For those who have been following the line of thinking about the potential collapse of civilization, starting with The Limits to Growth (1972) and including Jared Diamond's Collapse and Thomas Homer-Dixon's The Upside of Down (see Dahl 2008), there is an interesting new approach from a mathematical modeler. Peter Turchin, a mathematical ecologist in the USA, felt he needed a new challenge after modeling the rise and fall of biological populations, and turned to history as a field that had not yet been explored by mathematicians. He assumed that the growth of a civilization or empire depends on social cohesion. Collecting data sets on a number of past civilizations, using as an indicator collective violence, he looked for patterns and cycles. Among other things, he found a 200 year cycle in which population growth and technological innovation create wealth that is concentrated by a wealth elite or expanding upper class. Eventually an oversupply of labour makes it possible to drive the workers further into poverty, but the poor do not revolt. A generation later, the young people who no longer have access to the shrinking elite become the revolutionaries, producing factionalism, anarchy and ultimately collapse, before the cycle starts over again. He predicted a risk of political instability and impending crisis in Western Europe and the USA peaking in 2020. The only way to avoid this would be to reduce social inequality.
The reference to the original paper is Turchin, Peter, 2010. Political instability may be a contributor in the coming decade. Nature, vol. 463, p. 608 (4 February 2010). doi:10.1038/463608a. It was cited in Holmes, Bob, 2012, Revolutionary Cycles, New Scientist, 18 August 2012, p. 46-49.