Why Theory Is Important: Responding to two reviews by Scott Alexander

Peter Turchin


Join 36.9K other subscribers

Scott Alexander wrote two reviews of my work on the structural-demographic theory: Book Review: Secular Cycles and Book Review: Ages of Discord. The first one, on Secular Cycles, is quite positive, but the tone of the second review is a bit uneven–generally positive, but with some notes of critique and suspicion. His suspicions of the data and theory in Ages of Discord (AoD), however, are misplaced. Still, I would be the first person to admit that AoD is a highly technical and, thus, difficult book — full of models, formulas, tables, and graphics. I am constantly thinking about a popular version of AoD, but I haven’t yet figured out how to lay it out for the general public (I am also currently completely occupied with the analysis of Seshat data). Here I offer a few thoughts in response to issues raised by Alexander.

While what follows may sound critical of his review, I emphasize (and reiterate at the end) that I greatly appreciate the amount of time and effort he invested into reading and digesting my books.

First, Alexander starts the review of Ages of Discord with on overview of empirical patterns and only much later gets to theory. This is not how AoD is organized and, as a result, he ends up confusing himself and, I think, readers. AoD is only the latest installment in my work on structural-demographic theory (see Chapter 7 and 8 in Historical Dynamics published in 2003). By the time I wrote AoD, structural-demographic theory has matured to the point where one could (and I did) make predictions about novel cases. AoD, thus, is mainly an empirical test of predictions for the USA between 1780 and 2010. It is true that the US has gone through only 1.5 secular cycles in its history, but my identification of these cycles is not based on just these 1.5 “points”. The case of the US is an “out of sample prediction”, as it is known among the analysts.

Predictions were listed in Table 1.1 of Secular Cycles, published in 2009. I devoted two chapters at the beginning of AoD to extending the structural-demographic theory to industrializing societies and refining the predictions for the US. These two chapters, however, have very significant mathematical component. By his own admission, mathematics is not one of Alexander’s strong suits, which is probably why he jumps into data right away. But by doing this, his review fails to give justice to the logical structure of AoD.

Second, Alexander does some “spot-checks to see whether the data are any good”. This is somewhat strange — all data sources are listed in AoD, does he think that I have falsified them? Naturally, he concluded that “Turchin’s data all seem basically accurate.”

Next, in an attempt to check whether I “cherry-picked the data series that worked”, he looks at a variety of random indicators, for example, treasure bonds. Here again, by missing the theoretical part he doesn’t do justice to AoD. The variables I focus on all follow from general theory. There are fundamental variables in the theory that drive the dynamics (immiseration, elite overproduction, state strength, and socio-political instability) and there are “proxies” — variables closely correlated with the fundamental drivers. Then there are variables about which the theory is silent. Treasure bonds in no way are part of the theory, and I have no idea whether they would correlate with anything important, so I never looked at them.

There are also variables that are affected by fundamental ones, but are not part of the feedback web (they don’t affect the main drivers). For example, homicide rates. We expect that popular immiseration and elite overproduction would result in social pressures for instability, and one surface indicator of that could be growing homicide rates. I discuss homicide data in both Secular Cycles and AoD. However, one should keep in mind that there are many other factors, apart from structural-demographic ones, acting on this class of variables. Thus, we do not necessarily expect a perfect correlation. In fact, while structural-demographic pressures continued to grow during the last four decades, homicide rates actually declined during the 1990s. One possible explanation is that incarceration rates have quadrupled over this period of time. But the more important point here is that homicide rates are not a fundamental driver in the theory.

A particularly strange indicator that Alexander looked at is the USD/GBP exchange rate. I have no idea why he did that. Once again, in my research on structural-demographic dynamics I do not trawl through the thousands of time-series available on the web to look for correlations. Such trawling inevitably will yield correlations, but with high probability such correlations will be spurious.

Third, one has to be careful with data on meaningful indicators, and examine its provenance. In his first review (of Secular Cycles), Alexander starts by showing “Chinese population over time.” He doesn’t specify the source, but I recognize it — it’s McEvedy and Jones. 1978. Atlas of World Population History. Unfortunately, this resource is quite dated. Furthermore, it smooths out a lot of cycles in the data. Compare it with this chart (for provenance, see Historical Dynamics):

Quite a difference! Not everything in this graphic is solid, but the main point I am making is that one simply can’t grab the first available chart; it’s important to give thought to data sources and to understand their limitations.

Fourth, you cannot take on faith various opinions — myths is not too strong a word — propounded by social scientists; they have to be evaluated critically. This is particularly true of economics, because economists have an enormous vested interest in propounding theories that would please various powers-that-be. I’ve written about it in several of my blog posts, e.g.

Economics Superbus I

Economics Superbus II

When Alexander takes issue with one of the fundamental processes in structural-demographic theory, that oversupply of labor tends to depress its price, he says: “Hasn’t it been proven almost beyond doubt that immigrants don’t steal jobs from American workers”? Alexander refers to a survey of top economists for this. I’ve written about how much we can trust what economists say to the public here:

What Economics Models Really Say

So I am on the side of Harvard Professor George Borjas, who’s careful lifelong work leads him to conclude: “The best empirical research that tries to examine what has actually happened in the US labor market aligns well with economic theory: An increase in the number of workers leads to lower wages.”

Despite these disagreements, I want to emphasize that I quite appreciate the amount of time Scott Alexander invested in reading my work, especially because AoD, let me repeat, is not the easiest book to read.

Let me finish this post with a quote from Steve Sailer,

I think Turchin doesn’t get much attention because his books are too reasonable to be easily debunked and too enormously detailed to be easily digested and too ambitious to be easily trusted.

I am afraid I can’t argue against this assessment; the only thing I can do is to continue doing my work, to the best of my ability.

Notify of
Most Voted
Newest Oldest
Inline Feedbacks
View all comments

I think you are spot on with your assessment on economists.

I have had similar experience when I showed one of them parts of your work, the reaction was dismissive and almost arrogant.

The depressing thing is that I think given the high status of economics in the scoial sciences they have a big influence on policy makers and peoplea opinions in general.

But this means when people start to behave more in line with what economsits predict how people behave, society starts to unravel faster, so I think they actually have a negative influence of social cohesion of a society in general.

To me economics seems more like formalised ideology than actual science

Loren Petrich

I mentioned Peter Turchin’s work in an online forum that I frequent, and I got a sizable response that I’ll summarize as “Is this a joke? … It looks to me like this is the greatest case of theoretical overreach in Historical Analysis since Marx.”

Turning to elite aspirants, Turchin identified Donald Trump as one. But there is now a loose movement of EA’s on the Left, a movement that attempts to acquire seats in Congress, much like the Tea Party of some years earlier. Most of these EA’s have run as Democrats, trying to defeat established ones in the primaries. The best-known of them is Alexandria Ocasio-Cortez, She had a remarkable come-from-behind victory over 20-year-incumbent “King of Queens” Joe Crowley. She and three other such first-time Representatives have become close friends, nicknamed “The Squad”. All indications are that there are likely more to come in 2020.

Loren Petrich

It’s good to collect that data. Continental drift was not accepted for a long time because much of the evidence offered for it could be explained by land bridges and the like. But in the 1950’s, seafloor spreading and paleomagnetism data started coming in, and they decisively supported drift. Likewise for some other scientific revolutions, like dorsoventral inversion. Vertebrates as upside-down invertebrates was proposed off-and-on but not generally accepted until recently, when patterns of morphogens were discovered to also have that inversion.

Loren Petrich

https://en.wikipedia.org/wiki/Cyclical_theory_(American_history) – Arthur Schlesinger’s cycle of liberal vs. conservative, reform vs. stagnation. They fit fairly well into the SDT.
* Integrative: L Constitution, C Hamilton, L Jefferson, C After 1812 War (Era of Good Feelings)
* Disintegrative: L Jackson, C Slaveowners, L Reconstruction, C Gilded Age
* Integrative: L Progressive, C Roaring 20s, L New Deal, C Eisenhower
* Disintegrative: L Sixties, C Gilded Age II

The US also has Frank Klingberg’s foreign-policy cycles, but they run on a separate clock.

As to breaking the SDT, I think of it as like Isaac Asimov’s story “Nightfall”, where some astronomers attempt to break the cycle of periodic destruction of their civilization.

Loren Petrich

President Trump seems just the president for a disintegrative period. His presidency has been very disruptive, and not in a good way. It has also involved a lot of what I like to call moral ugliness. He also seems to want personal rule as a dictator, rather than as a democratically elected leader.

There is another bizarre feature of him: low conscientiousness – https://blogs.lse.ac.uk/usappblog/2016/11/02/what-trump-and-clintons-personality-traits-tell-us-about-how-they-might-govern-as-president/ This is evident from his impatience with briefing papers and his spending much “Executive Time” watching Fox & Friends and sending out Twitter messages (tweets). He also spends 10 times more times golfing than his predecessor Barack Obama did.

That is a personality trait in the https://en.wikipedia.org/wiki/Big_Five_personality_traits and it includes diligence and orderliness. C’ness is positively correlated with academic and career success. Steven J. Rubenzer, Ph.D., and Thomas R. Faschingbauer, Ph.D. in “Personality, Character, and Leadership in the White House: Psychologists Assess the Presidents” (2004) find that most presidents are medium to high in c’ness.

By comparison, Hillary Clinton, his opponent, has high c’ness. His predecessor, Barack Obama, also does https://core.ac.uk/download/pdf/3146078.pdf Looking at Congresspeople, Nancy Pelosi has a “legendary” “work ethic”, more c’ness https://www.businessinsider.com/nancy-pelosi-2013-3 and Mitch McConnell is also big on that https://www.nationalreview.com/2013/07/mitch-mcconnell-darth-vader-robert-costa/ Alexandria Ocasio-Cortez has shown great diligence in her career, and she has praised diligence as something that can make you do nice things https://www.instagram.com/p/B139mhTg4_1/ – c’ness yet again.

So most presidents, Barack Obama, Hillary Clinton, Nancy Pelosi, Mitch McConnell, and AOC all have high c’ness. Which makes Trump very exceptional. Which also raises the question of what has kept Trump’s career going – it has survived several bankruptcies.

Loren Petrich

Oops. I wanted the previous post to be a top-level one.

As to problems with economics, I once remember going to an Austrian-economics IRC channel and asking about the problem of businesses spending on advertising. If consumers’ preferences are arbitrary, as Austrian economics seems to hold, then advertising would be a pointless waste of money. I didn’t get a good answer to that.


Austrians (the believers of the theory), while holding a few insights, tend to be wrong about a lot of economics, so I don’t find that surprising.

Loren Petrich

Could a part of it be the ready availability of computer resources for doing simulations? That would be an outcome of having software for doing heavy statistical number crunching. Judging from http://www.99-bottles-of-beer.net/ much of such software is Turing-complete or close to it: R, SAS, SPSS, Stata, …

The simulations would be for understanding the dynamics of economies by creating systems of differential equations for toy economies and then solving those equations.

What has been done there? Has anyone tried to simulate boom-and-bust cycles or economic bubbles or other such economics effects?

Graeme Bushell

Loren, yes, Steve Keen has done a lot of work in this area with his “Minsky” simulator. It models endogenous money creation and credit cycles emerge naturally. I don’t think it includes anything about demographics.
No-one has put this all together yet: a dynamic simulation of the ecosystem (like World3) that includes population demographic effects (SDT) and real-world economics (Minsky).


As a regular reader of both your blog and SSC don’t expect your (entirely justified and IMO understated) criticism of economists to be taken without a tedious fight, Scott himself might be somewhat sympathetic but his commentators treat neoclassical free market propaganda as self evident.

Dick Illyes

I think the exposure of your ideas on SSC was valuable. This a very unusual period in human history. It is possible for humans to evaluate their society in ways never before possible. Your work is at the leading edge of that evaluation. Followers of SSC are very interested in those issues and I think the overwhelming majority were impressed positively.


I don’t think it’s over compensation at all.

I think because people are part of the same society everybody has an opinion on it and when the mechanisms are not easily visible conclusions can be wild, ending into quack land.

I think it’s very important to be as rigourous as possible to seperate yourself from that

Ross Hartshorn

There are some economists who are fundamentally analogous to scientists, trying to improve theory until it aligns with reality. There are others who are fundamentally analogous to lawyers, trying to improve arguments until they align with a predefined position. Not that physics, chemistry, etc. are immune to politics, but for obvious reasons the emotional attachment to positions is even higher in economics; the positions on how immigrants impact the labor market are a case in point. It is patently obvious that they must, but the political implications are unsavory, so economists do mental backflips to avoid the obvious conclusin. This is not to mention the incentives to argue in favor of positions which people with money to donate for research, endowments, etc. are in favor of (e.g. low tariffs if you are head of a company that wish to import or export).

I’m not aware of a change in any field of study, that was comparable in scope to that which cliodynamics requires, happening in less than a generation. So, it’s not too difficult to imagine that cliodynamics will take a while to rise. Also, the reality is that nearly everyone has some limit beyond which math is too complex for them. If they perceive that their field of study is headed in a direction which will go beyond their own, personal limit, they are not going to respond favorably. This also requires a newer generation of researchers, who are comfortable with the level of math required, moving into the field.

Which brings up a question for Dr. Turchin. Right now, if someone were interested in cliodynamics and interested in working in the field, does an introductory textbook exist? If a university wanted to add a course on cliodynamics, is there a curriculum that exists that allows them to do so?

steven t johnson

Steve Sailer is a notorious crank on race, IQ, etc. This doesn’t mean he is unable to still see blatant gaps and contradictions between economic and actual history. But it does raise questions about your ability to your ability to analyze historical data, particularly in identifying superior “nations” worthy of analysis and demographic trends of peoples not worthy of analysis. Do you endorse human biodiversity movement, evolutionary psychology, etc?

It’s basically the notion that a person who believes one conspiracy theory believes a lot of them. In particular, a crank attempt to replace Marx’s influence with real Popperian science could be motivated by commitments to reactionary ideals.

Haven Monahan

Steve Sailer has some notoriously banal takes on race and intelligence. First he insists that race is a real biological phenomenon, and secondly he states that observed differences in the intelligence levels of racial groupings is not all due to environmental factors, some of the differences are due to the heritability of intelligence. Every year more and more scientific papers are released that support these two hypotheses.

But it’s Sailer’s insistence on the law of supply and demand as it applies to labor where he is in direct conflict with the interests of the wealthy oligarchy in the US. How much of each billionaire’s fortune is due to the decrease in labor costs brought on by mass immigration into the US? George Borjas estimates billions of dollars each year are transferred to the wealthy through higher corporate profits brought on by wages depressed by immigration. So it is quite normal that the oligarchs have purchased PR people wearing the cloaks of economists to try to convince the public that oversupply of labor has nothing to do with stagnant salaries.

steven t johnson

The thing is, I don’t think an endorsement from Steve Sailer is on the point here either. And worse, his judgment is so dubious it counts more as an anti-endorsement. HBD is just the newest form of scientific racism and Sailer is both a crank and a political zealot eager to pretend to objectivity, in order to proselytize. It ‘s not really flattering. My judgment, of course.

The post on Pinker is fairly good, though it concedes far too much in treating Pinker as a legitimate authority. (I think his only true expertise is in linguistics.) For one thing, Pinker has a nasty habit of switching from Darwinian fitness (reproductive success) to simple survival, in the most tautological Spencerian version. Monasticism is a repeated and spontaneous development in a broad range of technologically developed societies that directly impacts Darwinian fitness. Invoking kin selection and reciprocal altruism here is mumbo jumbo. By the way, the ancient Spartan Equals/Peers are an example of an elite that sacrificed Darwinian fitness. One of the reasons for the relative decline of Spartan power was the relative decline of Spartan hoplite soldiers, the Equals/Peers.


You sound like the zealot here, all opposition to HBD is rooted in morality, not evidence or logic.


A belief in some ‘human biodiversity’ merely requires the belief that genes influence IQ and that there are (in origin) geographically-related genetic differences. I believe that the evidence for these is quite strong. My opinion is that some gene-based variation in IQ between semi-isolated populations pretty much has to be true.

The real question, in my eyes, is the extent to which populations differ in IQ due to genetics, rather than because of environment, which is extremely unclear, because the research into this is rather poor in quality, because environmental differences nearly always confound the research. So this difference can be small or not so small.

I personally don’t see Steve Sailer as being appreciatively more of a crank and a political zealot than very many others. His particular crankiness and zealotry is less politically correct, but many mainstream beliefs are less supported by evidence and/or have more counter evidence.

As for Steven Pinker, he is extremely sloppy. He is prone to only presenting evidence in favor of his thesis and not being critical enough of the evidence that seems to support his thesis. He falls in the common ‘soft sciences’ trap of taking on way more than he can chew & sacrificing scientific rigor to make up the difference. IMO, this lack of rigor is what allowed him to become famous, as most people dislike true rigor or even half-rigor. Writing that recognizes the limits of the evidence and that explains things properly tends to put off nearly everyone. Popularity is usually a bad sign.


The problem I have with the HBD scene isn’t so much the field but many of the people in it. Namely, many/most of them are junk scientists. “Prone to only presenting evidence in favor of his thesis and not being critical enough of the evidence that seems to support his thesis” describes their actions well and many/most of them are incredibly sloppy.
(When someone states that a cultural trait is prevalent in a locale when I am from said locale and know that to not be true, their credibility pretty much craters).



I fully agree, although because of the taboo on drawing even rather obvious and scientifically well supported conclusions about the possibility of IQ differences between groups, moderate and reasonable people rarely want to get involved in the debate, out of fear of being ostracized. Even the NYT wrote articles lamenting how genetic scientists are unwilling to engage with the audience, resulting in the available information on the subject being largely from the far-right. Yet they didn’t seem to understand that a likely reason is that these scientists may be in a conundrum, with their scientific conclusions clashing with modern (center-left) dogma. Nor do they seem to recognize that the NYT is itself part of a culture that vilifies certain beliefs for their possible implications and frequently assumes that beliefs are adopted because of their implications. Yet the logical result of such a culture is that even scientists stay silent when their findings clash with fiercely defended dogma, out of a very reasonable fear of losing friends and their job.

Peter van den Engel

The problem with economic theory is it is looking for constants in the equation for prediction/ while, because the economy is an evolutionary process as well, it is based on dynamics/ not on constants.
So, certain explanations for patterns, like the more labor supply you get, the lower its price, work well in an inflationary environment of labor demand/ but does not mean anything when the average is gone. It just cannot absorb more labor anyway, so it remains flat.

One rule of thumb is, because labor costs are contradictory to consumption income, the latter will always drive out labor to lower cost environments despite patent ruling after the economy grew/ and what remains is the currency value the other side will want to export to, but will never create new labor: it is in its consumption fase.
Without paying attention to the dynamic levels it reaches with evolutionary change, no economic theory will survive.


Economics has major problems as a serious domain of inquiry. Pierro Sraffa blew away marginal utility theory in the 1960’s, but that didn’t stop the economists:


If Sraffa is correct in this theoretical modeling that wages and profits are linearly related to each other and independent of pricing, it sheds light why there might be a political motive for neglecting his work.

I would note that Sraffa’s economics does seem consistent with Turchin’s neo-Malthusian approach to history. Elite enrichment occurs when profits grow and wages fall, and good times occur when profits shrink and wages rise, and this is independent of pricing, worker productivity or other factors.

j r

So I am on the side of Harvard Professor George Borjas, who’s careful lifelong work leads him to conclude: “The best empirical research that tries to examine what has actually happened in the US labor market aligns well with economic theory: An increase in the number of workers leads to lower wages.”

This is a pretty gross over-simplification of what economic theory says, which is odd, because you basically argue the same thing in that piece on Rodrik’s book. Economic theory says that the demand curve for normal goods slopes downward, which it probably does. But that describes a near-term condition (which economists generally define as the period in which no new factors can be brought into production). If one qualified person applies for a job opening, that person has greater leverage to negotiate compensation than if 20 people apply for that one job opening.

But as you say, social science isn’t one model. So, economic theory also says that once you get past the near-term, demand curves shift. And which way they shift is a function of a bunch of different variables. If a bunch of migrant workers show up in a place to work in a specific industry and those workers all live stacked in bunk beds and spend most of their money back home to family, the demand curve is going to shift in a different manner than if a bunch of workers show up to live, with their families and start consuming housing, food, entertainment, etc. If immigration increases either demand for more goods and services or the supply of goods and services (immigrants bringing new products or new innovations), then the demand curve shifts out and more labor is demanded.

Which one of these happened/happens in real life? That’s an empirical question, so we should probably look at the accumulated body of research, instead of just throwing out the economists you don’t like (because they can’t be trusted) and accepting the one who gives you the answer that fits with your theories.

Peter van den Engel

The only way to solve the problem is to recognize labor costs actually are nothing but consumption needs, which is a random given. Depending only on the environment and what particular given occurs in the situation, which explains itself.
A computer model; athough very helpful for analyzing; would only use the type of input it is fed with ofcourse, disregarding if its a true or a false equation.

The upward migration of money in the piramide is only due to its administration model itself. Which is also a random choice, basicly.
I think you should only use the statistics of sales to analyze supply and demand, which is not ones standard entropy related needs, which represents another constant.


AoD was a great book. Data choices, given the theory, could hardly be better.


I could not resist commenting. Very well written!
I could not resist commenting. Perfectly written! Way cool!
Some very valid points! I appreciate you writing this write-up plus the
rest of the site is really good. http://ford.com

David Vognar

Hi Peter,

Long time no read. Theory is indeed the best way forward. Induction can only get us so far. That being said, I remember your political violence prediction from 5 years ago. I think we need to start changing the way we talk with each other. I think poor people are really going to hurt with the next recession, especially with CPI on the rise and the $142 trillion in consumer debt. Credit cards are going to have to offer lower interest rates for social cohesion and stability purposes. They’re going to have to take a hair cut. If banks get bailed out again, there will be revolution.


Eh. Old people don’t do (violent bloody) revolutions:

We may have more chaos, dysfunction, mass shootings, but the worst we can expect is something akin to the relatively bloodless strife between Reds and Yellows in Thailand or the bloodless revolution of Tunisia.

David Vognar

I should clarify it is $142 trillion in over all debt. But consumers are driving the economy right now.

  1. Home
  2. /
  3. Cliodynamica
  4. /
  5. Regular Posts
  6. /
  7. Why Theory Is Important:...

© Peter Turchin 2023 All rights reserved

Privacy Policy