Hacker News new | past | comments | ask | show | jobs | submit login
Human cycles: History as science (nature.com)
55 points by ananyob on Aug 2, 2012 | hide | past | favorite | 34 comments

I'm very sceptical. I listened recently to economic experts talking in the 90s, they were all saying we were in the ascending phase of some macro cycle, and that we had thirty years of growth ahead. None did forecast the crisis we had since. They also did not say a word about the economic equilibrium bending east.

Please take these cycle theory with a grain of salt. It is too easy to find patterns and to invent explanations for them, but there is no more reason to have recurring patterns in history than to have patterns when playing with a dice.

A comparison with the Lotka-Volterra equation


which are the mathematical explanation of predator prey cycles, shows what's missing in the article and possibly in the research - the explanation for the cycles.

Predator-prey cycles can be tied back to biological constants - how quickly prey and predator reproduce, how the availability of prey affects predator reproduction, etc. Thus, if you want to model wolves and sheep rather than foxes and rabbits, you can do so, if you know the various biological constants.

I don't get the sense that Turchin has identified the underlying dynamics.

> I don't get the sense that Turchin has identified the underlying dynamics.

Then you should read his work. He describes what he thinks are the driving dynamics of the rise-and-fall process.

Ok, so I looked at the article again, and got that they focus on population, social structure, state strength, and political instability. It's even more interesting that they identify two cycles with at least some dynamics:

1. the "secular" cycle: in the beginning of the cycle, supply and demand for labour roughly balance out. In time, the population grows, labour supply outstrips demand, elites form and the living standards of the poorest fall. At a certain point, the society becomes top-heavy with elites, who start fighting for power. Political instability ensues and leads to collapse, and the cycle begins again.

2. the "fathers and sons" cycle: the father responds violently to a perceived social injustice; the son lives with the miserable legacy of the resulting conflict and abstains; the third generation begins again.

Thanks for the nudge.

Peter Turchin published a highly readable pop-science account of his work in a book called War and Peace and War. I read it a few months ago and gave a review.


What's left out of this link is that Turchin still hedges his findings heavily. He's not performing precise predictions; he can't. He's pointing out dynamics that acquire massive social inertia and eventually force the phase change of whole societies.

I hope, in this case, given the really close parallels, it wont be facetious to say: Hari Seldon! Fact following from fiction?

This is also one of the "big ideas" in the book _Eifelheim_, under the name "cliometrics".

SPOILERS (just little ones, but still)

Essentially, a historical visitation by aliens is revealed due to the failure of a certain group of people to fit the mathematical model of what their history should have been- the arrival of aliens was not considered as a factor in the theory.

He talks a bit about Hari Seldon in his work.

To give HN skimmers some context: Hari Seldon is a key character in Isaac Asimov's Foundation series. He develops a new science called "psychohistory", which enables accurate prediction of the behavior of extremely large (galactic scale) groups of people.

When I read the first book of the series, an idea that resonated with me was the marginalization of scientists. I kept seeing aspects of current society reflected in his work. What I wonder is if scientists felt the same way at the time Asimov wrote the series?

P.S. It is a fantastic series of books. I encountered them my chance in the last few years. Highly recommended!

I find it interesting that WWII and the Korean War do not show as spikes in his graph.

It's a US-centric graph, if I'm not mistaken. But it is interesting that WW2 doesn't trail effects. Although, there is a long-standing theory that WW1 and WW2 were just the same war.

Methinks Alkamie disguised a call of 'bullshit!' behind a sort of empty platitude such as "I find it interesting".

>It's a US-centric graph

The title is "Human cycles: History as science"

A US-centric graph that represents the cycles of humanity?


Do we need a fancy name for the phenomenon that aspects of history tend to repeat?

Yes, aspects of history tend to repeat, but there's no way to use that knowledge to determine periodicy, severity, when it will happen, if it will happen etc.

I'll say what Alkamie didn't. This is bullshit.

sssssh You'll ruin the illusion that (Societal Breakdown|End of Moores law|Space Travel|Strong AI|etc.) is only ~10 years away!

I wonder what he thinks of Steven Pinker's paradigm[1] that world violence has been declining for centuries and is the lowest it's ever been.

Besides, there will be too many old people in the US to have mass riots in the 2020's.

[1]One example: http://online.wsj.com/article/SB1000142405311190410670457658...

Sounds like a good long bet (http://longbets.org/). I'd be willing to bet that "Worldwide, the 2020s will be less violent than any preceding decade in the 20th or 21st century".

I'm going to have to track this guy's book down. Subject looks interesting, and the nature article leaves some information lacking, sadly.

For instance, Turchin mentions seeing this pattern in Ancient Rome and Egpyt (past and present) but he doesn't explain how he accounts for the Middle Ages and European colonial expansion. Arguably the number of "violent" events during those periods would be concentrated to the point of potentially breaking the father-son-grandson model he advocates. The chart in Nature leaves out things like the Reconquista of Spain, Ghengis Khan and heirs marching into Europe, the Crusades, the 50 years war, the French Revolution, and numerous other major upheaval events that occurred in smaller windows than those 50 year cycles he claims.

I guess I want to know what happens when you add in those time frames missing from his graph. Does it show the amount of "conflict" in the world decreasing with time as civilization (our understanding of it) expands around the world?

Humans see patterns where there are none.


I think that applying statistical methodology to the study of history might provide some insights into cause-and-effect and the situations surrounding historical events. But, I seriously doubt that there are any "cycles" to be found.

For one thing, major events in human history absolutely are (or have been) pivotal on just one or a few individuals. A fun book on this is, "What If?", by Robert Cowley. Also, "How the Scots Invented the Modern World". Being the right person, in the right place, at the right time has been a recurring theme throughout history. For example, the final vote for the unification of Scotland and England was led by one passionate, older man -- who died of a stroke later that night. Had he died the night before, it's likely that his opposition, an almost equally skilled and passionate orator, would have won the vote instead.

The fathers-and-sons cycle described in the article is a relatively recent phenomenon, where society is settling down into a more peaceful state overall, and even then, the two greatest wars in human history violate the cycle: World Wars 1 and 2. Popular sentiment in the U.S. was against being involved in World War 2, after the horrors scene by the previous generation in World War 1, so the cycle isn't completely without merit. But, as usual in human history, there are too many other factors involved in the outcome.

I also think that history changes a lot depending on how you look at it. There are so many human societies on the planet at any given time, each one doing so many different things, that it's easy to step too far back in perspective by ignoring many of the events of history, and then start to think that you an discern some kind of a pattern or cycle. One of my favorite books for this is "Chronicle of the 20th Century", a collection of short-form news articles and pictures from the U.S. from 1900 through 1987. Flipping through that, it's always stunning just how much is so easily forgotten, how rich and complex each day was for the people living through those years.

How does he account for the fact that generations do not proceed in lock-step across families? Some parents have children earlier than others - siblings can be spread across ~15 year periods, etc.

I'd hope he'd pretty easily be able to generate graphs like this for every society where the data is available - should be at least a dozen or so over the last 150-200 years. Though I'd imagine if his theory is correct, the time spans would vary slightly from society to society based on average age of parents at the time of childbirth.

3 violent "events" happening equidistant from each other does not mean a 4th is inevitable in the same timeframe. If I see proof back hundreds of years, then I would listen. Either way, pretty safe prediction that humans will act violently in the next 10 years.

His model proposes that the timing of the mesocycles of violence is set by the father-grandson dynamic and that the timing of the macrocycles of decline is 3 or 4 mesocycles once the empire has reached its zenith.

His prediction isn't just based on the three spikes though. And the deeper pattern he sees in the US holds in other countries.

Anyone who finds this interesting might also find the Strauss-Howe generational theory (1) of interest. It talks about the generational cycle in America.

(1) http://en.wikipedia.org/wiki/Strauss-Howe_generational_theor...

So the question is, if there comes a period of upheaval in the next decade ... what could a well-meaning entrepreneur work on now that will either help socially (secure communications, ad-hoc networking to avoid government lockdowns). Or maybe even position themselves for profit (3d printing tech to manufacture weapons, for example)?

It would mean political entrepreneurship would be more valuable than the fundamental kind - witness who tends to do well in post-Soviet states or even post-revolutionary America. Or, if one rejects political involvement, relocating to an expected stable state.

curiously, I have a friend who's family hails from georgia. Their old(ish) money came from after the civil war, when his great(N) grandfather started shipping grain and other desperately sought after food items to heavily affected civil war states/cities. A relatively small investment ended up multiplying and benefiting many generations :)

The police have never been more militarized, even simple protesting against something is more and more likely to at a minimum get you arrested if not shot or "less lethal" weapon used against you like pepper spray in the face despite you being unarmed and non-violent (while the police will receive a paid vacation for doing so).

In 1st world countries I think protests are doomed. Wait till they start rolling out the sonic canons and drones at the RNC and DNC conventions this year and see if you agree. Oh and as a bonus, the millions spent on "security" at the conventions is taxpayer funded - congress votes on it for themselves every time.

Tell that to the student protesters at Kent State in 1970. 4 killed, 9 wounded (one of whom was permanently paralyzed), in a peaceful protest. It was a time when the national guard was regularly called in to break up peaceful protests.

There isn't an increase in violence, there's an increase in information now. There weren't cellphone videos of police violence on youtube in the 1970s (neither existed). There weren't 24 hour news stations who needed any content they could get (violence sells). We're over-saturated with ways to see current violence, and it can make it seem like it's happening more now (there's more reports of it, so it seems like it must be happening more). That's not at all the case.

Are you aware of history at all? Police in Western world have much less power than they used to have. One example was the Riot Act in UK ( http://en.wikipedia.org/wiki/Riot_Act ) which gave local authorities the power to declare 12 or more people a riot, and if they didn't leave within 1 hour, they could be legally killed.

Protesters have had much harder times than "sonic cannons". Try soldiers shooting into a crowd (e.g. Bloody Sunday in 1972, Northern Ireland http://en.wikipedia.org/wiki/Bloody_Sunday_(1972) )

John C Dvorak has a theory, I am not sure if it is just his about cycles and a predicted pattern of events going back to the 1600's that indicates a dip in things starting in 2017.

Personally I am an optimist and think that as a whole we will continue to persevere and generally get it right more often, with more free information, we learn faster and eventually we evolve our outlook and efforts for the better.

Not because we want to, it's something about our species and tenacity to survive happy.

Psychohistory? Where is Hari Seldon when you need him?

In the fiction aisle.

Bankers are creatures of habit http://www.prolognet.qc.ca/clyde/pres.htm

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact