There is an interesting paper called How Complex Systems Fail. It’s a collection of 18 related observations about complex systems and about when and how they fail. The observations are as follows:
- Complex systems are intrinsically hazardous systems.
- Complex systems are heavily and successfully defended against failure.
- Catastrophe requires multiple failures – single point failures are not enough.
- Complex systems contain changing mixtures of failures latent within them.
- Complex systems run in degraded mode.
- Catastrophe is always just around the corner.
- Post - accident attribution accident to a ‘root cause’ is fundamentally wrong.
- Hindsight biases post - accident assessments of human performance.
- Human operators have dual roles: as producers & as defenders against failure.
- All practitioner actions are gambles.
- Actions at the sharp end resolve all ambiguity.
- Human practitioners are the adaptable element of complex systems.
- Human expertise in complex systems is constantly changing.
- Change introduces new forms of failure.
- Views of ‘cause’ limit the effectiveness of defenses against future events.
- Safety is a characteristic of systems and not of their components.
- People continuously create safety.
- Failure free operations require experience with failure.
It’s worth reading the paper in it’s entirety but the gist of the paper is that attempts to analyze failures in complex systems are often themselves doomed to failure because of false assumptions about the nature of these systems. The main point is that because complex systems are inherently redundant, attempts to reduce collapse to a single point of failure are incorrect. In fact multiple failures have to occur simultaneously for the system to fail completely and furthermore, complex systems tend to contain multiple failures at any one time anyway. It’s the confluence of just the right combination of failures that causes the entire system to collapse.
Ultimately, a holistic approach must be taken to the health of the system.
Interestingly, complex systems in the sense meant by this paper tend to be systems with human components. Hospitals, stock markets, firms, and metros would be examples of complex systems. It’s not mentioned in the paper, but I suspect that if these human institutions constitute complex systems, wouldn’t bigger human institutions such as governments and entire civilizations constitute the same? If so, and I think this is the case, then this analysis would apply to these bigger things as well. Attempting to figure the cause of the collapse of civilizations such as the Roman Empire or the Maya would run into the same roadblocks as analyisis of those other sorts of systems. You would not be able to pinpoint the cause of collapse to any one factor or reason.
Perhaps the most famous history book of all time is Gibbon’s The Decline and Fall of the Roman Empire. This is a long and comprehensive account of Rome from its supposed peak in the second century to the final collapse of the Byzantine Empire. One of the most interesting things about this book is that Gibbon attempts to forward a thesis about why the empire ultimately collapsed. He believes that the decline of civic and martial virtue over time, hastened by the rise of Christianity, lead to Rome’s eventual inability to defend itself from the Barbarian invaders, which is a quite straightforward explanation.
However if the complex systems theory of civilizations is true, it would suggest that Gibbon’s explanation for the fall of Rome is far too simple and that many other factors likely contributed. And in fact, it looks like that might be the case. Many historians have taken Gibbon’s account to task and his explanation is no longer very popular, but the number of explanations proffered in its place are numerous. In fact, there are at least 210 given reasons for the Fall of Rome, and this kind of figures. Why should there be a simple, straightforward explanation? Rome as a civilization transformed itself almost entirely multiple times of the course of millennia, yet continued to run up until the very end. When one strength would disappear it would compensate with another. It would take a long series of disasters to finally destroy the empire.
Likewise, when people talk of lost civilizations such as the Maya or the Indus Valley Civilization, and wonder what happened to the them, perhaps there isn’t a simple satisfying answer such as ‘war’ or ‘famine’ or ‘a decline in civic virtue’. Perhaps the real answers is the much less satisfying ‘many things ultimately came together over hundreds of years to force the collapse of the civilization’ and perhaps the attempt to draw a simple lesson from these collapsed civilizations is an exercise in vanity? When a confluence of many factors leads to such a dramatic collapse it’s easy to pick out the ones we want and to use them to construct an easy moral which conforms to our preconceived notions. And so maybe the thing we learn about when we study civilizational collapses is really ourselves.