I must begin with a warning. This post is a bit long and boring but please indulge me.

In my previous post – Circles – I highlighted the problems involved in using technology to help the transition to a circular economy.

In my next post, I want to share a ‘systems map’. A graphic, showing my personal view of the interplay between the economic, commercial and technological issues that stand in the way of the circular economy. I’m not saying that it is finished and correct (far from it). I’m simply offering it as a straw man to be discussed, criticised, taken apart and re-built with the benefit of fresh eyes and expert opinion.

However, before I share it, I want to use this post to be sure that we have a common understanding of what this system map represents and how it has been arrived at. That’s the boring bit.

Wicked problems

Horst Rittel, the late Professor of The Science of Design at Berkley coined the term ‘wicked problem’. He used it to describe a problem that is near impossible to solve because the contradictory, complex, multi-faceted and interconnected nature of it means that there can never be a single solution. I’m sure he didn’t have the technical barriers to the circular economy in mind (this was 1973) but the characteristics he described of a wicked problem fit our challenges perfectly.

Wicked problems are:

Unique – every wicked problem is one-of-a-kind; there is no applicable or practical comparison.

The result of multiple root causes – there is always more than one explanation for what’s causing a wicked problem, and different actors might have conflicting perspectives on this.

Interconnected – wicked problems are not just characterised by interconnected root causes, they can be linked to other problems and even be perpetuated by each other. So, defining the boundary of a wicked problem is challenging in itself.

Unstoppable – it is impossible to know whether the problem is fully resolved and under control, and there is no visible point at which to stop iterating and searching for further/better solutions. Which means that…

Relative – they have no fixed measure of success. A wicked problem cannot have ‘true’ or ‘false’ solutions, only ‘better’ and ‘worse’.

Difficult to measure – there is no short term test to see if a solution is making things better or worse. There will be time delays and attribution issues. After all, if we cannot pin a single root cause, then we cannot pin a single source of improvement. Which means that…

Wicked problems have no definitive set of solutions that will fully resolve it. Which means that…

Solutions tend to require high-investment, without an opportunity to learn from trial-and-error (hence the ‘linear lock-in’ issue discussed in the previous post).

But none of that means that we shouldn’t strive for ‘better’ and there is a way to make wicked problems slightly less scary.

Systems Thinking

The process of systems thinking begins with drawing out a map of the problem’s elements in order to better understand the relationships between them. In geek speak, these elements are called ‘nodes’ and they are defined as “the outcome of an influencing dynamic”. 

In English, a node is something that can influence or be influenced by something else.

Nodes and system dynamics

Naming nodes is important. Ideally, nodes shouldn’t just be a noun. They should have a notional qualitative or quantitative value that can be increased or diminished when influenced.

‘Waste Streams’ isn’t a node. What is it about ‘Waste Streams’ that we want to understand? Their visibility? Their value? Their size?  Their quality? All can be influenced by other nodes. For example, ‘Quality of Waste Streams’ might be seen to be influenced by ‘Effective Waste Separation’ and to be an influence upon ‘Value of Recovered Materials’.

Connections

Once we’ve listed all the nodes, the task is then to understand these influences – ‘system dynamics’ – that connect them. A system dynamic is any form of causal relationship that exists within a system. So, if more X causes Y, it’s a system dynamic. If more A means less B, that’s a system dynamic too, as would be, if more C means more D.

In order to keep a clear understanding of the dynamics, it’s important to maintain the ‘if more’ perspective. We should look at each node and ask the question ‘what would having more of this directly cause?’.

Each node does not have a direct causal relationship to every other node (if it does, then it’s not a wicked problem to solve is it?) but one node may have a direct influence on more than one other.

A node can have a direct effect on another in two ways – positive or negative – and the effect can be immediate, or delayed.

All connections are either positive or negative but there’s no value judgement implied by the terms. Positive doesn’t mean ‘good’ and negative doesn’t mean ‘bad’ – it just indicates the direction of the influence. Delayed just marks a connection that is expected take time to become significant rather than having perceptible short term results.

So, in a system map, nodes, connections and their dynamic are shown like this (click to expand)

• 

• 

• 

Loops

When positive influences, negative influences, and delays combine, they can create loops. Two simple examples are ‘reinforcing’ and ‘stabilising’ or ‘balancing’ loops.

In a reinforcing loop, change in one direction is compounded by more change. A positive example might be:

In a balancing loop, change in one direction is countered with change in the opposite direction. Ideally the outcome is an equilibrium; a climate control system for example:

Because the goal in ‘+’ balancing loop is often implicit, it can be difficult to realise that a ‘-‘ balancing process is at work. It’s important to find the implicit goals in the system from a particular perspective – the ‘rationale’ that is driving actions, as this will determine the nature of the corrective action.

For example, think about a cup of hot coffee sitting on a table. The ‘goal’ of the coffee is to cool to its desired level – room temperature. Left alone, the natural corrective action, is a heat transfer between the coffee and the surrounding air. Coffee cooling is thus a natural balancing process that will occur without any explicit corrective action. 

However, the person that enjoys hot coffee has a different goal. Their desired level is not coffee at room temperature and so their corrective action is to actively prevent heat transfer between their coffee and the surrounding air (or, they could drink it before it cools). So, understanding the nature of the corrective action is important as it determines whether it requires an explicit action.

So, there we have it. The wicked problem of the circular economy and how systems thinking can help us to, not solve it, but progress. In my next post, I’ll share the 41 nodes that make up our problem and their map.