The business of futures is to see what possibilities lay ahead to better anticipate how to meet them when or if they become reality. When this story line follows a linear path this is a lot easier; when it follows a more complex path reality can bite.
Foresight models look at trends and curves in trajectories of things including those that might disrupt the status quo. Using tools and frameworks (PDF), foresight professionals and futurists seek to better understand the contributors (drivers) and patterns associated with decisions, activities, and circumstances to anticipate what might come and better prepare for it (strategic foresight). Foresight is being used in fields ranging from natural resource management to energy policy to healthcare planning.
A rational look at foresight finds many reasons to embrace it for an organization. Who wouldn’t want to have a better sense of what is coming and prepare for it? The problem foresight poses is that it can lead people to look for the right things in the wrong way and that has everything to do with our human tendencies to see narrative arcs in the stories we tell ourselves instead of seeing either exponential or j-shaped curves.
Both of these models for data have enormous consequences for how we understand some of our greatest challenges as humans and as organizations as we shall see.
A linear distribution or data structure is what humans see most easily. It’s the maintenance of a status quo, gradual change, or the progressive rise and fall of something over time. It’s what we see when we see in most trends and patterns. This perspective has the tendency to view much of the system in which this change takes place as relatively stable.
Stability is largely a matter of perspective. Everything is in motion to some degree; its the rate of change that we notice. In linear systems, that rate of change is relatively consistent or at a pace we understand while exponential change (or growth curves) are more challenging to see — and potentially more dangerous as the video below illustrates.
Al Bartlett’s lecture and other notes provide just one example exponential growth and how our perception is challenged by these kind of data structures in the world and the systemic effects they can bring.
Without an understanding of the growth dynamics associated with a particular phenomenon, we are at risk of grossing under-estimating the potential implications of what might happen. In these cases we need fixes, but not just any fixes as we shall see.
Another type of curve that can distort foresight models is the ‘j-shaped curve’. This curve describes situations where there is a long-term trend that is briefly countered in the short-term. An example is the case of alcohol consumption and health. There is evidence that alcohol consumption (e.g., a glass of wine or beer) can have a beneficial effect on a person’s health (at the population level, individual results might vary significantly). However, beyond that certain amount — that varies by person — and alcohol becomes toxic and can substantially contribute to a variety of health problems, injuries, and premature death. The j-shaped curve forms from data showing a mild reduction in health risks associated with modest alcohol intake as illustrated below.
For alcohol use, a single drink can lower your mortality risk before the risk starts rising again. Contrast this against cigarette use where a linear pattern of risk is seen: the more you smoke at any level, the higher your risk. Both patterns have linearity to them, but one is far more deceptive in it’s short and long-term implications.
Where this can fool foresight researchers is that there may be a trend that is showing a certain set of properties assumed to be on the trajectory like that on the left hand side of the graph when it is really similar to the right. Depending on the time horizon you use to inform your decisions based on this data the implications could be markedly different and potentially catastrophic.
Our fixes or strategies to anticipate change based on the wrong model could actually serve to amplify the very problem we sought to solve. A possible example of this is the move to ban single-use plastic bags. While the evidence of the environmental impact of plastic is considerable, a shift from plastic bags has its own negative implications, including the increased manufacturing of (with resulting waste and potential increased consumerism from) reusable tote bags or the increased use of forest products to support paper bag production.
The loss of plastic shopping bags which are often re-used (despite being called single-use) as garbage liners is now resulting in more purchases of plastic-intensive garbage bags. If the systemic implications are not considered in the design of such policies, these well-meaning fixes can profoundly fail. What is needed is a change in the way we consume, store, and buy goods, not just carry them home.
Systems change changes systems
The idea that you could be surrounded by literally thousands of people, connected to most of the planet through a device that fits in the palm of your hand, and still experience profound loneliness would once be considered the most profound oxymoron to anyone born before 1980.
Yet, here we are in a state where the very fixes for connection are failing us. The benefits of social media, social connection, artificial intelligence, and new production methods (e.g. 3D printing) are now starting to show some negative effects on our social and economic systems. Are these linear progressions of technological advancement that are simply generating a few of the inevitable bumps along the way? Are they exponential trends about to explode and profoundly transform the way we live? Or are these j-shaped curved trends that once provided us the benefits of finding connection in the modern world only to entrench our social systems into being online, not off?
We are creating systems that are changing themselves and having profound effects on the fundamentals around us. Retail conveniences created by online shopping means changing the relationship we have with our local merchants and that changes their viability. Handheld computers like an iPhone are engineered to hold our attention; what happens when we stop paying attention to the world around us?
These are systems questions and ones that foresight — when applied well — holds some promise in allowing us to anticipate and maybe deal with before its too late.
We can’t see these things coming if we hold models of the future that are based on a linear framing of what is happening now and what is to come. We also can’t adapt if we assume that even non-linear change will take place and persist within the same system it started in. Systems change changes systems.
Data models are fundamental to foresight and understanding them is the key to knowing whether your ahead of the curve, behind the curve, or sitting in the middle of the letter J.