Geographic Influences on Social Evolution: Ecology and Empire


Join 36.6K other subscribers

A week ago I was interviewed by a BBC person who is working on a TV series about the role of geography in history. The questions were prompted (of course!) by Jared Diamond’s magisterial Guns, Germs, and Steel. I love the book, even though I disagree with many of Jared’s views. (There is no contradiction here, in science it is much more important to be interesting than right.)

Today I’d like to discuss one of Diamond’s ideas that I find very productive. In the chapter entitled “Spacious Skies and Tilted Axes” Diamond pointed out that ecological zones (the technical terms is ‘biomes’) tend to be stretched from east to west, because climates, soils, etc. are more similar as one travels east or west, compared to north/south. As a result, crops and domestic animals can spread more easily along the lines of latitude (in the East-West direction). The Eurasian landmass is oriented along an East-West axis; it is also huge. Different plants and animals, domesticated in different parts of Eurasia, spread East or West quite readily. The end result was that each particular region could profit not only from crops and animals domesticated within it, but also from many other species domestcated in distant, but ecologically similar regions. As an example, peach was domesticated in China but it spread to Europe already in the Antiquity. Earlier, cereals, such as wheat and rye, readily spread from the Fertile Crescent west into the Mediterranean Europe. But it took millennia longer to spread north into Russia, although the distance that needed to be traveled was actually shorter.

An argument can be made that it should be easier for many other things to spread along the East-West axis – not only cultivars, but also human genes, artifacts, ideas, and even political power. As an example of the latter, think of the Roman empire. The Roman state evolved within the Mediterranean biome. Once it expanded beyond Italy, it rapidly spread West and East to other regions with the same ecology, from Spain to the Levant. Pushing beyond the Mediterranean ecological zone, however, proved to be much harder. The Roman push into the forests of northern Europe ended in the disaster of the battle of Teutoburg Forest in 9 C.E., in which 20,000 legionaries led by Publius Quinctilius Varus were wiped out by an alliance of Germanic tribes. Being used to the Mediterranean shrubland/woodland habitat, the Romans had a lot of trouble with conducting military operations in northern forests, where “the trees grew close together and very high” (in the words of the Greek historian Cassius Dio). In the end, the Romans decided that conquering the Germans was more trouble than it’s worth.

In the east, the Roman expansion was stopped by the Syrian and Arabian deserts. The famous Roman defeat there was the battle of Carrhae (53 B.C.E), in which the invasion force of 35,000 legionaries plus supporting troops led by Marcus Licinius Crassus was virtually obliterated by Parthian cavalry.

Although deserts can be a serious barrier to people who are not used to them, for others, like the Arabs, they were essentially highways of expansion. After the Prophet Muhammad died, his successors expanded the Islamic Caliphate West into North Africa and East into Persia and Central Asia. Again, as in the Roman case, the shape of the resulting empire was stretched in the East-West direction. However, the biome at its core was not the Mediterranean, but the hot subtropical desert. Yet another example is the Mongol Empire, which stretched from Ukraine to Korea, with the Great Eurasian Steppe at its core.

I can multiply examples, but this wouldn’t be science. What needs to be done is to test the idea empirically – collecting data and doing statistics. In short, we need to do some cliodynamics. Incidentally, this is one of the criticisms that I have for Guns, Germs, and Steel – it is long on ideas, but short on testing them in a systematic fashion.

A few years ago we decided to do a proper scientific test of the hypothesis that extending state power should be easier East-West than North-South (this was research co-authored with Jonathan Adams and Tom Hall). We compiled a list of all large historical empires with territories exceeding a million square kilometers, and calculated the ‘latitudinal index,’ which measures the extent to which territories are stretched along the East-West axis. The details are here:

Our results indicated that the physical and biological environment had a very strong effect on the shapes of historic states. As you can see from the graph below, the great majority of historical empires have a positive Latitude Index (meaning they are stretched in the East-West direction).

The two empires with a strong north-south orientation were in areas where ecological zones were stretched in the North-South direction. The New Kingdom of Egypt had at its core the valley of a major river oriented North-South, the Nile. The Inca empire was located on the west coast of South America where ecological zones are stretched North-South due to the Andean mountain chain. In other words, these exceptions to the East-West rule conformed to the more fundamental rule that projection of military and political power was easier within the same ecological zone.

So Diamond’s argument is supported by data. More importantly, it turned to be a productive idea, because it prompted us to collect data to test empirically a particular extension of the argument. I know of at least one other article (by David Laitin and co-workers), currently under review, that applies Diamond’s hypothesis to the shapes of modern nations.

To reiterate what I said above, the best of science is not to be right (one can be right in obvious, boring ways), but to put forth productive ideas; ideas that lead to further research. In this Jared Diamond succeeded very well.

  1. Home
  2. /
  3. Cliodynamica
  4. /
  5. Regular Posts
  6. /
  7. Geographic Influences on Social...

© Peter Turchin 2023 All rights reserved

Privacy Policy

%d bloggers like this: