Thinking in Scales

Systems exist in the full realm of our reality. They grow in time, occupy space, and connect to other systems on a variety of levels. That is why we analyze systems across the Time, Space, and Context dimensions. Because a full overview is important to prevent new externalizations, we need to evaluate our actions in all three. Let us look at how scales operate within each dimension, and at the three levels of impact that describe different orders of magnitude.

Spatial Scales

When mapping a system, it helps to include a variety of spatial scales. When looking at the redevelopment of a building, it makes sense to map on the scale of the direct neighborhood to figure out stakeholders, potential sources for recyclable materials, and so on. After this, mapping the city district, the state, and perhaps its influence on the world may provide remarkable insights. In practice, 3 to 5 scale steps are sufficient for the spatial dimension.

Temporal Scales

In time, it makes sense to look at the past as well as the future. Looking back for any aspect of a system shows us how the system has behaved before, which we can learn from in trying to look forward. History is the best teacher we have. It is easy to look back, much easier than looking forward. Make use of this wherever you can. The typical division is hours, days, weeks, years, decades, and about 50 years before and after.

Contextual Scales

Scales in context can give insights that space and time are unable to provide. Context maps plot relationships within a system. Context map scales relate to which degree you trace the relationships. For example, when you map yourself in your family context and draw your family tree, the scales refer to how many steps you go back: parents are 1st degree, grandparents are 2nd degree, and so on.

In the case of larger systems, scales refer more to a level of magnitude. For a multinational company, the first level may be the company itself, the second level the company and all its primary suppliers and customers, and the third level all people affected by the organization.

The Global Reporting Initiative (GRI) uses a standard for supply chain context scales. Tier one covers all departments of the organization itself. Tier two is tier one plus all primary suppliers. Tier three covers tier one and two plus all remaining suppliers in the chain. The complexity of information in this tiering is exponential.

Three Levels of Impact

Beyond the scales within each dimension, there are three levels of impact that describe different orders of magnitude:

1st level: Direct impacts. These are effects that flow directly from physical properties, operations, and direct resource use. They can be measured objectively in quantitative data: kWh, temperature, GDP. A large set of these indicators, such as those used in Life Cycle Assessment, can give a reasonable overview of direct impacts. While relatively easy to quantify, they tell us little about the influence an object or action has on the actual system, such as societal resilience and social justice.

2nd level: Network impacts. The second level sits between object and system, and relates to the effects of and on the network. By building a school far from a residential community, you increase transport movements. That is a network effect. This network effect in turn causes object level effects: the energy used, emissions, and time spent traveling. Other examples include increased connectivity of people due to social networking, or availability of food in an area due to the distribution of agricultural industries.

3rd level: System impacts. At the highest abstraction level, system behavior can be perceived. Here live the systemic behaviors that govern our past, present, and future: long-term societal resilience, educational value, adaptive capacity. While harder to express and usually difficult to quantify, third level impacts are by far the largest and therefore essential to focus on in our goal setting. System effects always have a large influence on the entire set of network relationships and all objects within.

What to Map?

We want to avoid the pitfalls of band-aid sustainability solutions that end up making matters worse. An effective approach is to investigate on what level the effects of planned interventions will play out. When the effects are only in the first degree, there is a good chance they will not have a significant beneficial effect in the long term.

As a general rule of thumb, you want to have at least one strong positive network effect for any intervention. Preferably, you create a strong positive systemic effect, but that is not always achievable.