Olson in Context
Putting Olson's paper in conversation with some other hits of early state formation literature.
My last post was a simple celebration of Mancur Olson’s state formation theory, so in this post I’m going to put it in a bit of context and try to respond to some of the questions people have asked me about it.
Was Olson’s Idea Original?
In a word, no.
The core of Olson’s theory is fully anticipated by the German sociologist Franz Oppenheimer in his 1908 book The State. The State is an absurdly confident bit of turn-of-the-century social science. Much of it doesn’t hold up at all, but the opening chapters on early state formation are fun and thought provoking and they do indeed give us something pretty close to Olson’s theory.
Oppenheimer thinks that the prehistoric world consisted of three groups of people: hunter gatherers, nomadic herdsmen, and fixed field cultivators. For reasons not entirely clear to me, he thinks that nomadic herdsmen would most easily develop wealth differentials and thus status hierarchies. He also thinks that they would tend to get good at warfare because they could build large groups with mobile food sources.
Nomadic herdsman were thus, in Oppenheimer’s view, naturally driven to become roving bandits. When they would go out raiding they would have limited success fighting hunter gatherers, because huntsmen didn’t have much to tie them down and could easily run away into hills and forests. Fixed field agriculturalists by contrast had to defend their land. were basically sitting ducks and nomads could easily come and wipe them out and take all their food stores.
Oppenheimer then describes the same mechanisms of state formation as Olson, without the language of public choice economics. He says that after some amount of raiding, nomadic groups figure out that they can make more money by subjugating the fixed-fielders rather than merely pillaging them. They set up institutions and rituals to regularize their theft as taxation and they call themselves kings. He adds some confusing detailed steps whereby the purported social hierarchies within nomadic tribes make this easier, but the basic model is the same as Olson’s.
Does the model require roving pillagers?
After reading my last post another friend asked me whether I thought the distinction between the fixed location of the stationary bandit and the mobility of the roving bandit was significant. Do roving bandits have to literally be people on the move? Oppenheimer certainly thought so, but I don’t think Olson’s theory turns on that, and it turns out the archaeological record doesn’t support it. Anthropologist Robert Carnerio’s influential 1970 paper “A Theory of the Origin of the State” is a useful contrast here.
Carneiro says that states arise where there are geographically circumscribed areas that provide greater than subsistence potential for a large number of small villages and clans. In such a situation, little villages and clans fight with each other for greater portions of land, and the ones who end up good at fighting ending up subjugating the people around them. Because the area is geographically circumscribed, the subjugated people don’t have an easy option to run away. They have to stick around and accept their tribute status.
This view is fairly easy to graft on to Olson. Basically, you just need to postulate that in the earliest state formation contexts, the “roving bandits” were probably just guys from the next village over. These roving bandits weren’t Oppenheimer’s nomads. Instead they were other fixed fielders who had developed the tools and inclinations to go pillage their neighbors. It still makes sense that these people will eventually realize that enslaving their neighbors can make them more money than merely robbing and killing them would.
The way that Carneiro describes this dynamic fits the roving/stationary bandit model well enough, but he also highlights what international relationship scholars call the “security dilemma.” The settlers of a circumscribed productive area know that their neighbors might one day come and kill them for their stuff, so they might reasonably respond by developing tools for violence. The reciprocal impressions that your neighbors are preparing for violence and could be willing to murder you can lead to a spiral where conflict becomes basically inevitable. Once a certain amount of fighting between villages gets going, there could be a lot of selective pressure for the best fighters to win out and end up subjugating the whole territory.
How this all lines up with Olson’s story about the under-provision of public defense is unclear. Perhaps the views fit comfortably. Perhaps the security dilemma really becomes a problem once settlements become big enough that they’re hard to defend and there are big and poorly defended prizes for people to go out and raid. Of course, this could happen differently in different contexts, so there does not need to be a universal narrative.
How well does this line up with the archaeological evidence?
I can’t answer that question generally. However, I can compare Olson to the one recent famous popularization of early state archaeology that I’ve read: James Scott’s Against the Grain.
States came long after Sedentary agriculture
Scott wants to dispel any lingering myths that states arose through cooperative agreements that people needed to put together to make agriculture work. He points out that Mesopotamia was a lot less arid between ten and five thousand years ago and that agricultural settlements did not require any great irrigation works. But more importantly, we have no evidence of anything like an early state for the first few thousand years of agricultural settlements in the area.
Scott highlights that these settlements grew in population faster than the mobile groups of mixed foragers, herdsman and cultivators around them. However, the fertility edge was slight. Fixed field cultivators were exposed to epidemic disease, perhaps for the first time in human history, and paid a horrible price in mortality for it. The fixed fielders also probably were not taking in significantly more calories, perhaps less per unit of work even. What seems to have given them the fertility edge was merely that they didn’t practice abortion and infanticide to nearly the level that mobile tribes did.
This seems to line up with Olson’s claims. Agriculture led to the state only insofar as it it made large settled population centers possible. There’s nothing too surprising about that taking a few thousand years.
Early States Were Fragile Extraction Machines
Scott’s big thesis in the book is that the most developed capacities of early states were to requisition grain and enslave people. One of his favorite theories is that grain is so much easier for government agents to see and collect that states emerge more easily around grain growers and prefer to force people to grow grain when they can. He loves contrasting grain farmers to foragers and herdsmen who evade state structures much more readily (here he echoes Oppenheimer, but sadly like Olson, he doesn’t cite him!).
Scott wants to hammer home the fact that early states were quite fragile though. They seem to often have been overwhelmed by violence from within or without and they often succumbed to epidemic disease or environmental collapse. He generally thinks that life outside of states was much better and that more people have lived outside of state control over the last few millennia than most of us ever realize because all our history is written from a statist perspective.
This might seem like it is in tension with Olson’s claim that state formation would generally be an improvement for vulnerable agricultural settlements. Olson makes much of the fact that you’d rather live under a stationary bandit than free in a world of roving bandits. If states are so bad and people avoided living in them if they could, then maybe Olson has something wrong.
But here I think we see some of the limits of Scott’s brand of myth inversion. He gets himself very worked up about how great it was to live outside of state structures, but when he describes the “Golden Age of the Barbarian” as he calls it, it turns that most of what he thinks was so great is that there were a couple thousand years where roving groups of warrior herdsmen could raid the fringes of agricultural states. He traces this from Germanic raiders around the Roman periphery up to mid 19th century tribes like the Comanche on the western edge of the United States. He loves quoting lines from famous pillagers saying things like “raiding is our agriculture.”
It thus seems that all Scott is really saying here is that he doesn’t respect stationary bandits, and that he would rather be a roving bandit than a peasant. But this seems more like a matter of taste, and not actually a disagreement with Olson. There’s nothing in Scott that implies that conditional on being a fixed field-cultivator, it’s bad to live in a state.