Novelty
"The future ain't what it used to be."
Description
To address a problem, we need to know how its system will behave in future. This is easiest if the system will be the same in the future as it has been in the past, because we can rely on historical information to work out how it would respond to our actions.
However, systems can change. The most obvious changes are changes to structure, such as adding or removing elements, relationships or delays. However, even if the system's structure doesn't change, the system's behaviour could become significantly different if the actions we take are different — especially if the system's behaviour is complex. Even small actions, if repeated, can have a cumulative effect that drastically changes system behaviour (this is called a tipping point).
Predicting if and when a system will undergo significant change is practically impossible. If it happens, adaptation is the only way to deal with it. (Mintzberg 2000, p. 231)
Of course, the problem system may be entirely new, rather than pre-existing.
It's very difficult, and often impossible, to determine the behaviour of a new or changed system by looking at known systems (Roberts 1978, Godet 1987). Since we can't rely on history, we need to rely more heavily on theoretical analysis of the novel system. Except for simple systems, this is difficult and error-prone. We may not even be able to deduce the structure of the new system, far less its behaviour.
Examples
Element: Reverse Gear
The inclusion of one additional element (an ‘idler gear’) in a gearbox changes the direction of travel from forward to reverse; i.e., it changes the gearbox’s behaviour to the opposite.
Relationship: Steering
Control systems, such as a driver steering a car, work because of a relationship that corrects the system’s behaviour when it does something we don't want.
If either of the relationships in this system fails (i.e., we don't turn the steering wheel or doing so has no effect), we won't be able to travel in a straight line for very long. Adding or removing relationships in a system can totally change system behaviour.
The signs (+ or –) of relationships are also critical. If our car's steering mechanism were assembled such that rotating the steering wheel turned the car in the opposite direction to what we expected, we could no longer control the car if we drove normally. Rather than counteracting the car's drifting, our steering would increase it.
Delay: Sloppy Steering
Normally, we can steer a car along a straight road by making subtle movements of the steering wheel, resulting in the car travelling in a straight line with only minor deviations.
What would happen if a loose connection in the car’s steering system suddenly resulted in a delay of 0.25 seconds from when the steering wheel is turned until the road wheels changed direction? If we base our assessment of the car’s new driving behaviour on its previous behaviour, we would probably assume that we could still drive in a more-or-less straight line, but with larger deviations.
In fact, the delay is likely to result in ever-increasing over-corrections leading to a total loss of control. The system’s behaviour changes from stable to unstable.
Global Warming
The good news for the climate system is that its structure isn't changing. Despite the prospect of global warming, the climate is still driven by the same elements and relationships as it always was.
However, the novelty we're now facing is the possible impact of human activity. This primarily takes the form of much larger greenhouse gas emissions than previously. In system terms, this is a significantly changed action applied to the system.
There is some evidence that the behaviour of the climate system is changing. If its behaviour were consistent with history, we would be entering an ice age. Temperatures would be dropping, but they actually seem to be increasing. Since the climate system has changed, we can't rely directly on historical records to be able to predict what will happen or how to influence it. Instead, we have to put more emphasis on theoretical models (e.g., simulations), with their attendant risk of error.
"Even if the models can reproduce climate history, we shouldn’t expect them to reliably tell us about the strange new future we’re facing."
Because we can't have great confidence in our ability to predict and manage the climate, there is potential for people to adopt beliefs ranging from denial to doom. When convincing evidence is lacking, beliefs can be unduly influenced by thinking traps such as instinctive self-interest.
Key Point
These examples show that small changes to a system can drastically alter its behaviour. This means that we can't rely on knowledge of its past behaviour to work out how it will respond in future. We need to gather more experience with the new system and/or rely on theory.
Related Issues
This section describes similarities and differences between this issue and related problem factors and thinking traps.
Complex Behaviour
Systems with complex behaviour are more likely to exhibit significantly different behaviour when they change. Such systems are also more difficult to analyse theoretically, meaning that the absence of real-world information about their new behaviour is more problematic.
Probability & Uncertainty
Our lack of knowledge about how an unfamiliar system will behave means that we need to think in terms of probability and/or uncertainty.
Instinct
Our instincts are based on the behaviours of systems that our ancestors experienced. If we are confronted with a novel or changed system, our instinct about how it will behave may be wrong. Since instinct tends to be overconfident, we might assume that we know when we actually don't.
Intuition
Intuition works by recognising patterns. It assumes that things that seem similar behave similarly. However, only a small change to a system's structure, or the actions applied to it, can totally change its behaviour. Intuition can underestimate the significance of such changes and assume that the system's behaviour will remain largely unchanged. As with instinct, intuition tends to be overconfident, so we might assume that we know when we actually don't.
Unconfidence
Given the difficulty of solving problems that involve novel systems, we should be wary about pushing known systems (such as the environment) into uncharted territory, and we should be wary about the actions we could take to fix problems. However, caution does not necessarily mean that we should do nothing.
Risk Factor Misjudgement
When we can't rely on historical information to assess the probability and consequences of events, we need to rely more on judgement. Unfortunately, we are vulnerable to many types of misjudgement.
Related Engine Processes
Later, we will describe a thinking framework that comprises multiple processes. This section points forward to the processes that deal with this page's topic.
Project Management
If actions to fix a novel problem don't need to be undertaken immediately, we can defer taking action for a while so we can learn about how system behaves. If the system is no longer novel due to our newfound understanding, our actions are more likely to be successful. Of course, we need to be careful not to defer action until it's too late.
Adaptation
Adaptation is especially relevant because:
- predicting if and when a system will undergo significant change is practically impossible, and
- the actions we initially take may not be ideal because we don’t have enough understanding of the system to know how it will react.
Further Reading
Chivers, p. 103.
Mintzberg 2000.
Roberts 1978.
Godet 1987.
To Do
Murdered Darlings
Rename 'unfamiliarity' or similar??
David Hume's realisation that science assumes and somewhat depends on the assumption that the future will be consistent with the past, even though there's no proof of this (but only historical data). [Chivers p.103]
Examples:
- Recipe missing one minor ingredient (eg, yeast). Not a 'problem'. Or, 'one bad egg spoils a bunch' [sic].
- Load-bearing components suddenly subject to greater loads. But this isn't structural, but changed actions (inputs). This is more like a tipping point, akin to GW caused by increased CO2 action.
- unchanging system converted to exponential growth by addition of feedback loop; such as population size and birth rate.
- car steering reinforcing CLD is exponential, same as population growth.
- GW included: (although land use changes and other actions are also significant)