This is part 1 of my Civilization series.
Rise and fall
Successful civilisations throughout history have grown rapidly, appearing unstoppable. Some (like Rome, and Han China) existed stably for hundreds of years. Others, like Alexander the Great‘s Asian Empire did not. But in every kingdom, empire and state that has ever existed (excluding the few surviving today, which are probably not as eternal as we’d like to think), something has eventually changed. Once great empires are defeated by a smaller rival (e.g. Sassanid Persia), and others crumble all by themself. Changing fortunes are one of the clearest signals in history.
Yet we have no mechanistic understanding for the collapse phenomenon. It is entirely absent from “Vanilla” Civilization, which assumes that growth gives power and power allows further growth, with the result that empires grow at an “exponential” rate. This is a good first approximation: over time, population has grown exponentially as more people plant more food which feeds more people. The same applies to technology, and social development. Yet these trends clearly do not apply to states and empires, even though it is often within the largest states that population density, technology and social development are highest. So why do empires fall?
There are many mechanisms in history that may cause the strong to become weak. Perhaps their strength was illusory. Since success at war pays well in the short term, winners of scuffles can hire more soldiers and thus will get greater military strength. But if the infrastructure for effective taxation, trade and public works to support the empire is not in place, when the expansionist wars stop (as they must) the empire will be overstretched. Warlords from within or without the empire can take over distant provinces and the empire collapses. Conversely, smaller civilisations are more efficient and competitive per person, people are more involved in the state, greater mobilisation is possible and they have unity of purpose.
Yet these stories are just that – stories. When does the first apply, leading to large states conquering small, and when do the small states gain an upper hand?
Modifications to Civilization such as “Revolution” and “Rhys and Fall of civilisation” both make a good attempt at adding this dynamic. The basic idea is that civilizations that are large (relative to what their infrastructure and governmental system can handle) will experience both inefficiency and stability problems. This is basically correct – but is added as a fudge, entirely without a mechanism. Some large empires (China for example) have managed to avoid internal strife (or at least, it hasn’t lasted all that long). Yet other, similarly sized empires collapsed almost without trace. What is the difference?
Peter Turchin has some ideas – he says that societies have “asabiya” – a unity of purpose that allows for collective action. Groups of people that live together in strife (such as desert nomads) – and particularly those on cultural divides – develop strong bonds and are able to commit very strongly to struggles. This allows these small groups to conquer much larger states, because the groups with such states defect and bicker and are not willing to die for their cause.
Once in power, the asibiya of such a group declines as the need for collective action decreases, and the lure of luxuries increases. The powerful state weakens to be eventually taken over by a smaller, more dedicated group. See his book “Historical Dynamics: Why States Rise and Fall” and his later books and his webpage for ideas, which stem from the works of the 15th century Arabic scholar Ibn Khaldun.
There is clearly something to this theory, but it is incomplete as it stands. Did 18th century England have higher asabiya than 18th century Venice (or Germany for that matter)? It is impossible to tell. Clearly, England was on the rise in a way that Venice wasn’t, but we can’t measure asabiya. The small nomadic groups described by asabiya are not in play here; there are more standard, more economic and sociological forces. What are they?
Essentially, we don’t know yet. I’ve been involved in some modelling, where we show that treating a state as a number of semi-independent groups can lead to this dynamic. Each group is coerced into the state, but gains benefits from being a part of it. Whilst everyone is on board, everything is fine – culture and development increase, and great achievements are made by the elite. But the groups at the bottom of the pile get a worse and worse deal – because they have less political power, and political power tends to lead to more political power. When one group rebels, the benefit to the rest is reduced, and it can set off a chain reaction leading to state collapse.
The model allows groups to act in an entirely selfish way – there is no fluffy concept of asabiya, but asabiya arises as a natural result of within-group competition. It is nice for several reasons. Empires that look strongest from the outside – those with high development and culture – are exactly the least stable. Collapse can be triggered internally, or by external stresses such as war. These problems can all be overcome by continual growth of the state, so that everyone is getting a better deal… but no growth is forever.
Unfortunately, the only reference for this is a Masters Student thesis. But watch this space…
Relation to civilization
The implication for the modelling of artificial societies is that group dynamics matter. Giving people a shared identity leads to greater capacity for action – but this shared identity is eroded by distance in culture, status, religion, nationality, money and power. Each new realisation of a state can form its identity on new grounds – by language, or religion, or race, or ideology, shared suffering or shared experience. This is why modern states have had to be more inclusive, to create a shared identity amongst people who in other circumstances would see themselves as very different.
This implies that there is a natural lifecycle to states. It is not necessary for a state to collapse and be rebuilt, but the people must find a new way of feeling unified. The natural dynamic in a game is to have groups with identities that are shared to a greater or lesser extent, and that have different desires based on these. Weaker groups will be less able to bargain for their desires and therefore will want to rebel… leaders have the option of silencing them, but at the cost of alienating further groups. It is not a long term strategy to kill your own citizens, and eventually, the state must be restructured, perhaps smaller, to create a new shared identity and (most importantly) a more even distribution of power. The greater the change, the more dramatic the collapse will be. Making demands of citizens – such as drafting for war – or not protecting them adequately will result in rebellion of some groups, leading to the rebellion of more.
It is important to make this a dynamic that is visible, rather than spawning random rebellions that just happen because some counter got too high. The groups within the empire should be visible, broken down by town, and each one should be unique in some way. Groups can be absorbed when their needs coincide strongly, which will be most common during rebellions or redefining of statehood. They cause trouble only when their needs are not met – be they financial, political, or religious. They can be eliminated by force – at the cost of scaring other groups – or be brought in line through education or incentives. Left alone, strong groups will get stronger and weak ones weaker, at a rate that depends on the degree of accountability in the society. Therefore different types of government will have a different ability to cope with differing needs.
Differences between groups are defined in terms of differences within the state versus differences without. If there is a clear enemy, people will remain united. Diasporas will lead to shared demands over multiple states, but also lead to shared bonds and greater trade of both goods, knowledge and cultural values. Multiple cultures within a state lead to less focus but greater rates of innovation and cultural growth.
Adding this one feature will in one swipe unite the process of cultural diffusion (to be dealt with later) with the internal process that causes collapse. “All” that remains to be done is to implement the mechanic in a balanced way that is interesting to engage with and produces realistic effects…
Essential reading for this section
*P. Turchin: “Historical Dynamics :Why states rise and fall” (2003) (A must read!)
P. Turchin: “War and peace and war” (2005)
Kandler and Levand: An investigation of the relationship between innovation and cultural diversity (2009)
Kogel and Prskawetz: “Agricultural productivity growth and escape from the Malthusian trap” (2000)
Kohler: “Simulating ancient societies” (2005)
Neeraj Oak (and Daniel Lawson) “Explaining the collapse of political states” (Masters Thesis, 2011)
Forward to “diffusion and impact of culture”