Social Evolution Forum
FIND sef:
Unproductive War II

In the previous post I argued that although it looks like the period between 1 and 1500 AD was one of stagnation, and even regress (European ‘Dark Ages’, etc.), under the surface there was a lot of technological and social progress. In China, again, agricultural techniques continued to evolve. And while the geographic extent of Chinese empires, measured in millions of square kilometers, did not increase much, if at all, the scale of the Chinese society measured in the number of people expanded quite dramatically. During the first millennium AD the population of Chinese empires (under Han, Sui, and Tang Dynasties) peaked at the level between 50 and 60 million. But in the early second millennium, under the Sung Dynasty, as a result of significant advances in agricultural technology the population reached the level that was double of the first millennium peaks, 120 million. And peak populations achieved by subsequent dynasties continued to increase even before 1500.

Similar advances took place on the other side of the Eurasian land mass. In the region where I was born (just southwest of Moscow) there was no agriculture to speak of during the first millennium AD. But in the second millennium, as a result of such innovations as the heavy plow and new cultivars (e.g., rye) agriculture spread to central Russia, and provided a resource base for a very substantial state.

Most importantly, military technology was not at a standstill. Cavalry improvements (stirrup, heavily armored cavalry, and sabers to cut through the armor) is one example. But the most far-reaching consequences were due to improvements in firearms, starting with Greek fire and Chinese incendiary weapons. In fact, the perfection of gunpowder weapons was what brought the era of the steppe archer dominance to its end.


At a deeper level, it’s wrong to think of the period between 1 and 1500 AD as unproductive, because it’s not really war in itself that is either productive or unproductive. The productive force, in terms of driving social evolution, is instead competition between groups and societies, which has the effect of weeding out dysfunctional groups. It’s just like firms competing in the marketplace. If you stop such economic competition, as happened in the Soviet Union, you will end up with a lot of dysfunctional firms. Same principle applies to the competition between social groups.

Violence is unproductive when it pitches people against other people within societies, taking forms such as murders between individual people or civil war between organized groups. It can be ‘productive,’ despite killing people and destroying property, when it is whole societies fighting other societies. But the key is not killing people, it’s between-group selection – competition that eliminates societies, whose members are unable to cooperate with each other, or to invent and adopt innovations and acquire or sustain prosocial norms and institutions. It’s not being good at killing.

Even in war one society can defeat another by outproducing the enemy, rather than by outfightning them. During World War II Germans were unparalleled as fighters on the battlefield. Quantitative studies by military historians have documented that, for example, Americans had to bring 40 percent more troops than the Germans just to get an even chance of winning. This is a little-known historical fact. What is better known is that Americans simply outproduced the Germans by churning out more aircraft, tanks, artillery shells, etc.

Another little-known fact is that this was also how the Soviets fought the Nazi Germany (which is more important, because the Third Reich lost over 80 percent of its casualties on the Eastern Front). Before 1944, when the battle-hardened Red Army finally matched the fighting prowess of the Germans, the Soviet Union had to rely on outproducing the Reich – I don’t have the numbers handy, but it was like producing two, three, or even four times as many tanks, planes, assault rifles, ammunition, etc.


These examples deal with war, but while historically competition between societies usually (but not always) took bloody forms, it does not logically need to be that way. This is, by the way, one reason why I think our future as the species is not completely dark.

So between-society competition is not just about violence, and primarily not about it. Conversely, high rates of violence, as those found in the Inuit (where 30 percent of adult male deaths were due to murders), or incessant between-village warfare in New Guinea and the Amazon (Yanomami!) have been completely unproductive for the evolution of large-scale cohesive and productive societies.

new guinea warfare

If you are interested in a more detailed, although technical argument, take a look at this article of mine.

In that article I identify several conditions that turn war into a potent force of social evolution. One of them is improvements (if this is the right word) in military technology. As I have written before in this blog, the rise of the warhorse was one of the most important such technological innovations before the Age of Gunpowder, And that is why we see tow jumps in the scale of largest empires in the figure above after the invention of, first, chariotry, and, a thousand years later, cavalry. And there was a third jump, after 1500, due to the invention of cannon and ocean-crossing ships.

So I disagree with Ian who sees the rise of steppe horse-riders as the development that turned productive war into unproductive violence. It was these horse-riders who drove the evolution to larger and wealthier world empires during the first millennium BC. The warhorse increased the selection intensity on societies, which brought the typical size of world empires to that range of around 3 million square kilometers around which it fluctuated for the next two thousand years. Then, the evolution of gunpowder weapons and sailing ships elevated the pitch of selection even higher, resulting in the rise of such huge transoceanic empires as the Spanish and the British.

We now live in a new age. The Spanish and the British empires, as well as the Soviet one are now gone. Only one is left, and it is in deep trouble. The European Union, which for a while looked like a new and different way of voluntarily integrating multinational conglomerates of people, is also in trouble. Does this mean that we have again run out of steam, just as in 1 AD, as Ian Morris suggested for an earlier period? Interesting questions for which I don’t have answers.


Join the discussion


  1. Martin Hewson says:

    Have we run out of steam? Military competition has subsided, so all else being equal that should allow for some political down-sizing. (One way this may occur is that more secure-feeling states will tend to allow more autonomy to minority nationalities, or even allow them independence.) But we are also in an era of intense international economic competition. The political up-scaling of the EU was largely driven by the desire to compete with the economic behemoth of the US, then Japan, now also China.

    A case can be made that war was productive (in a positive sense) not primarily when it led to an up-scaling of empires but when it led to the invention of smarter, better kinds of political cooperation like the nation-states of gunpowder-age Europe, or such experiments as the Swiss and American federations – all of course forged by external competition. Arguably, these forms of cooperation have yet to be bettered.

    • Peter Turchin says:

      The EU case is an interesting one. I think that the Cold War had a lot to do with the initial impetus at political integration. More recently, it became clear that ecnomic integration cannot work without political integration. Economy and polity are not really separable.

  2. In general I agree with this, certainly about those steppe nomads (vs. Morris), but it seems to me that a great deal of intersociety war is just random violence too, and the meaner and more vicious people win instead of the most productive or advanced. Also, even if it is the more advanced that wins, there may be an overall lose-lose sitiuation because said advanced society exterminates all the others, costing the world all their accumulated knowledge. It would take some case-by-case comparison to test all this.

    • Peter Turchin says:

      There is no contradition between ‘meanness’ and being more advanced. Americans of European extraction won over Native Americans by both being more advanced in technology etc, and by being ruthless when it came to exterminating Indians.

  3. Nathan Dean says:

    It would seem speed is an underlying concept here. The faster one produces, the faster one maneuvers, and the faster one wins. Conflicts and wars have certainly became shorter as time has progressed, from Roman multi-century wars, to medieval century wars, and now our decade wars. Society evolved where wars become quicker so they remain more productive than long, drawn-out, stagnated wars. Of course, there are many different facets to the cause but I believe it is a good marker to judge an evolved society and government.

    The possibilities if conflicts become even more quick would one: that weapons have advanced to the point where the opposing army is instantly vaporized, or two: that they do not even happen because the war does not last even milliseconds meaning that there really is no war. The second possibility being the more likely as an end result than the first as the second would encourage a more productive outcome without major infrastructure damage and genocide.

    It is not necessarily that we will run out of steam but that possibly violent conflicts will be solved in little time, allowing for continual competition but less death and destruction.

    • Peter Turchin says:

      Speed can be thought of as one of the underlying concepts. I would, though, reformulate it as ‘logistics’ – ability to project power across distance rapidly, effectively, and massively. But this is just one aspect of what makes some societies better in intersocietal competition. Other historians suggested that energy is the key (e.g., Leslie White). And I can think of other ‘proxies.’

  4. For a somewhat different slant on the relationship between social evolution and warfare, I’d like to bring up a view that John Arquilla and I have elaborated before: The history of military organization and doctrine is largely a history of the progressive development of four basic forms of engagement: the melee, massing, maneuver, and swarming. Briefly, warfare has evolved from chaotic melees in which every man fought on his own, to the design of massed but often rigidly shaped formations, and then to the adoption of maneuver. Now swarming is on the rise; it appears at times in past history, but its major advances as a doctrine will occur in the coming years.

    If this formulation looks helpful and interesting, go here to download our old Rand study (it’s free) on “Swarming and the Future of Conflict” at:
    link to
    Chapter Two (pp. 7-23) is about the evolution of military organization and doctrine — from melee, to massing, then maneuver, and lately swarming — with particular reference to the roles of information and information technology in the evolution of these four forms.

    What that write-up does not show, except in a passing footnote, is what I’d stress here: Our formulation derives from a view of social evolution — one I call TIMN for brevity — which holds that, across the ages, societies have progressively given rise to four cardinal forms of organization: tribes, hierarchical institutions, markets, and networks. Thus, early tribes are associated with melees, hierarchical institutions with the rise of massed formations, then the rise of market-oriented societies with the turn to maneuver doctrines, and now the age of networks with swarming. This progression in organization and doctrine appears in all realms of war: land, sea, and air. The progression also appears in the evolution of insurgencies and social protest movements. In other words, the relationship between social evolution and warfare is about who can out-organize.

  5. Peter Turchin says:

    David, thank you for this link. I downloaded the publication and look forward to reading it. I am a bit sceptical about rigid classificatory schemes (tribes, etc), but let me read first before I criticize! Approaching military organization from the eovlutionary point of view is certainly very interesting and productive.

  6. if you want to take a quick look at TIMN in its current state of formulation, there’s a conference video talk here:

    link to

    a briefing-like text overview is here:

    link to

    in my view, it’s categories are more flexible than rigid (or should be, if I spell it out right).

    • oops, sorry, i never meant to post the video, just the url to it. feel free to edit the video out while keeping the url pointer, if possible.

      • Peter Turchin says:

        No problem – I personally prefer to read text, rather than watch videos (I read 3-4 times faster than a person can speak), but others may want to listen to the talk.