figure 6.1.


At a more macro level, the boundary conditions of an organisation also includes its strategy – what does it deliver, to who and how. Under the oligarchic assumption, strategy is something defined by the leaders and executed by the followers. Under a polyarchic assumption, strategy is defined by the dynamic of the relationships within which the organisation exists. At the end of the day, that dynamic needs to be clarified, so the techniques used in clarification may be the same as within an oligarchic assumption, although the process may differ (for example, the use of mass intervention techniques as described in Chapter 3). The establishment and maintenance of boundary conditions takes time. So whilst in the more traditional days of oligarchy a leader is looking down into the organisation and defining the strategy and resultant organisation, the polyarchic leader will be looking across and outside the organisation, realising that his organisation is itself a node in a complex network.

Figure 6.1 Network of relationships is key


  • Continuous feedback. Every individual knows at any time where they are in relationship to achieving their objective. And they are able to interpret the data they see – in other words they can judge distance. They ‘sense’ what the reference points are doing and ‘respond’ – a very real experience of being adaptive. This relates to Key Performance Indicators (KPIs). There are huge debates over the use of measurement, both what should be measured, how it should be done and how reported. Suffice it here to say that an organisation needs a range of measures, and individuals need both to understand them and to understand how they influence them. These should not only include ‘hard’ measures such as financial and operational metrics, but also ‘soft’ measures ranging from staff attitude (at an organisational level) to personal behavioural 360 degrees feedback (at an individual level).

  • Skill/will of participants. Each person has the ability to judge distance and move accordingly (skill) and also wants to do the exercise (will). Translated into organisational terms, this would mean each person has the skills needed to do their job as well as the motivation to do it well. Again the field of skills development and motivation theory is a huge area. People are generally more skilled and motivated than their leaders would often suppose (if their actions are anything to go by, as explored in Chapter 8). The best way a polyarchic leader can develop skills is to ensure a process of knowledge transfer is available, as well as necessary development courses. And the best way a polyarchic leader can develop motivation is to understand how leadership behaviour can demotivate – and then eradicate such behaviour! this issue is explored further in the application of the skill/will model in Chapter 9 and the complex Adaptive Leadership model described in Chapter 10.

  • Discretion and freedom of action. Each person is free to act without having to wait for ‘permission’, or needing guidance on which way to go. This takes both an organisational culture and a personal attitude which encompasses Stuart’s Law of Retroaction (it is easier to seek forgiveness than obtain permission). Having discretion and freedom of action within well-defined boundaries is critical for complexity to work. And people need to feel confident in taking risks and using initiative.

  • Underlying purpose. There is an implicit and unifying underlying purpose to the exercise. When i run this exercise it is normally as part of a course in a business school. I do not need to remind people they are in a business school wanting to learn new things – it is the underlying purpose that unites people and thus it is easy for me, the professor, to ask people to do a strange exercise. If we had all been strangers on a bus, then the underlying purpose would not support the exercise (that is, wanting to go somewhere, and not get off the bus to do a crazy exercise, no matter how compelling the learning).

  • Ambiguity and uncertainty. Within the exercise there is a degree of chaos, and the situation is far from equilibrium. Whilst for some, if not all, there may be an uncomfortable feeling that things are looking chaotic, and that the exercise may be impossible, people still enter into the flow. This ‘far from equilibrium’ and uncertainty is the very essence of life itself, is therefore natural and something to be embraced rather than avoided, even if we prefer order and control. And the paradox is that we should not abandon order and control either!

There is also a Yin/Yang element to the eight aspects above. The eight principles can be paired, with each pair having two principles which are complementary and in some way paradoxical to each other.

Figure 6.2 Yin/Yang and the Four + Four model for leading complexity


These four pairs are in some way paradoxes as they seem to contradict (that is, implicit–explicit, freedom–enclosure, people power–rules power, ambiguity uncertainty– unambiguous measurement). And surrounding the pairs is the uncertainty of deeper reality, as well as the fact that it is uncertain that these rules work all the time! It is not for nothing that the sub-title of this book is ‘Embracing paradox and uncertainty’.

Chapter Summary

  1. The more complex things are, the less traditional directional leadership one needs if certain principles are in place.

  2. These principles form four pairs of paradoxical principles which together allow complexity to work within an organisation.

Submit your own content for publication

Submit content