One of the things that popped up in the chat while I was talking immediately caught my attention, given my interest in long waves and futures. It was an article by Luke Kemp and colleagues at the BBC Future website that discussed the timespans of states, and their patterns of decline. 200 years is about the average, which tallies with the 100-300 year window used by Peter Turchin and his colleagues working on ‘secular dynamics.’

Perhaps this isn’t surprising since Turchin draws on the Seshat model, which Kemp and his colleagues use as one of their sources.

Signs of decline

Thinking about civilisations as being time-limited immediately leads you to a particular perspective about them, which is in effect, that as civilisations age they are more likely to show signs of decline, in the same way that we can make the same kinds of assumptions about people.

And from that, of course, we can start to build a view of what those signs of decline might be. But I’m getting ahead of myself here. Let me go back and discuss their approach to research and analysis first.

Luke Kemp is a research associate with the Notre Dame Institute for Advanced Study, and a research affiliate with Cambridge’s Centre for the Study of Existential Risk.

Pre-modern societies

The research here is into pre-modern societies, since if you are measuring how long they have lasted for you need to know that they have come to an end. They restricted their analysis to”states”, which they defined as organisations that enforced rules over a given territory and population.

To collect data, they created their own database, called MOROS, which “contains 324 states over 3,000 years (from 2000BC to AD1800).” The data was assembled from other databases, an encyclopedia on empires, and so on. MOROS, by the way, is the Greek God of Doom. Academics are such wags.

As mentioned above, they also consulted Seshat, which has been curated over time by archaeologists and contains data on 291 polities. (Seshat? The ancient Egyptian goddess of wisdom, writing, and knowledge.)

‘Survival analysis’

The two datasets, in other words, produce similar results. To do their research, they used a technique called “survival analysis”, which analyses the spread of lifespans:

If there is no ageing effect, then we can expect an “ageless” distribution in which the likelihood of a state terminating is the same at year one and 100. One previous study of 42 empires found exactly this. In our larger dataset, however, we found a different pattern. Across both databases, the risk of termination rose over the first two centuries and then plateaued at a high level thereafter.

This, apparently, is a similar finding to another recent study that found—in an analysis of historic crisis events—that the average polity had a life of 201 years.

In the academic article they summarise their findings like this:

In conclusion, the distribution of longevities reveals that the risk of termination rises over the first two centuries, and then remains roughly constant. That is, newly established states appear to have some benefit of youth in terms of survival chances, but this effect fades away over the first ~200 y[ears].

‘Critical slowing down’

In other words, as you approach 200 years, the risk of termination increases, but if you get past 200, the risk of termination flattens out. One of the effects of ageing is that societies recover more slowly from disturbances. This is known in the systems literature as “critical slowing down”, and is seen across many different types of system:

Before a complex system undergoes a large-scale shift in structure, or a “tipping point”, it often begins to recover more slowly from disturbances. The ageing human body is similar: injuries can take a longer toll when you’re older.

The point is that “critical slowing down” is an indicator of loss of resilience. The article discusses two examples of this from the archaeological literature—neolithic farmers in Europe and the Pueblan societies in what is now the south-western United States. Taking the second of these:

The Pueblo societies were maize farmers who erected the largest non-earth buildings in the US and Canada before the metal-framed skyscrapers of Chicago in the 1800s. The Puebloans also underwent several cycles of growth and contraction, ending with crisis events in around AD700, 890, 1145, and 1285. During each of these events, population, maize, and urbanism fell, while violence rose. On average, these cycles took two centuries, in line with the wider pattern we found.

State termination

In each cycle, recovery from crisis was slower than in the previous crisis.

Of course, there are some caveats here. One is the definition of a “state termination”. These take multiple forms, from a new ruling elite to a full-scale collapse, in which there is loss of government, written records, shared institutions, and population decline.

In addition, state termination may not be a bad thing. In some cases, state collapse led to greater prosperity for the majority:

Many pre-modern states were grossly unequal and predatory. By one calculation, the late western Roman Empire was three-quarters of the way towards the maximum level of wealth inequality that is theoretically possible (with one individual holding all the surplus wealth).

(By the same score, I’d note that the elite coup d’etat in England—later repackaged as the “Glorious Revolution”— that engineered the replacement of James II by William and Mary definitely led to an upturn for the English state.)

Longevity factors

The numbers are also based on generally accepted start and end dates, which may not be completely reliable, given their distance in history. Even more recent state terminations can be problematic. Constantinople fell in 1453 to the Turks, but it was also partitioned in 1204 by marauding Crusaders. (I’d imagine they might both be terminations, although the 1453 one was rather more final.) The researchers deal with this by providing a range of estimates.

Their next steps in their research will be to examine the factors that create longevity.

States could be losing resilience over time due to variety of factors. Growing inequality, extractive institutions, and conflict between elites could heighten social friction over time. Environmental degradation could undermine the ecosystems that polities depend on. Perhaps the risk of disease and conflict rises as urban areas become more densely packed? Or loss of resilience may be due to a combination of different causes.

Cascading effects

Depressingly, both extraction and environmental degradation are seen in the collapse literature as causes of collapse, along with over-complexity.

Of course, modern states are not the same as pre-modern states. They are more complex and more connected. But that may not reduce their vulnerability; it may just mean that they are at greater risk of collapse or termination from cascading systemic effects. We barely need to rehearse the litany of crises, from climate change to global wealth inequalities to failing planetary boundaries to the plethora of powers with nuclear weapons:

[T]he instability of a superpower, such as the US, could trigger a domino effect across borders. Both Covid-19 and the 2007-2008 global financial crisis have shown how interconnectivity can amplify shocks during times of crisis. We see this in many other complex systems.

(Source: The Long Now Foundation)

The Governance layer

But one of the things this research allows us to do is to attach a number to one of the layers in Stewart Brand’s pace layers model, which when discussed is largely conceptual. As a reminder, the top three layers are the ‘fast’ layers, the bottom three ‘slow’ layers.

Fast learns, slow remembers.  Fast proposes, slow disposes.  Fast is discontinuous, slow is continuous… Fast gets all our attention, slow has all the power.

In other words, the top three are sources of innovation, the bottom three sources of stability. Governance is the fourth layer down. We can assume now that the Governance layer has a clock speed of around 200 years.

The Collapse presentation can be downloaded here.

—

A version of this article is also published on my Just Two Things Newsletter.

The gap between the way these services are delivered and the technologies we use on a daily basis is wider than it’s ever been. Though it’s likely to be worse in Germany than in many other places, there’s actually such a gap everywhere. Technologies continue to grow exponentially whereas our public services, traditional organisations, legal categories and social norms change much more slowly, in a linear fashion. (Her emphasis).

Obviously this is the thesis of the Azhar’s book, and in the newsletter we get the usual ‘exponential’ tropes: the ‘wheat and chessboard’ story, wheeled out to demonstrate that our brains aren’t equipped to think about exponential growth. (While checking this just now, I discovered that this story dates from 1256).

So maybe it’s worth interrogating the ‘exponential’ story and its limits. Here’s some quick notes.

First, this is a Silicon Valley story, driven by the fact that over quite a long period of time Moore’s ‘law’ anticipated that the number of transistors on a microchip doubled every year. And so it did—it was even baked into the industry roadmap. So people like the Singularity Hub imagined what would happen until At least, it did until it stopped doing it.

And for chips, there are some strong negative exponentials at work: for example, the cost of the fabrication plant for each new generation of chips doubles each time around—that gets pretty expensive pretty quickly.

Second, people imagine that—maybe because this is a ‘Silicon Valley is different’ story—that this exponential rule on technology improvement is specific to digital technologies. It’s not.

Research by the Santa Fe Institute found that Moore’s ‘Law’ was a version of ‘Wright’s Law’, developed in 1936 when microchips were the stuff of science fiction. The doubling effect is a function of economies of scale and scope (learning). There comes a point when you run out of scale.

Third, and building on this, most technologies follow an S-curve pattern, accelerating towards a mid-point inflection and then slowing down. This is more obvious when you remember that an S-curve is the derivative of a diffusion curve; the mid point—at least when talking about markets—represents the point when half the available market has been reached.

But the reasons for that inflection point aren’t just to do with market capacity. The other thing that happens when technology applications accelerate rapidly is that they start to generate externalities, and these act as a brake on continuing accelerating growth. I wrote something about this in a piece reflecting on Alvin Toffler’s Future Shock, one of the 20th century ur-texts on the idea that change was going too fast for us to cope with:

S-curves or logistics curves, one of the basic building blocks of much futures thinking, show an acceleration in the first part of the S-curve, and then a deceleration in the second part. Much thinking that writes about exponential change, as Toffler does (in Future Shock) without using the word, imagines that the rapid acceleration seen in the second quartile of the S-curve, when growth is at its fastest, is a new normal, rather than a system that is about to start meeting its limits.

And if we look at the S-curve of technology and Vitaud’s ‘linear’ change in other social categories, we find that, in fact, regulation catches up. This is the reason why the question of how to manage our monopolistic tech giants is suddenly a live discussion.

The final note is a reflection on Stewart Brand’s ‘pace layers’ model, published in his book ‘The Clock of the Long Now’. The pace layers model looks like this:

Brand explains the difference between the top three layers (the fast layers, where tech also lives) and the bottom, ‘slow’, layers, like this:

Fast learns, slow remembers. Fast proposes, slow disposes. Fast is discontinuous, slow is continuous. Fast and small instructs slow and big by accrued innovation and by occasional revolution. Slow and big controls small and fast by constraint and constancy. Fast gets all our attention, slow has all the power. All durable dynamic systems have this sort of structure. It is what makes them adaptable and robust.

In an era of climate change, this is maybe a little more complex. Nature might be regulating the fast layers by being fast—even enforcing the occasional revolution. And indeed, Paul Saffo makes a similar point in a 2017 podcast conversation (57 minutes) with Stewart Brand about the pace layers, referencing the San Francisco earthquakes.

All the same, re-reading his article on this recently before I did some training, I was struck by a section in his final paragraph:

The division of powers among the layers of civilization lets us relax about a few of our worries.  We don’t have to deplore technology and business changing rapidly while government controls, cultural mores, and “wisdom” change slowly.  That’s their job.  Also, we don’t have to fear destabilizing positive-feedback loops (such as the Singularity) crashing the whole system.  Such disruption can usually be isolated and absorbed.  The total effect of the pace layers is that they provide a many-leveled corrective, stabilizing feedback throughout the system.

So why does the idea of ‘exponential’ continue to have such a hold? I have a benign version and a less benign version. The benign version is that humans are susceptible to the idea that ‘this time is different’. There’s a reason why ‘The Emperor’s New Clothes’ is such a compelling story. The less benign version: ‘exponential’ is an effective sales tool for consultants and technologists alike.

A shorter version of this article was also published on my Just Two Things Newsletter.