This Time Is Different (Maybe)

Peter Turchin


Join 36.7K other subscribers

In my research I studied many historical societies that experienced structural-demographic crises. In each case there was a multitude of causes for the explosion of political violence, but I see two commonly recurring themes, fundamental problems that prevented people from finding a nonviolent way out. First, and most importantly, they didn’t really understand the nature of the crisis they found themselves in. Second, and related, their lack of understanding of deep structural causes driving their societies to collapse caused them to blame specific individuals and groups as causes of the crisis. Instead of cooperating to solve the problem, these societies descended into partisan conflict, which eventually exploded into a full-blown conflagration.

Sounds familiar? In the United States today most everybody knows who the culprit is. Depending on who is speaking, it’s Trump, or Clinton, or Black Lives Matter, or Blue Lives Matter. As passions and hatred of the opponents boil over, barriers to violence go down, and we get perilously close to the point where the cycle of strike and counter-strike, revenge and counter-revenge takes over in an autocatalytic manner.



So what’s to be done, to repeat the question from the previous post? The only way to get out of the crisis is to generate broadly based collective action for positive change. It cannot come from within either of the parties, which are by the nature divisive. You cannot overcome a disintegrative structural-demographic trend by playing party politics.

Even more important is to have good working understanding of how we got into this mess. Without such understanding, we will not be able to escape it. The only hope I see for us, and why this time may be different, is that we have a much better understanding of the causes of this crisis.

Consider also that the current crisis has been decades in developing. For decades we have been doing some things (and perhaps many things) wrong. Drastic change means that almost everybody will be negatively affected, at least, in the short run. We need a good working model of the society to figure out which of the unpopular measures are really needed. There will be enough pain to go around, and we don’t want to impose unnecessary suffering on anybody.

This means that the fundamental problem number one, our lack of clear understanding of how we got into this mess, is really the most important one that needs to be solved.



My friend and colleague Kevin Feeney has a particularly clear idea of how collective action for positive change should be structured. First, we need to develop much better science, because without good science we are doomed. Although the structural-demographic theory provides us with a decent set of theoretical tools to understand the deep causes of the current crisis, I would be the first to admit that it’s just a beginning. Different social sciences, e.g. economics, have learned a lot about different aspects of how our societies function,. These insights need to be integrated within a more holistic understanding of how different factors—economic, social, demographic, cultural etc.—work together, both to drive us to the precipice, and hopefully get us away from it. This means establishing a transdisciplinary scientific center or institute is a key part of the solution.

We also need advocacy organizations that would get the message coming from science out to the public, put pressure on the elected representatives, and propose concrete policies.

The third element of the solution is an organization that translates science to policy. We need a mechanism that will evaluate proposals on policies and reforms. Will they move us in the right direction? What are their unintended consequences? Human societies are nonlinear dynamical systems, and sometimes a policy intended to achieve a certain result can lead to precisely the opposite effect—due to those nonlinear feedback loops.

Why does the science part and the policy part need to be separate, and independent organizations? Because, again as Kevin points out, they have different goals, methods, and values that don’t mix well. The science institute is about truth, and must be able to speak truth even when it is inconvenient. The policy organization is about good, about how to increase social wellbeing. There is a tension between what’s true and what’s good. We really need two separate organizations to pursue each.

So there you go. We need a science institute, a science-to-policy think tank, and advocacy organizations—separate, but working together. Then we somehow need to forge a common consensus and adopt painful reforms. And in the process we have to convince, or overcome special interests whose narrow-based wellbeing is vested in the status quo. Who said it’s going to  be easy?

Notify of
Most Voted
Newest Oldest
Inline Feedbacks
View all comments
Luke Lea

You might like this answer to the problem, especially chapters 3 and 4:

Rich Howard

Only if it’s biased.


“So what’s to be done,…”

Perhaps the correct answer is to have the crisis occur early rather than later. I would think that social pressures would merely build up if you put off an inevitable breakdown and rebuilding.

This strike me as similar to activist monetary policy. It’s too early to tell whether the long term experiment in metastasized Keynesian thinking will pan out.

An odd thing to me about corrective social policy, if that is what is being suggested, is that it seems to avoid building up the glue that cements a culture or nation. Lacking a strong core, something like a shared religion, Roman virtues, ethnicity (with a scary ‘other’ nearby), or all those little things like a shared base of cultural thinking, social policy is just whistling past the graveyard.


I’m not so sure that truth and good are so different. There is an interesting take on that question in the chapter 8 of Landemore’s Democratic Reason, that would greatly interest you (and by the way, the whole book is fascinating).

Ross Hartshorn

I think that an advocacy organization that is based on “structural demographic theory tells us that X, so we should…” is going to be a pretty hard sell, by which I mean it will not work. However, perhaps SDT could tell us something about how to make something that does work.

Also, let’s face it, we do not have tens of millions of SDT enthusiasts to put behind this. So it will have to be something other than mass mobilization to influence elected officials.

However, I did find it interesting in your recent “Ages of Discord” that much of the work involved trying to find proxies for the various variables of interest, such as the Political Stability Index and so on. I was left wondering, could there be a real-time index of PSI?

The various economic metrics (DJIA, unemployment rate, inflation rate, etc.) get a lot of attention, simply because they put a numerical, and approximately real-time, measure of a more nebulous entity (“the economy”) that is hard to pin down. We actually care more about the overall economy than any one metric, but the metric gets the attention.

I wonder if devices like the Dictator Game or the Ultimatum Game, from behavioral economics, could be scaled up to measure something like PSI. Or, using a variety of regularly updated, public data (e.g. certain kinds of crime statistics, voter turnout, or whatever else can be found) to generate a publicly viewable PSI which is tracked over time. Perhaps also, tracked regionally (the PSI in Texas seems to be going up/down but the California PSI seems to be going down/up, so what is going wrong/right in Texas different than in California?).

Making a publicly viewable, graphed over time, measure of PSI which is tracked at the national, regional, and perhaps eventually local levels, may not seem as exciting as “advocacy”. But, I think it could end up guiding the conversation, in the same way that unemployment stats or inflation rates can. As W. Edwards Deming said, “What gets measured, gets improved.” He may have been too optimistic there, but I’m pretty sure that what doesn’t get measured, doesn’t get improved. Right now we’re burning asabiya (or alternatively, building up PSI) without any measurement of the fact that we’re doing it, which is likely to result in us not taking the problem seriously until it blows up. It might be helpful to at least see clearly the cliff we are approaching.

Rich Howard

I’d like to see evolutionary and behavioral science added to this model. I think ancient survival mechanisms kick in even when the threats are very “modern” like financial security. Likewise isn’t our need for “tribal membership” the driver behind fear of modern diversity?
These don’t sound like population level mechanisms, but they are if they are biological: we all have them. What are their unbiased indicators?

Vladimir Dinets

I don’t think it’s doable, considering the overall trend towards idiocracy in so many countries. The only practical way to deal with the crisis might be to prioritize personal survival. Accidentally, we are moving to Okinawa; will be working at OIST starting next September 🙂
(Yes, I know it’s impossible to run away from USA’s influence on global economy and environment, but hopefully it will buy us some time).

Loren Petrich

Part of the problem is, I think, what is broadly convincing. Some people think that we can’t find any patterns in humanity’s history, and they are at least somewhat justified from some of the less-convincing theories of history that some people have proposed.

We can see this in the history of some now-well-established theories. Continental drift, for instance. It’s long been noted that Africa and South America fit together very well. But after Alfred Wegener proposed it in detail, most geologists rejected it. How do continents plow through oceanic crust? Why did this only take place over the last few hundred million years? There were other hypotheses that could work, like sunken land bridges and former continents. But by the 1960’s, a lot of discoveries resolved these issues. Seafloor spreading meant that continents drift with oceanic crust rather than plowing through it. Also discovered were where oceanic crust was created, mid-ocean ridges, and where it was destroyed, oceanic trenches. There were some sunken land bridges discovered, like Beringia, but not many. Continental drift was also evident in polar wandering, and most recently, quasar and satellite tracking. Once geologists knew what to look for, they quickly found evidence of pre-Pangaea continental drift. In summary, continental drift got accepted only after the evidence for it got strong enough to outweigh alternative explanations.

I wish to note another Isaac Asimov analogy, with his story “Nightfall”. An alternate Earth gets subjected to periods of darkness every 2000 years where people burn their cities to make light for themselves, destroying their civilization. Some astronomers discover that it’s due to eclipses, and they prepare to have some people try to survive it — and to observe it themselves. Like to try to see the “stars” that allegedly appear.


Do you know any example form history where a society experienced a structural-demographic crisis and avoid a catastrophic outcome?

  1. Home
  2. /
  3. Cliodynamica
  4. /
  5. Regular Posts
  6. /
  7. This Time Is Different...

© Peter Turchin 2023 All rights reserved

Privacy Policy