The new history

Here is the original draft of the end-of-year essay I published in the last 2011 issue of Nature.
___________________________________________________

2011 shows that our highly networked society is ever more prone to abrupt change. The future of our complex world depends on building resilience to shocks.

In the 1990s, American political scientist Francis Fukuyama, now at Stanford, predicted that the world was approaching the ‘end of history’ [1]. Like most smart ideas that prove to be wrong, Fukuyama’s was illuminating precisely for its errors. Events this year have helped to reveal why.

Fukuyama argued that after the collapse of the Soviet Union, liberal democracy could be seen as the logical and stable end point of civilization. Yet the prospect that the world will gradually replicate the US model of liberal democracy, as Fukuyama hoped, looks more remote today than it did at the end of the twentieth century.

This year we have seen proliferating protest movements in the fallout from the financial crisis – not just the cries of the marginalized and disaffected, but genuine challenges to the legitimacy of the economic system on which recent liberal democracies have been based. In the face of the grave debt crisis in Greece, the wisdom of deploying democracy’s ultimate tool – the national referendum – to solve it was questioned. The political situation in Russia and Turkey suggests that there is nothing inexorable or irreversible about a process of democratization, while North Africa and the Middle East demonstrate to politicians what political scientists could already have told them: that democratization can itself inflame conflict, especially when it is imposed in the absence of a strong pre-existing state [2,3]. Meanwhile, China continues to show that aggressive capitalism depends on neither liberalism nor democracy. As a recent report of the US National Intelligence Council admits, in the coming years “the Western model of economic liberalism, democracy, and secularism, which many assumed to be inevitable, may lose its luster” [4].

The real shortcoming behind Fukuyama’s thesis, however, was not his faith democracy but that he considered history to be gradualist: tomorrow’s history is more (or less) of the same. The common talk among political analysts now is of ‘discontinuous change’, a notion raised by Irish philosopher Charles Handy 20 years ago [5], and alluded to by President Obama in his speech at the West Point Military Academy last year, when he spoke of ‘moments of change’. Sudden disruptive events, particularly wars, have of course always been a part of history. But they would come and go against a slowly evolving social, cultural and political backdrop. Now the potential for discontinuous social and political change is woven into the very fabric of global affairs.

Take the terrorist attack on the World Trade Centre’s twin towers in 2001. This was said by many to have proved Fukuyama wrong – but on this tenth anniversary of that event we can now see more clearly in what sense that was so. It was not simply that this was a significant historical event – Fukuyama was never claiming that those would cease. Rather, it was a harbinger of the new world order, which the subsequent ‘war on terror’ failed catastrophically to acknowledge. That was a war waged in the old way, by sending armies to battlegrounds (in Afghanistan and Iraq) according to Carl von Clausewitz’s old definition, in his classic 1832 work On War, of a continuation of international politics by other means. But not only were those wars in no sense ‘won, they were barely wars at all – illustrating the remark of American strategic analyst Anthony Cordesman that “one of the lessons of modern war is that war can no longer be called war” [6]. Rather, armed conflict is a diffuse, nebulous affair, no longer corralled from peacetime by declarations and treaties, no longer recognizing generals or even statehood. In its place is a network of insurgents, militias, terrorist cells, suicide bombers, overlapping and sometimes competing ‘enemy’ organizations [7]. Somewhere in this web we have had to say farewell to war and peace.

Network revolutions

The nature of discontinuous change is often misunderstood. It is sometimes said – this is literally the defence of traditional economists in their failure to predict the on-going financial and national-debt crises – that no one can be expected to foresee such radical departures from the previous quotidian. They come, like a hijacked aircraft, out of a clear blue sky. Yet social and political discontinuities are rarely if ever random in that sense, even if there is a certain arbitrary character to their immediate triggers. Rather, they are abrupt in the same way, and for the same reasons, that phase transitions are abrupt in physics. In complex systems, including social ones, discontinuities don’t reflect profound changes in the governing forces but instead derive from the interactions and feedbacks between the component parts. Thus, discontinuities in history are precisely what you'd expect if you start considering social phenomena from a complex-systems perspective.

Experience with natural and technological complex systems teaches us, for example, that highly connected networks of strong interactions create a propensity for avalanches, catastrophic failures, and systemic ruptures [8,9]: in short, for discontinuous change.

So it should come as no surprise that today’s highly networked, interconnected world, replete with cell phones, ipads and social media, is prone to abrupt changes in course. It is much more than idle analogy that connects the cascade of minor failures leading to the 2003 power blackout of eastern North America with the freezing of liquidity in the global banking network in 2007-8.

Some see the revolts in Tunisia and Egypt in this way too, dubbing them ‘Twitter revolutions’ because of the way unrest and news of demonstration were spread on social networks. Although this is an over-simplification, it is abundantly clear that networking supplied the possibility for a random event to trigger a major one. The Tunisian revolt was set in motion by the self-immolation of a street vendor, Mohammed Bouazizi, in Sidi Bouzid, in protest at harsh treatment by officials. Three months earlier there was a similar case in the city of Monastir – but no one knew about it because the news was not spread on Facebook.

It was surely not without reasons that Twitter and Facebook were shut down by both the Tunisian and Egyptian authorities. The issue is not so much whether they ‘caused’ the revolutions, but that their existence – and the concomitant potential for mobilizing the young, educated populations of these countries – can alter the way things happen in the Middle East and beyond. These same tools are now vital to the Occupy protests disrupting complacent financial districts worldwide, from New York to Taipei, drawing attention to issues of social and economic inequality.

Social media seem also to have the potential to facilitate qualitatively new collective behaviours, such as the riots during the summer in the UK. These brief, destructive paroxysms are still an enigma. Unlike previous riots, they were not confined either to particular demographic subsets of the population or to areas of serious social deprivation. They had no obvious agenda, not even a release of suppressed communal fury – although there was surely a link to post-financial-crash austerity policies. One might almost call them events that grew simply because they could. Some British politicians suggested that Twitter should be disabled in such circumstances, displaying not only a loss of perspective (some of the same people celebrated the power of networking in the Arab Spring) but also a failure to understand the new order. After all, police monitoring of Twitter in some UK cities provided information that helped suppress rioting.

What all these events really point towards is the profound impact of globalization. They show how deep and dense the interdependence of economies, cultures and institutions has become, in large part thanks to the pervasive nature of information and communication technologies. And with this transformation come new, spontaneous modes of social and political organization, from terrorist and protest networks to online consumerism – modes that are especially prone to discontinuous change. Nothing will work that fails to take this new interconnectedness into account: not the economy, not policing, not democracy.

The path forwards

Such extreme interdependence makes it hard to find, or even to meaningfully define, the causes of major events. The US subprime mortgage problem caused the financial collapse only in the way Bouazizi’s immolation caused the Arab Spring – it could equally have been something else that set events in motion. The real vulnerabilities were systemic: webs of dependence that became destabilized by, say, runaway profits in the US banking industry, or rising food prices in North Africa. This means that potential solutions must lie there too.

Complex systems can rarely if ever be controlled by top-down measures. Instead, they must be managed by guiding the trajectories from the bottom up [10]. In a much simpler but instructive example, traffic lights may direct flows more efficiently if they are given adaptive autonomy and allowed to self-organize their switching, rather than imposing a rigid, supposedly optimal sequence [11]. The robustness of the Internet to random server failures is precisely due to the fact that no one designed it – it grew its ‘small world’ topology spontaneously.

This does not imply that political interventions are doomed to fail, but just that they must take other forms from those often advanced today. “Complex systems cannot be steered like a bus”, says Dirk Helbing of the Swiss Federal Institute of Technology (ETH) in Zurich, a specialist on the understanding and management of complex social systems. “Attempts to control the systems from the top down may be strong enough to disturb its intrinsic self-organization but not strong enough to re-establish order. The result would be chaos and inefficiency. Modern governance typically changes the institutional framework too quickly to allow individuals and companies to adapt. This destroys the hierarchy of time scales needed to establish stable order.”

But these systems are nevertheless manageable, Helbing insists – not by imposing structures but by creating the rules needed to allow the system to find its own stable organization. “This can’t be ensured by a regulatory authority that monitors the system and tries to enforce specific individual action”, he says.

That’s why theories or ideologies are likely to be less effective at predicting or averting crises than scenario modelling. It’s why problems need to be considered at several hierarchical levels, probably with multiple, overlapping models, and why solutions must have scope for adaptation and flexibility. And although cascading crises and discontinuous changes may be unpredictable, the connections and vulnerabilities that permit them are not. Planning for the future, then, might not be so much a matter of foreseeing what could go wrong as of making our systems and institutions robust enough to withstand a variety of shocks. This is how the new history will work.

References
1. Fukuyama, F. The End of History and the Last Man (Penguin, London, 1992).
2. E. D. Mansfield & Snyder, J. Int. Secur. 20, 5–38 (1995).
3. Cederman, L.-E., Hug, S. & Wenger, A., in Democritization (eds Grimm, S. & Merkel, W.), 15, 509-524 (Routledge, London, 2008).
4. National Intelligence Council, Global Trends 2025: A Transformed World (US Government Printing Office, Washington DC, 2008).
5. Handy, C., The Age of Unreason (Harvard Business School Press, Boston, 1990).
6. In H. Strachan, Europaeum Lecture, Geneva, 9 November 2006, p. 12.
7. J. C. Bohorquez, S. Gourley, A. R. Dixon, M. Spagat & N. F. Johnson, Nature 462, 911-914 (2009).
8. Barabási, A.-L. IEEE Control Syst. Mag. 27(4), 33-42 (2007).
9. Vespignani, A. Nature 464, 984-985 (2010).
10. Helbing, D. (ed.), Managing Complexity: Insights, Concepts, Applications (Springer, Berlin, 2008).
11. Lämmer, S. & Helbing, D., J. Stat. Mech. P04019 (2008).