A Quantitative Prediction for Political Violence in the 2020s

Peter Turchin

39 Comments

Join 36.9K other subscribers

In 2010 I made the prediction that the United States will experience a period of heightened social and political instability during the 2020s. Recently, several people challenged me to make this prediction more quantitative. There are all kinds of caveats, and I will get to them eventually.

But first, the TL;DR version.


David Andolfatto suggested on Twitter (@dandolfa) that I should simply use the quantitative metric that was established in my 2012 JPR article. This is a sensible suggestion because the methods are clearly described in the article and anybody will be able to check that there is no funny business once we get to comparing the predictions with data. Here’s the relevant figure from the article:

Structural-demographic theory (SDT) suggests that the violence spike of the 2020s will be worse than the one around 1970, and perhaps as bad as the last big spike during the 1920s. Thus, the expectation is that there will be more than 100 events per 5 years (see the upper panel in the figure). In terms of the second metric (the lower panel) we should expect more than 5 fatalities per 1 million of population per 5 years, if the theory is correct.

And there you have it. If violence doesn’t exceed these thresholds by 2025, then SDT is wrong.


And now for the caveats.

First, as I said on many occasions, it’s not a prophecy but a scientific prediction. So if you see me testifying before the Congress in the next few years, all bets are off. However, and realistically, political violence would probably have to reach the predicted levels before people start listening and politicians introducing reforms (if then).

Second, a critical defining feature of a political violence event, in order for it to be counted in the database, is that someone is killed (see the article why). Because modern medicine is much better at treating gunshot wounds, victims of political violence today are more likely to survive being shot, compared to 100 years ago, and that introduces a certain statistical bias.

Third, it’s important to understand that the nature of the dynamical processes generating collective violence imposes severe limits on our ability to predict them. Wars, whether between states, or internal to states, are like earthquakes. Small variations in the magnitude of the initial rupture can result in it either dissipating without much effect, or amplifying to truly catastrophic consequences. In technical language, magnitude of collective violence is  governed by a fat-tailed distribution. Large-scale events (measured by the number of people killed), while relatively uncommon, are much more probable than we think. This idea has been popularized by Nassim Nicholas Taleb in The Black Swan. While I disagree with Taleb on his pessimistic view on the prospects of a science of history (not everything in social dynamics is governed by fat-tail distributions), his point is perfectly valid when we consider the possibility of predicting major ruptures, like the American Civil War, or who knows what awaits us after 2020.

The comparison between political violence and earthquakes is not a poetic one. Statistical analysis of the frequency-severity distribution showed that the number of deaths per instability event in the United States between 1780 and 2010 (again, severity is simply the number of people who are killed) is characterized by approximate scale invariance in which the frequency scales as an inverse power of the severity:

What does it mean for making predictions? In this blog post, following my 2012 article in the Journal of Peace Research, I use two metrics in quantifying political violence: the number of such events per 5-year interval, and the number of people killed in such incidents, divided by the total US population. Both measures have problems, as is discussed in the article. In particular, the second measure is quite unstable, because of its sensitivity to uncommon, but extreme events. A single large event, such as Oklahoma City bombing, can generate a spike all on its own.

This is why I use these historical data to frame my prediction about the 2020s in both metrics.

Finally, what we really need is not ability to predict the future, but ability to predict the consequences (including unintended ones) of our interventions.

Subscribe
Notify of
guest
39 Comments
Most Voted
Newest Oldest
Inline Feedbacks
View all comments
Gene Anderson

I would agree. Every reason in terms of cycles to expect breakdowns in the 2020s.

Ross Hartshorn

Well, there it is! Dr. Turchin has placed his chips on the roulette wheel table, on red. I have to say, this looks at first blush like a make-your-reputation-or-break-it prediction. I hope you do not take it personally if I say that I hope you turn out to be incorrect. Unfortunately, while that’s what I hope, it’s not what I think will probably happen. I fear you will turn out to be correct.

An interesting article, as always.

Question: Why doesn’t your graph of “fatalities” per million match the FBI’s homicide data?

FBI data 1995-2014: https://ucr.fbi.gov/crime-in-the-u.s/2014/crime-in-the-u.s.-2014/tables/table-1

Graph:comment image?itok=-_z6lBiI

al loomis

politicians respond to violence with repression. indeed, they often seize on political violence to increase their power. which might explain why they are not often seen preventing the causes of unrest.
‘psychohistorians’ who succeed in making accurate short-term predictions will be inconvenient to such politicians.

giorgios papadopoulos

China that is a ancient country, and Russia, a big country with history of centuries, will not have to face serious internal problems in the future while USA will have, so and this will play a rule in the internal conflicts.

Richard

Not sure what you are basing that assertion on. China and Russia also have their own cycles. I’m not sure what phase Russia is in, but China is also heading in to the destructive phase of their cycle.

giorgios papadopoulos

Russian community and Chinese also, as older and organsed on historical religious, have more strength than the society of USA has, the society of USA is not able to overcome problems that has impact in all society, it is what i believe

Steve H.

giorgios, what you believe is contradictory to Turchin’s work, but he provides evidence. Please provide yours within the frame of his work.

giorgios papadopoulos

Turchin is scientist, i express just my opinion having read this ineresting blog. I will give one example, the fall of Soviet union was a great historical event that affected negatively all people of Russia, but in 10 years only Russia managed to become great again. one event with the size of the fall of Soviet union if happens in USA its society will be in chaos and i dubt if will be able to gain social balance again

Richard

In 2001, Russia was great again?

That would be news to most Russians.

Wasn’t the life expectancy of Russian men still declining dramatically?

giorgios papadopoulos

Russia had and war in Chechnya and if you remember the first systematic terrorist attack aginst the mass was in metropolitan metro of Moscow, but yes in 10 years entered in orbit of the reorganization and growth. For the life expectancy you have right, but in Russia people can suffrer anything for the sake of Motherland, can the same happen and in USA;

giorgios papadopoulos

Russia had and war in Chechnya and if you remember the first systematic terrorist attack aginst the mass was in metropolitan metro of Moscow, but yes in 10 years entered in orbit of the reorganization and growth. For the life expectancy you have right, but in Russia people can suffer anything for the sake of Motherland, can the same happen and in USA; Other example, What would happen in USA if the sosiety suffered for 8 consecutive years as Greek society suffers; I think all would be in chaos

Richard

Yes, the Greeks are very forbearing, but I still don’t understand your logic on Russia.
Suffering makes Russia great again?
How is Russia great if the men there are still drinking themselves to death?

giorgios papadopoulos

Yes but the big countries can deputize itheir population, the population of Russia is about 150 millions, the people that speaks the Russian language are more than 200 millions, and anyway the quality of life has priority in a few countries in all world

Richard

So quality of life isn’t what makes Russia great again? Then what makes Russia great again? Deputization?

Are you basically just calling for endless wars?

giorgios papadopoulos

I do not call for endless war, this is what the west makes for more than 5 centuries. If Russia was not great again it would have be destroyed by USA, the years that USA could underestimate Russia belong to the past.
For a citizen of USA the quality of life is to live many years, if it was possible and for 100 years, with pleasure, conveniences, holidays, travels, houses etc, on the earth live more than 7 billions people, for the majority of them quality is just to have somewhere to live, something to eat and a sovereign state

Richard

So not being destroyed by the US is what made Russia great again?

giorgios papadopoulos

The opposite, because of the event that Russia become strong and great again in the era of Putin it was not colonized and destroyed by USA as USA had planned

Equationist

I expect that Turchin would agree that Russia went through its crisis in the 90s and now has rising asabiya.

Winston

Actually life expectancy in Russia has risen!

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0138021

akarlin

Negative angle: Political violence is becoming normalized by the Left, at least against Nazis. Problem – everyone that disagrees with them is a Nazi. This makes for depressing reading.

Positive angle: The US still has a long way to fall to even tumble down to the level of Greece, where political assassinations do happen. But even in Greece, despite their huge challenges, and being at the forefront of the immigration crisis, nothing particularly serious has developed as a result.

Edward Turner

these people simply do not understand what is happening because they rely for their talking points and agenda on elite controlled corporations. they do not compare and contrast from a variety of sources. they do not think for themselves.

for example, CNN offers this captured group images of the inauguration crowd that appear to show sections empty implying that Trump didn’t have as large a crowd for his inauguration as Obama.

this was a lie that was denied by those there on the day and has been disproven by other photographs. the shameful lie was intended to enrage the uninformed, to program them to believe that President Trump is not really popular, when he is.

the people who take their beliefs from CNN are repeaters. all you need to do is change the broadcast for these repeaters to start repeating something else.

the broadcast will change as the reality of the deception of the previous regimes becomes glaringly exposed by the Trump administration. everything and anybody associated with the deception will be political toast without any laws or action needed by government or anyone else.

Gaelen

How’s the Kool-aid. I’m partial to purple myself.

In all seriousness, the available metrics disprove Trump’s claim’s regarding the historic size of his inauguration. These include photo’s of the day in question, ridership numbers, and nielson ratings.

Edward Turner

In terms of the photographic evidence CNN were forced to retract their inaccurate claims so I’ll pass the Kool-aid back to you. What CNN did was show a photograph when they were still letting people in before the maximum crowd size was reached. To present the photo as the maximum crowd size was a lie from a source that has previous history of publishing fake news.

The Nielson ratings do not include viewership on internet platforms youtube and on Facebook. Even so it was an historical viewership on TV alone for a Republican president beating all previous figures except Obama’s. That’s historical.

Obama’s inauguration was on MLK day so your ridership numbers are skewed.

Take a straw.

Richard

False. Obama’s first inauguration was on Tuesday, January 20, 2009. MLK Day is always on a Monday. That was the inauguration that had record-breaking numbers.
Obama’s second (public)inauguration was on MLK Day, but Americans typically ride public transit _less_ on holidays.

Trump’s inauguration had less ridership than both.

It also drew less ridership than the Women’s March the following day:
http://fortune.com/2017/01/22/dc-transit-statistics-inauguration/
And note that Americans typically ride less on weekends than weekdays as well.

Edward Turner

When using ridership figures as proxy for interest consider that Washington DC is a state that voted over 90% for Clinton and had violent protests barricading entry.

Now imagine if Washington DC voted 90% for Trump and there were no protestors threatening people as they went to view the inauguration. I could imagine the ridership being higher in that situation.

Those who made it to view the inauguration were very brave. CNN’s own TV images show the crowd stretching back to the monument. It was an historic day that pulled in massive numbers.

The image they presented for comparison with Obama’s inauguration was taken at 8:30. That they believed they could get away with the deception shows they have lost their grasp on the real world.

MLK day was the day before Obama’s inauguration. MLK was Monday, inauguration was Tuesday. It was very close to MLK day so increased turnout for him.

paolo

i think that a key factor is food and/or energy price (i don’t know if it is inside Ages of discord, in travel but it seem a loooong way to Italy black hole :)).

in 2008 and 2011 food price was http://www.fao.org/fileadmin/templates/worldfood/images/home_graph_3.jpg
Am J Agr Econ (2014) 97 (1): 1-21…..”Results indicate that for the period 1990–2011, food price increases have led to increases in social unrest, whereas food price volatility has not been associated with increases in social unrest.”

steven t johnson

Seems to me the levels of political violence by the state should include excess mortality inflicted by the state, in arrests and premature deaths in prison. Obviously there are enormous difficulties since these statistics are typically not even kept, for political reasons. But it seems to me that there might be a great deal of uncertainty.

Perhaps a focus on mortality statistics because of their relative completeness and reliability is a little like a drunk looking for his keys under the street lamp because that’s where the light is? No doubt it is difficult to assess the data (if it’s even collected,) but percentage of population in prison seems like a relatively useful proxy for repression by the state. Levels of censorship, expropriation of property, termination of employment, suppression of trade unions and political organizations could possibly provide an alternate methodology?

Ross Hartshorn

I think the idea is that these particular kinds of violence are of special interest because:
1) they’re not just one person mad at a specific other person, they involve many people on either the aggressor or the victim side (or both); thus they have a social element, they’re involved with the attitudes towards society
2) they’re not organized or directed by the government

This doesn’t mean that other kinds of violent death are any less bad, especially if you’re the victim or loved them. But, it does indicate a different kind of cause. It relates to the ability of the government (or the elite generally) to channel and direct violence, in those cases where it can’t or won’t eliminate it. So, if a government starts a war with a neighbor as a pretext for oppression, or sends out the troops to fire on peaceful protesters (to take two of many possible examples), it’s just as bad, but it doesn’t indicate that they are losing control of the situation. Ditto if there’s an increase in people getting drunk and arguing with their cousin until someone shoots the other; that’s bad, but it doesn’t indicate that the ruling elite are losing control of the society.

This particular kind of violence, one-on-many or many-on-one or many-on-many, is being used as a metric for the ability of the ruling elite to keep control of things, rather than any more general metric of whether bad things are happening or not.

Nonetheless, your idea on other metrics for detecting the signal of a government getting desperate and worried about their ability to keep a lid on things, is a good one. I don’t know about percentage imprisoned, since (for example) the Prohibition of alcohol in the U.S. was said to have increased imprisonments related to organized crime. Some of your other ideas might work, though. Any ideas on how to measure such things, across time?

Caroline Charlese Scott

Just this week, two explorations of the very believable specter of political violence.

The New Yorker, A Reporter at Large, January 30, 2017

“Doomsday Prep for the Super-Rich
Some of the wealthiest people in America—in Silicon Valley, New York, and beyond—are getting ready for the crackup of civilization.”

http://www.newyorker.com/magazine/2017/01/30/doomsday-prep-for-the-super-rich?utm_source=The+Federalist+List&utm_campaign=4f3831db66-EMAIL_CAMPAIGN_2017_01_28&utm_medium=email&utm_term=0_cfcb868ceb-4f3831db66-83888249

In the web journal, Naked Capitalism, an opinion essay followed by many comments concurring that we are facing a descent into conflict, figurative or literal civil war.

James F. (commenter) on Hatred in Our Divided Nation: Anger at Flyover Country
Posted on January 27, 2017 by Yves Smith

The essay opens:
“I, too, am worried by our descent into prewar hatred. … Civil wars seem implausible until they start and then they follow the devil’s logic. People like my friend tend to die in them or turn into something less than they were in order to survive. …

http://www.nakedcapitalism.com/2017/01/james-f-on-hatred-in-our-divided-nation-anger-at-flyover-country.html?utm_source=feedburner&utm_medium=email&utm_campaign=Feed%3A+NakedCapitalism+%28naked+capitalism%29

Paolo Ghirri

i live in italy so i don’t know if this can be keep seriously but …….http://www.yescalifornia.org/

Loren Petrich

The “Calexit” activists are trying to get enough signatures to get their secession proposal on the 2018 ballot. This proposal is to have a vote on secession in early 2019. But actual secession would require an amendment to the US Constitution, and that is very hard to amend. It will need 2/3 of both houses of Congress and 3/4 of state legislatures agreeing, or else a political joker in the deck, a Constitutional Convention.

Secession is a loose end in the US Constitution. Its original creators never addressed that issue, and neither did any of its amenders. The closest thing to settling it was the US Civil War, provoked by the South’s unilateral secession. That was resolved by the South’s defeat. Yet I’m sure that secession is feasible if enough politicians agree. Either a split by mutual agreement or else kicking out some states.

California is far from alone. Numerous states and regions have had secession movements, most recently Texas, Alaska, Hawaii, Vermont, New Hampshire, and the Pacific Northwest. But most have not gotten very far.

Paolo Ghirri

thanks Loren. i know that secession movement are present in a lot of states. but in 1860-1861 the southern secession is a reaction to Lincoln election. On November 9, 1860 the South Carolina General Assembly passed a “Resolution to Call the Election of Abraham Lincoln as U.S. President a Hostile Act” and stated its intention to declare secession from the United States.

the Calexit say (15 nov 2016) “Last week, the Americans chose Donald Trump as their president. We Californians overwhemingly cast our ballots for another candidate who ended up winning the popular vote but losing the election. This campaign is not just about Donald Trump, though. It is about the American people who elected him. If they could do such a thing, then the United States is not our country and Trump is not our president. We will vote to secede from the Union.”

so my question is: there are resons to thinks that they have realistic possibility to win referendum?
and (to Truchin our teacher) can be this considered a instability event (even if don’t lead to a ciwil war 2.0) ?

[…] Twitter50Facebook60Google+2In 2010 I made the prediction that the United States will experience a period of heightened social and political instability during the 2020s. Recently, several people challenged me to make this prediction more quantitative. There are all kinds of caveats, and I will get to them eventually. But first, the TL;DR version. David Andolfatto suggested on Twitter (@dandolfa) that I should simply use the quantitative metric that was established in my 2012 JPR article. This is a sensible suggestion because the methods are clearly described in the article and anybody will be able to check that there is no funny business once we get to comparing the predictions with data. Here’s the relevant figure from the article: Structural-demographic theory (SDT) suggests that the violence spike of the 2020s will be worse than the one around 1970, and perhaps as bad as the last big spike during the 1920s. Thus, the expectation is that there will be more than 100 events per 5 years (see the upper panel in the figure). In terms of the second metric (the lower panel) we should expect more than 5 fatalities per 1 million of population per 5 years, if the theory is correct. And there you have it. If violence doesn’t exceed these thresholds by 2025, then SDT is wrong. And now for the caveats. First, as I said on many occasions, it’s not a prophecy but a scientific prediction. So if you see me testifying before the Congress in the next few years, all bets are off. However, and realistically, political violence would probably have to reach the predicted levels before people start listening and politicians introducing reforms (if then). Second, a critical defining feature of a political violence event, in order for it to be counted in the database, is that someone is killed (see the article why). Because modern medicine is much better at treating gunshot wounds, victims of political violence today are more likely to survive being shot, compared to 100 years ago, and that introduces a certain statistical bias. Third, it’s important to understand that the nature of the dynamical processes generating collective violence imposes severe limits on our ability to predict them. Wars, whether between states, or internal to states, are like earthquakes. Small variations in the magnitude of the initial rupture can result in it either dissipating without much effect, or amplifying to truly catastrophic consequences. In technical language, magnitude of collective violence is  governed by a fat-tailed distribution. Large-scale events (measured by the number of people killed), while relatively uncommon, are much more probable than we think. This idea has been popularized by Nassim Nicholas Taleb in The Black Swan. While I disagree with Taleb on his pessimistic view on the prospects of a science of history (not everything in social dynamics is governed by fat-tail distributions), his point is perfectly valid when we consider the possibility of predicting major ruptures, like the American Civil War, or who knows what awaits us after 2020. The comparison between political violence and earthquakes is not a poetic one. Statistical analysis of the frequency-severity distribution showed that the number of deaths per instability event in the United States between 1780 and 2010 (again, severity is simply the number of people who are killed) is characterized by approximate scale invariance in which the frequency scales as an inverse power of the severity: What does it mean for making predictions? In this blog post, following my 2012 article in the Journal of Peace Research, I use two metrics in quantifying political violence: the number of such events per 5-year interval, and the number of people killed in such incidents, divided by the total US population. Both measures have problems, as is discussed in the article. In particular, the second measure is quite unstable, because of its sensitivity to uncommon, but extreme events. A single large event, such as Oklahoma City bombing, can generate a spike all on its own. This is why I use these historical data to frame my prediction about the 2020s in both metrics. Finally, what we really need is not ability to predict the future, but ability to predict the consequences (including unintended ones) of our interventions. – Read full story at Hacker News […]

Mike Alexander

Since most of the current events in the USPV database are rampage events, which are pretty deadly, I suspect all such events will be entered into the database. If we compare the USPV database for rampage events to the Mother Jones mass shooting database we see the following:

Period USPV MJD Ratio
1982-1990 12 4.5 2.7
1991-2010 30 11 2.6
2011-2016 70? 27

If the USPC database follows the track of MJD we would project ca. 70 rampage events per 5 years in recent years–more than half way there.

[…] First of all, I think the revolution would start in the US due to its current political tensions, and its penchant for change and fluidity. I also think it’ll happen sometime this decade, or at the very least the next, according to Peter Turchin’s mathematical predictions of societal crises. […]

  1. Home
  2. /
  3. Cliodynamica
  4. /
  5. Regular Posts
  6. /
  7. A Quantitative Prediction for...

© Peter Turchin 2023 All rights reserved

Privacy Policy