Evidence for the Continental Axis Hypothesis

One of the most striking hypotheses in Jared Diamond’s Guns, Germs, and Steel was that technology diffused more easily along lines of latitude than along lines of longitude because climate changed more rapidly along lines of longitude making it more difficult for both humans and technologies to adapt. Thus, a long East-West axis, such as that found in Eurasia, meant a bigger “market” for technology and thus greater development.

A few pieces of evidence are suggestive:

Laitin and Robinson (2011) and Laitin et al. (2012) report that linguistic diversity has been historically more persistent across lines of latitude than longitude, suggesting that population movements were more prevalent East-West relative to North-South. Ramachandran and Rosenberg (2011) report similar evidence based on the geographic distribution of genetic variation. While these studies speak to greater movements of populations East-West relative to North-South, they do not speak directly to the diffusion of technologies and development. Alternatively, Olsson and Hibbs (2005) provide a cross-country analysis in which a variable measuring East-West orientation of major landmasses correlates significantly with present-day income levels. This finding explicitly links continental orientation to income levels. However, it does not speak directly to the  mechanisms (e.g., more diffusion of technologies) leading to this correlation.

In Did Technology Transfer More Rapidly East-West than North-South?, from which I just quoted, Pavlik and Young offer more direct evidence on the natural direction of technological diffusion:

We employ Comin et al.’s (2010) data on ancient and early modern levels of technology adoption in a spatial econometric analysis. Historical levels of technology adoption in a (present-day) country are related to its lagged level as well as those of its neighbors. We allow the spatial effects to differ depending on whether they diffuse East-West or North-South. Consistent with the continental orientation hypothesis, East-West spatial effects are generally positive and stronger than those running North-South.

Very cool!


I'll bet when the primordial tribes set out on adventures, it was in summertime when there was more food. And it was in the direction of a rising or setting sun-god.

The so-called climate change if there was any is so small that it cannot be detected. We depend upon AGW activist and NASA revisionists to tell us there is actually any climate change. So it is quite unlikely that this imaginary phenomena caused any political or social problems except of course some screaming socialist who hope to exploit it to steal your money.

And yet, detected it we have: https://reason.com/archives/2006/09/22/confessions-of-an-alleged-exxo

They find what they must to continue the extortion. In the 50's they were finding global cooling. In the 80's it was global warming. After the turn of the century nothing was happening so they changed the name of the scare campaign to climate change so they could point to any weather and declare it to prove their wild ass claims. But yet it hasn't shown up so they need to rewrite the historical temperature records and current reading to continue to scare the ill informed.

Putting aside the causes, and whether they are man-made, are you saying there aren't noticeable changes in the planet's climate over the last couple of centuries? Ice coverage at the poles, ocean temperatures, etc?

They don't mean "climate change" as used in the current sense but changing climates from more extreme latitudes. Moving east west tends to keep climate stable, I.e. consistent with cultural practices. Crossing latitudes means broader changes in weather that challenge those practices.

Much of the US, for example, is similar to Europe in terms of agriculture, length of winters and summers, rainfall, sunshine, winds, etc.

North America was a perfect match for Britain and the Netherlands, South and Central America for Spain and Portugal.

Almost. The winds in Great Britain and Western Europe come from the Atlantic and are moderated by the Gulf Stream.

In Eastern North America the winds also come from the west, but pass over the North American continent. As a result summers are hotter and winters are colder in Eastern North America than they are in the British Isles or Western Europe.

IIRC, Diamond discussed the ease of movement of plants and animals on the East-West vs North-South axes.

Diamond noted that jungles along the equator severely restricted north to south travel, whereas east west trade routes were prosperous. The Silk Road was an example.

Hence, "Go West, young man," and not "Go North."

Except for North to Alaska, go north, the rush is on.

I'm looking at a lonely, lonely man.

I'd trade all the gold that's buried in this land,
For one small band of gold to place on sweet little Jenny's hand.

Because I'm tired of digging for nuggets with you.

Cause a man needs a women

like a fish needs a bicycle.

Speak for yourself soy boy!

Seems like an impossibly small sample size from which to draw a conclusion. One continent out of five was different, and hundreds of differences available to explain why it was different. Does the hypothesis have any use at predicting the future? Probably not. So I dont care, just as I've never cared about Diamond's kind of analysis attempting to explain huge civilizational differences via anti-intellectual causation. Does it make various tribal groups feel better to think they might have built skyscrapers too if only they had been lucky enough to be in the right place and caught the right disease at the right time? Or that the reason other people are so much more advanced than them is an accident rather than a continual process of accumulating better human capital via chosen effort?

It actually does has use in predicting the future. Now that the spread of technology no longer depends on people walking. One would expect under developed reagions to catch up. Also if the earth continues to warm, the ability to move north south will be more important.

I think most IQ heads are apologizing for their own middle rank.


IQ doesn’t explain literally everything, therefore it’s useless and fake! Mmmmk.

Let’s take a look at this article you linked but didn’t read:

“Together all these variables predict 48 percent of the variation in income between respondents. Adding AFQT scores bumps that number up to 50 percent. In other words, after accounting for the usual demographic suspects, AFQT only accounts for 2 percent of the variation in income. The unique contribution of native intelligence appears to be small.”

He adds a bunch of variables that correlate extremely highly with IQ and then adds IQ. Lol what the hell, no economist would make such an elementary econometrics error?

Nicholas Wolfinger - Professor of Sociology at University of Utah, author of “Do Babies Matter” and “Soul Mates.”

Oh. Not surprised.

Carry on bear.

Its particularly dumb because Garrett Jones (about the only serious IQ fueled economist out there at the moment), tend to think that IQ tends to explain between society differences ("Hive Mind") and is pretty open that they are of minimal use for predicting individual income, on their own.

(There's also a contention that they explain between ethnic group differences, which implies that none of the other intra-individual differences that diminish IQs value in a within ethnic sample, really are ethnic differences.)

There are some people that don't like the fact that IQ doesn't explain that much of intra-individual differences, but those people are almost a strawman set against the more serious proposition Jones presents, that keeping average IQ high matters for a society, while your own individual IQ doesn't really have *that* much to do with your success.

Tip: when I say "most IQ heads are apologizing for their own middle rank" I am setting it up for someone to be that guy.

Don't be that guy.

And 2% is pretty low, handwaving aside.

Garett Jones will be Garett Jones even if you suggest he shouldn't be. And that's definitely an "Everest regression" in that paper.

Since I linked to Jones agreeing with Tabarrok on the folly of measuring an IQ correlation after "controlling" for deaths IQ causally affects, it only seems fair to link to some more conciliatory comments on the same paper:

I have only vague memories of the "countries" work, but if I recall correctly, it relied heavily on proxy data, given the lack of truly equivalent time series, especially in Africa.

The nice thing about Wolfinger's post is that it uses high quality data for a large group, straight up, no proxies

I am surprised that Alex makes the same mistake as our friend Hmmm.

"Good but it's odd to write a post challenging IQ as a magic variable and then say it doesn't add much when you add marital status, education, number of children, employment status, employment history, hours worked, age...when the arg. is that all of these are influenced by IQ."

Do you see the error?

You can't just name a bunch of correlates and then just pluck out one and declare it the causation.

Wolfinger is not doing that. He's being agnostic on the correlations and how they might be interrelated.

You call it an "error", but that does not make it so. We measure IQ at one point in time and see that it predicts things at a later time (like the ones listed). Things that happen in the future don't cause the past. We also measure things like GWAS alleles & skull size that don't change much over certain periods of time. Additionally: if you find a large correlation with variable X and then find that it's greatly dimished when controlling for Y, then perhaps X wasn't that important on its own and the importance of Y was overlooked. But if you have to control for A, B, C, D, E, F, G, ... in order to bring down the correlation, that's less meaningful.

Nicholas Wolfinger updated his post, specifically thanking Lee Jussim, Robert VerBruggen, Alex Tabbarok & Garett Jones for their comments:

I mean, if you were going to casually knock around ideas, you might wonder whether "stick-to-it-ness" is upstream of a few of those others, including IQ test taking.


From his description he seemed to have only naively look at the increment in Rsq. Why he did not show the output from the stats analysis?

"Together all these variables predict 48 percent of the variation in income between respondents. Adding AFQT scores bumps that number up to 50 percent. In other words, after accounting for the usual demographic suspects, AFQT only accounts for 2 percent of the variation in income. The unique contribution of native intelligence appears to be small."

What he has to look at are change in the the significant level (p-values) for the coeffs of all the independent variables, was the coeff of AFQT significant and has the addition of AFQT dumped down the significant of other independent variables or made them statistically insignificant because of correlation among them.

What else do you expect from a sociologist. And the suckers that believe him.

I think I understand and agree with that, but as I say, there might be some assuming about which are independent variables. See below on 1200 genetic variations associated with staying in school. These are early days.

Assuming that east-west is just indicator combined approximate index for similar climate and vegetation zones, you could find even more correlation by using them directly.

European immigration into into the Americas seems to follow this pattern. People from Scandinavia and Finland settled in large numbers in Canada, Minnesota and Oregon regions where the climate and vegetation was familiar to them.

It would have not been that hard for Portuguese or Spaniards to sail north and build settlements there, but they choose climate, temperature and vegetation that was closest to their origin countries, then moved south.

No. The Spaniards and the Portuguese were looking for gold and glory. Their cultures were inferior and their motives evil, and everything they touched was corrupted. It's not an accident that the USA is the richest nation in history and the countries that comprise the Anglosphere - USA, Great Britain, Canada, Australia, and New Zealand - are the richest, healthiest, and freeist countries in history.

The Swiss are likely too polite to point out that they too live in one of the richest, healthiest, and freeist countries in history.

Switzerland proves that efficient government is the most important component of a prosperous people. The Germans in Switzerland are richer than the Germans in Germany, the French in Switzerland are richer than the French in France, and the Italians in Switzerland are richer than the Italians in northern Italy. Smaller countries are generally less corrupt than big countries where groups fight each other for political spoils.

Switzerland is an interesting example of the benefits of a small, stable country with established communities, placed aside from being a nation state or not, or being multinational. Sort of an opposite of the EU, where the normal slurs of "insularity" and "mono-ethnic nationalism" lobbed at European non-EU members can't diminish it.

The Swiss built their economy on international banking with a reputation for anonymity. If you hide your assets from the tax man, of course, you will be richer. Ironically, the Swiss are also proof that wealth taxes on their own citizens can work.

Switzerland is in the minor leagues.

Nice cheese though!

new zealand is in the big leagues?

Er... You think Mexico's climate is like Spain?

Presumably this is why Anglos never settled in New Mexico... All those TVs shows set in Texas, Arizona, New Mexico and the West must've been a fever dream. No Brits in Australia either.

People do seem to migrate due west in the US. I live in Oregon and many people here are descended from upper midwesterners who came here to build ships and planes in WWII. When I lived in the southwest I ran into many people from the southeast. There seems to be a resistance to moving north — air conditioning effect?

Visit Florida sometime. Much of the non-foreign population migrated down from the northeast of the midwest.

“than a continual process of accumulating better human capital via chosen effort”

I take it that you are referring to eugenics. If you are right, imagine the scale of it. Therefore I’m doubtful.

Your take is totally wrong. Human capital is knowledge and values, culture, not genes. Think of symphonies vs tribal chants, and then extrapolate to all fields of human activities.

Well that I might agree with, though I do think culture is quite malleable, but based on a very uniform human nature.


The great trick, going from squabbling tribes to prosperous (trading) nations, is to expand that "group loyalty" as far as you can.

"The great trick, going from squabbling tribes to prosperous (trading) nations, is to expand that "group loyalty" as far as you can."

True. That expanded group must have some shared value(s).

Hamburgers, baseball, and fast cars.

How about the Constitution, Bill of Rights, Declaration of Independence, the Federalist Papers, and an official language? And all those above taught in schools with an expectation students will learn them.

Yes those too!

The term "human capital" was coined in analogy to capital equipment that is purchased/created as an investment for producing an economic output. Education is commonly given as an example of this sort of investment in human capital. But Bryan Caplan's "The Case of Education" argues that there's relatively little human capital formation going on in education, and that the higher incomes of the educated are due more to signalling & selection effects. Recent genetic studies have shown that years of education is significantly correlated with a number of genetic variants, including within families. So to the extent that's still considered "human capital", God of Thunder may have a point.

"In one of the largest genetic studies ever done, an international consortium of scientists has found more than 1,200 genetic variants associated with educational attainment, which is defined as the number of years individuals spent in school or university."

Ability to sit in one place, stop fidgeting, and act like you are paying attention genes.

You might be intending to be funny, but there is probably something to that. Educational attainment is correlated with IQ but not perfectly so, and it doesn't seem to just be statistical noise but other personality traits that play a role.

Sample size: 3 continents

Confidence level: 99.99% because confident narratives sell books

If Newton had insisted on such high standards he would have died of apple related brain hemorrhage before being certain about the theory of gravity.

Hah, unfortunately, they're not making any more...

You could consider a fourth example of the Southern language and accent extending from Texas to South Carolina.

I love his books - especially GGS and "Collapse ...".

"since it happened that way, it had to have happened that way."

The big problem I have with this hypothesis is that it suggests that trans-Eurasian contacts should be why Eurasia pulls ahead of the Americas.

But that's not the case.

Although exchange between East and West Eurasia does help Eurasia pull home its advantage, agriculture and state history are just earlier in West Eurasia well before there's any exchange.

1. https://pseudoerasmus.files.wordpress.com/2015/09/op.jpg?w=640 - older emergence of agriculture and state in West Eurasia
2. https://pbs.twimg.com/media/DnjIEpyXsAEFTIx.jpg - correlation between emergence of agriculture and state across world

For the most part, until about 500 BCE, new technology is developed by societies within the West Eurasian zone (mainly the Greater Middle East, and to a lesser extent the Indus Valley, the steppes and SE Europe) and to a great extent, exchange is basically one way from West to East Eurasia.

West Eurasia is already well ahead of the Americas, well before any importance of the long axis is possible. Eurasia didn't *need* a long axis, when it already had a solid first mover advantage.

I think the elephant in this particular room is that trade developed first and foremost along shorelines, rivers, inland seas (eg. the Mediterranean), etc. I would guess that overland freight volumes were small compared to maritime.

I suspect you are right that ships were more efficient for volume. But I also think that land piracy was far more dangerous than see piracy at the time.

Diamond has good biogeographic arguments for his hypothesis - the relative continuity of climate, flora and fauna along lines of latitude contrasted with the opposite along lines of longitude.

Humans, for example didn't move out of Africa until Eurasia started warming and its biogeography became more Africa like.

Isn't altitude a consideration at least as much as latitude?

Yes - there are large areas of Africa that are as temperate as Europe thanks to altitude. But these never developed technological civilisations or imported technology from Europe. The British tried hard to introduce such technology during colonial days, but when they left it was generally abandoned. And as other people mention above when the continent was colonised by Western Europeans (like Australia) there was no problem in having an advanced technical civilisation. So geography clearly doesn't equal destiny, except in the sense of a genetic winnowing ground. I think certain civilisations have be bred to work in a capitalist society - it is not necessarily IQ, it is other things as well like work ethic. The western pride in being such is probably a little misplaced - its like someone being proud of being a good slave. Someone from outside looking at the way 50 year old men with plenty of assets are busting their assess to keep working would think they are pretty foolish.

In the movie "Black Panther," Wakanda is at such a high elevation that one province is covered in snow.

Director Ryan Coogler was likely inspired by Lesotho in Southern Africa, which has a ski hill. Coogler visited Lesotho, a constitutional monarchy, shortly before making Black Panther. I think Coogler theorizes that given enough time, Africans in wintry Lesotho would have developed a more technological culture.

I don't know if that's true, but it's pretty a fun and interesting idea for a comic book movie.

Good comment. Geography is not destiny but it sure shapes the edges of it. I would argue that a capitalist society is not necessary for a technological one as the USSR and modern China has shown. Also, North Korea had to re-invent much of the engineering of nuclear weapons and rocket science since no one will just give them the knowledge on how to do those things. Even a theocracy like Iran is in the situation to do the same.

The African highlands were mainly inland which cut them off from sea transport. Ethiopia was something of an exception since it was a relatively short journey from Axum down to sea ports in modern Eritrea which were plugged into the vast maritime trading network that stretched from the western Mediterranean to the seacoasts of China

Crops take longer to transfer north-south than east-west due to changes in the length of the growing season. For example, corn, the most important pre-Columbian North American crop for allowing a densely populated civilization, only arrived at Plymouth Rock a century or less before the Pilgrims arrived. Given more time, Amerindians in North America would likely have developed the dense corn-based civilizations of Amerindians in Meso America.

A twist in the tale - https://phys.org/news/2018-12-scientists-overhaul-corn-domestication-story.html

It seems that teosinte (proto-corn) had spread all the way into the Southwest Amazon from Mexico, and then was separately domesticated into maize by both separate Native American groups in Central America and the Amazon!

That suggests the reason why corn took so long to get into North America was because the plant wasn't fully domesticated until much more recently than believed... And the Native North American groups weren't so keen on getting in on the early domestication process.

And note, from Mexico the Southwest Amazon is pretty fucking far (much further than to North America). In distance, it's certainly on a part with wheat and barley from the Fertile Crescent to North China. (What's that about a North-South axis making spread of crops easier again, Jared?)

While I think Diamond's theory is pretty good, let me point out that Africa in the Sahel is 4,500 miles east to west, while North America and South America are quite wide too.

Also, it's unrealistic to point at the immense width of Eurasia as an obvious boon.

The mountains of Central Asia are an immense barrier to ground transport (e.g., that's why the Chinese Central Asian infrastructure project is so expensive). You can go around the Himalayas and other giant ranges on the Eurasian steppe to their north, but that vast area can mostly only support nomads, who tended to be more destructive of civilization than destructive.

I'm surprised you didn't mention Michael H. Hart's "Understanding Human History", which directly addresses Diamond's thesis, comparing the Americas to Africa.

Here's my 2007 review of Hart's "Understanding Human History:"


Diamond's celebrated factors are reasonably plausible for explaining the spectacular Spanish conquests of the Aztecs and Incas in the 16th century. But they fail to shed much light on famous subjugations within the Old World, such as the various invasions of India and China or the 19th century European imperialization of Africa. Refuting Diamond, Hart points out that sub-Saharan Africans, being part of the Old World, were more privileged than New World Indians in terms of the factors that Diamond emphasizes.

In contrast to Mesoamerican Indians, Sub-Saharan Africans had more disease-resistance than Europeans (for example, they had genetic adaptations for surviving malaria). Plus they could make iron; possessed domestic cattle, sheep, and goats; had been exposed to literacy on their northern edge in places like Timbuktu; and possessed a continent that is 4,500 miles wide from Senegal to Somalia—not that much narrower than Eurasia's 6,200 miles.

And yet, Africans didn't build anything close to comparable to the hidden city of Machu Picchu (Incan) or the pyramids of Chichen Itza (Mayan) and Teotihuacán (Central Mexican). ...

One important fact that Hart has ascertained:

"Throughout history, most of the instances of people from one region attacking and conquering substantial portions of another region have involved 'northerners' invading more southerly lands."

(The biggest exception: the Arabs of the 7th Century A.D. And the Romans conquered in all directions.)

This overall pattern of north conquering south has long been apparent from the historical record—even though northern lands are generally less populous, due to shorter growing seasons. ...

Hart offers a simple, deliberately reductionist model for explaining this (and much else): Foresight is needed to survive cold winters. So harsher, more northerly climates select for higher average intelligence. And intelligence is useful in war. ...

Enough about conquest. What about contributions?

The most productive centers of cultural innovation have tended to move north over the millennia, for example, from the Fertile Crescent to Ancient Greece to Renaissance Northern Italy to Enlightenment Northern Europe. Hart attributes this to agriculture tending to arise first in low-to-medium latitude locations with long growing seasons then spreading northward. In hunter-gatherer economies, every man must hunt. But in farming economies, enough food can be produced to support urban sophisticates.

It's cold.

There's a desert.

Sailer: "(The biggest exception: the Arabs of the 7th Century A.D. And the Romans conquered in all directions.)"

With Europe and North Eurasia too many more examples to be found. England vs Scotland and Ireland, Swedes conquering Finns, Russia expanding into Siberia, etc. Mainly happened when southerners needed to extinguish the north as a military threat though, or when there were opportunities for logging, fisheries and fur hunting, or to spread the faith. Not many targets to extract wealth from.

Idiots. We all know that climate change did not start until the late 1800s with the Industrial Revolution, and that we all have 12 years to live.

G Cochran wrote a great review of GG&S at his Westhunt site:

worth the read. tldr: "this book isn’t serious. The thesis is a joke, and most of the supporting arguments are forced ( i.e. wrong). Perhaps the most important thing we can learn from Guns, Germs, and Steel is that most people are suckers, eager to sign on to ridiculous theories as long as they have the right political implications."

but do read the whole review, very informative.

Comments for this post are closed