New England Complex Systems Institute

Cambridge, MA, United States

New England Complex Systems Institute

Cambridge, MA, United States
Time filter
Source Type

Many political observers have coalesced around a prevailing theory that they say explains Donald Trump's early struggles. Put simply, they say he's too dumb to be president. These are judgment calls made by policy experts, politicians, journalists, and pundits. However, there is also an emerging branch of theoretical physics and mathematics that suggests—in cold, analytical, nonpartisan terms—that these pundits are exactly right. The hypothesis is not that Trump is stupid, it's that human society has become too complex for any president to be effective, according to Yaneer Bar-Yam, the president of the New England Complex Systems Institute, which partners with MIT and Harvard. By leaving hundreds of key advising positions in the State Department, Office of Science and Technology Policy, and the broader executive branch unfilled, Trump in particular has reduced the number of people advising him on difficult policy decisions, putting more strain on an already broken system. Trump's promises of easy solutions to complex problems—Build a wall, healthcare for all, ban Muslims, Make America Great Again—is why he was so appealing to people who have become disillusioned with government. But now that he's part of the system, he's getting destroyed by it. "People believing Trump can be a solution to the problems we're facing is a serious issue because they have expectations that are not consistent with the structural issues we're facing," Bar-Yam said. Bar-Yam's area of study is to literally explain why the world works the way it does, and he has repeatedly hit the nail right on the head with his hypotheses and explanations, which use big data sets and supercomputers to separate the signal from the noise. For instance, he used global food price indexes to warn the US intelligence community that the Arab Spring would happen several days before it did. He then went back and explained that several relatively esoteric and seemingly unconnected American regulations led to the spikes that caused the initial riots. I explained Bar-Yam's hypothesis about governance and the math behind it in depth shortly before the election, but the gist is that society has reached a point of complexity that his research suggests is too complicated for the representative democracy we've set up. Power is concentrated in a small number of individuals who do not have the cognitive capabilities or information available to make well-considered decisions with any sort of reliability. With that framework in mind, he says that any president would be likely to fail in our current system. But Trump has made the US government fundamentally less distributed and is therefore putting more stress on American institutions that have evolved to be more complex alongside society. "The basic challenge is we still have to learn what the impacts of our decisions are actually going to be" For example, Trump's hiring freeze has left thousands of policy expert jobs across the executive branch unfilled. Disconcertingly, the administration has left nearly 200 top jobs open at the State Department, which would normally be filled with policy experts who understand geopolitical machinations. This means that decisions about domestic and global policy are being made by Trump and his closest advisors, as opposed to a more lateral system of experts advising them. Reports suggest that Trump also makes decisions on-the-fly, dispensing with the advice and plans of some of his closest advisors. "To the extent that decision making is limited to a few individuals, the complexity of the possible actions taken is limited to the complexity of that team," Bar-Yam told me. "The need to distribute decision making across government is a key part of more effective decision making." Trump's assaults on the sovereignty of the courts and Congress's hesitance to put pressure on Trump further undermine the complexity of the Constitutional system that was set up by the founding fathers, pushing us toward a more rudimentary form of government that's less prepared to meet the challenges of a globally connected society. "Checks and balances in government are a key part of the system that create more complexity, and shared decision making is what checks and balances are about," he said. "To the extent that checks and balances are not being actively used, it's a simplification of the decision making. Generically, dictatorships are better than disorder, but democracies are better than dictatorship. What we'd want to replace this system with is a more effective system, not a less effective one." Trump himself has acknowledged that he's not prepared for the complexity of governing the United States in 2017: In April, he told Reuters that he "thought would be easier." In February, he said "nobody knew that health care could be so complicated." "That's a classic statement of misunderstanding the nature of complexity," Bar-Yam said. With health care, he says politicians are discussing solutions—single payer or a market-driven system—without understanding the nature of healthcare itself, which is a result of many inputs and policies across every sector of American government. "The fundamental problems with health care have to do with the fact that it is a complex system and these problems cannot be addressed with simple choices—one has to restructure the system entirely to have individual solutions for people that are specific to them." "In the environment in which we live, the complexity progressively becomes higher and higher and it's basically like we're making random choices" Talking to Bar-Yam is both comforting and disconcerting—his assertion that current systems of governance around the world simply don't work because the world is too complicated for a couple people to make smart policy makes as much sense as any other explanation out there. But it's also a hypothesis that doesn't have an easy solution—if he's right, we can't simply impeach or elect our way out of this mess, we would literally have to write a new Constitution that allows for more teams of policy experts to make decisions for the country without devolving into authoritarian rule or utter chaos. Throughout the years we've created band-aid solutions to this lack of complexity within the current system—it's why under President Obama and previous administrations there have been so many policymaking task forces, commissions, and panels. But Bar-Yam says that while those sorts of teams have the ability to make government work better within the existing system, the ultimate decision-making power still filters up into a few individuals. Ideally, the teams he imagines would have the power to enact policy themselves. "It's important to understand violent revolution is regressive rather than progressive—you cannot create a complex structure quickly," Bar-Yam said. He also suggests that the left-right divide in the United States and around the world is kind of a red herring; the real problem is that we often don't even know what the outcomes of policy decisions will actually be. We are quickly moving toward a world that is so complex that politicians can't be expected to anticipate the outcomes of their decisions, which means that there's less of a divide between right and left than you might think: "In the environment in which we live, the complexity progressively becomes higher and higher and it's basically like we're making random choices. The imperative to share decision making becomes greater and greater over time. The difference between even an adult and a child making decisions is shrinking," he said. "People can have different values and seek different things, but if the decision about what you want to do creates policies that don't actually achieve what you were intending, it doesn't matter what your values or ideologies are—you're just ineffective," he said. "The basic challenge is we still have to learn what the impacts of our decisions are actually going to be."

News Article | November 23, 2016

Try, for a moment, to envisage a world without countries. Imagine a map not divided into neat, coloured patches, each with clear borders, governments, laws. Try to describe anything our society does – trade, travel, science, sport, maintaining peace and security – without mentioning countries. Try to describe yourself: you have a right to at least one nationality, and the right to change it, but not the right to have none. Those coloured patches on the map may be democracies, dictatorships or too chaotic to be either, but virtually all claim to be one thing: a nation state, the sovereign territory of a “people” or nation who are entitled to self-determination within a self-governing state. So says the United Nations, which now numbers 193 of them. And more and more peoples want their own state, from Scots voting for independence to jihadis declaring a new state in the Middle East. Many of the big news stories of the day, from conflicts in Gaza and Ukraine to rows over immigration and membership of the European Union, are linked to nation states in some way. Even as our economies globalise, nation states remain the planet’s premier political institution. Large votes for nationalist parties in this year’s EU elections prove nationalism remains alive – even as the EU tries to transcend it. Yet there is a growing feeling among economists, political scientists and even national governments that the nation state is not necessarily the best scale on which to run our affairs. We must manage vital matters like food supply and climate on a global scale, yet national agendas repeatedly trump the global good. At a smaller scale, city and regional administrations often seem to serve people better than national governments. How, then, should we organise ourselves? Is the nation state a natural, inevitable institution? Or is it a dangerous anachronism in a globalised world? These are not normally scientific questions – but that is changing. Complexity theorists, social scientists and historians are addressing them using new techniques, and the answers are not always what you might expect. Far from timeless, the nation state is a recent phenomenon. And as complexity keeps rising, it is already mutating into novel political structures. Get set for neo-medievalism. Before the late 18th century there were no real nation states, says John Breuilly of the London School of Economics. If you travelled across Europe, no one asked for your passport at borders; neither passports nor borders as we know them existed. People had ethnic and cultural identities, but these didn’t really define the political entity they lived in. That goes back to the anthropology, and psychology, of humanity’s earliest politics. We started as wandering, extended families, then formed larger bands of hunter-gatherers, and then, around 10,000 years ago, settled in farming villages. Such alliances had adaptive advantages, as people cooperated to feed and defend themselves. But they also had limits. Robin Dunbar of the University of Oxford has shown that one individual can keep track of social interactions linking no more than around 150 people. Evidence for that includes studies of villages and army units through history, and the average tally of Facebook friends. But there was one important reason to have more friends than that: war. “In small-scale societies, between 10 and 60 per cent of male deaths are attributable to warfare,” says Peter Turchin of the University of Connecticut at Storrs. More allies meant a higher chance of survival. Turchin has found that ancient Eurasian empires grew largest where fighting was fiercest, suggesting war was a major factor in political enlargement. Archaeologist Ian Morris of Stanford University in California reasons that as populations grew, people could no longer find empty lands where they could escape foes. The losers of battles were simply absorbed into the enemy’s domain – so domains grew bigger. How did they get past Dunbar’s number? Humanity’s universal answer was the invention of hierarchy. Several villages allied themselves under a chief; several chiefdoms banded together under a higher chief. To grow, these alliances added more villages, and if necessary more layers of hierarchy. Hierarchies meant leaders could coordinate large groups without anyone having to keep personal track of more than 150 people. In addition to their immediate circle, an individual interacted with one person from a higher level in the hierarchy, and typically eight people from lower levels, says Turchin. These alliances continued to enlarge and increase in complexity in order to perform more kinds of collective actions, says Yaneer Bar-Yam of the New England Complex Systems Institute in Cambridge, Massachusetts. For a society to survive, its collective behaviour must be as complex as the challenges it faces – including competition from neighbours. If one group adopted a hierarchical society, its competitors also had to. Hierarchies spread and social complexity grew. Larger hierarchies not only won more wars but also fed more people through economies of scale, which enabled technical and social innovations such as irrigation, food storage, record-keeping and a unifying religion. Cities, kingdoms and empires followed. But these were not nation states. A conquered city or region could be subsumed into an empire regardless of its inhabitants’ “national” identity. “The view of the state as a necessary framework for politics, as old as civilisation itself, does not stand up to scrutiny,” says historian Andreas Osiander of the University of Leipzig in Germany. “The view of the state as a necessary framework for politics does not stand up” One key point is that agrarian societies required little actual governing. Nine people in 10 were peasants who had to farm or starve, so were largely self-organising. Government intervened to take its cut, enforce basic criminal law and keep the peace within its undisputed territories. Otherwise its main role was to fight to keep those territories, or acquire more. Even quite late on, rulers spent little time governing, says Osiander. In the 17th century Louis XIV of France had half a million troops fighting foreign wars but only 2000 keeping order at home. In the 18th century, the Dutch and Swiss needed no central government at all. Many eastern European immigrants arriving in the US in the 19th century could say what village they came from, but not what country: it didn’t matter to them. Before the modern era, says Breuilly, people defined themselves “vertically” by who their rulers were. There was little horizontal interaction between peasants beyond local markets. Whoever else the king ruled over, and whether those people were anything like oneself, was largely irrelevant. Such systems are very different from today’s states, which have well-defined boundaries filled with citizens. In a system of vertical loyalties, says Breuilly, power peaks where the overlord lives and peters out in frontier territories that shade into neighbouring regions. Ancient empires are coloured on modern maps as if they had firm borders, but they didn’t. Moreover, people and territories often came under different jurisdictions for different purposes. Such loose control, says Bar-Yam, meant pre-modern political units were only capable of scaling up a few simple actions such as growing food, fighting battles, collecting tribute and keeping order. Some, like the Roman Empire, did this on a very large scale. But complexity – the different actions society could collectively perform – was relatively low. Complexity was limited by the energy a society could harness. For most of history that essentially meant human and animal labour. In the late Middle Ages, Europe harnessed more, especially water power. This boosted social complexity – trade increased, for example– requiring more government. A decentralised feudal system gave way to centralised monarchies with more power. But these were still not nation states. Monarchies were defined by who ruled them, and rulers were defined by mutual recognition – or its converse, near-constant warfare. In Europe, however, as trade grew, monarchs discovered they could get more power from wealth than war. In 1648, Europe’s Peace of Westphalia ended centuries of war by declaring existing kingdoms, empires and other polities “sovereign”: none was to interfere in the internal affairs of others. This was a step towards modern states – but these sovereign entities were still not defined by their peoples’ national identities. International law is said to date from the Westphalia treaty, yet the word “international” was not coined until 132 years later. By then Europe had hit the tipping point of the industrial revolution. Harnessing vastly more energy from coal meant that complex behaviours performed by individuals, such as weaving, could be amplified, says Bar-Yam, producing much more complex collective behaviours. This demanded a different kind of government. In 1776 and 1789, revolutions in the US and France created the first nation states, defined by the national identity of their citizens rather than the bloodlines of their rulers. According to one landmark history of the period, says Breuilly, “in 1800 almost nobody in France thought of themselves as French. By 1900 they all did.” For various reasons, people in England had an earlier sense of “Englishness”, he says, but it was not expressed as a nationalist ideology. By 1918, with the dismemberment of Europe’s last multinational empires such as the Habsburgs in the first world war, European state boundaries had been redrawn largely along cultural and linguistic lines. In Europe at least, the nation state was the new norm. Part of the reason was a pragmatic adaptation of the scale of political control required to run an industrial economy. Unlike farming, industry needs steel, coal and other resources which are not uniformly distributed, so many micro-states were no longer viable. Meanwhile, empires became unwieldy as they industrialised and needed more actual governing. So in 19th-century Europe, micro-states fused and empires split. These new nation states were justified not merely as economically efficient, but as the fulfilment of their inhabitants’ national destiny. A succession of historians has nonetheless concluded that it was the states that defined their respective nations, and not the other way around. France, for example, was not the natural expression of a pre-existing French nation. At the revolution in 1789, half its residents did not speak French. In 1860, when Italy unified, only 2.5 per cent of residents regularly spoke standard Italian. Its leaders spoke French to each other. One famously said that, having created Italy, they now had to create Italians – a process many feel is still taking place. “At the revolution in 1789, half of France’s residents did not speak French” Sociologist Siniša Maleševic of University College Dublin in Ireland believes that this “nation building” was a key step in the evolution of modern nation states. It required the creation of an ideology of nationalism that emotionally equated the nation with people’s Dunbar circle of family and friends. That in turn relied heavily on mass communication technologies. In an influential analysis, Benedict Anderson of Cornell University in New York described nations as “imagined” communities: they far outnumber our immediate circle and we will never meet them all, yet people will die for their nation as they would for their family. Such nationalist feelings, he argued, arose after mass-market books standardised vernaculars and created linguistic communities. Newspapers allowed people to learn about events of common concern, creating a large “horizontal” community that was previously impossible. National identity was also deliberately fostered by state-funded mass education. The key factor driving this ideological process, Maleševic says, was an underlying structural one: the development of far-reaching bureaucracies needed to run complex industrialised societies. For example, says Breuilly, in the 1880s Prussia became the first government to pay unemployment benefits. At first they were paid only in a worker’s native village, where identification was not a problem. As people migrated for work, benefits were made available anywhere in Prussia. “It wasn’t until then that they had to establish who a Prussian was,” he says, and they needed bureaucracy to do it. Citizenship papers, censuses and policed borders followed. That meant hierarchical control structures ballooned, with more layers of middle management. Such bureaucracy was what really brought people together in nation-sized units, argues Maleševic. But not by design: it emerged out of the behaviour of complex hierarchical systems. As people do more kinds of activities, says Bar-Yam, the control structure of their society inevitably becomes denser. In the emerging nation state, that translates into more bureaucrats per head of population. Being tied into such close bureaucratic control also encouraged people to feel personal ties with the state, especially as ties to church and village declined. As governments exerted greater control, people got more rights, such as voting, in return. For the first time, people felt the state was theirs. Once Europe had established the nation state model and prospered, says Breuilly, everyone wanted to follow suit. In fact it’s hard now to imagine that there could be another way. But is a structure that grew spontaneously out of the complexity of the industrial revolution really the best way to manage our affairs? According to Brian Slattery of York University in Toronto, Canada, nation states still thrive on a widely held belief that “the world is naturally made of distinct, homogeneous national or tribal groups which occupy separate portions of the globe, and claim most people’s primary allegiance”. But anthropological research does not bear that out, he says. Even in tribal societies, ethnic and cultural pluralism has always been widespread. Multilingualism is common, cultures shade into each other, and language and cultural groups are not congruent. Moreover, people always have a sense of belonging to numerous different groups based on region, culture, background and more. “The claim that a person’s identity and well-being is tied in a central way to the well-being of the national group is wrong as a simple matter of historical fact,” says Slattery. Perhaps it is no wonder, then, that the nation-state model fails so often: since 1960 there have been more than 180 civil wars worldwide. Such conflicts are often blamed on ethnic or sectarian tensions. Failed states, such as Syria right now, are typically riven by violence along such lines. According to the idea that nation states should contain only one nation, such failures have often been blamed on the colonial legacy of bundling together many peoples within unnatural boundaries. But for every Syria or Iraq there is a Singapore, Malaysia or Tanzania, getting along okay despite having several “national” groups. Immigrant states in Australia and the Americas, meanwhile, forged single nations out of massive initial diversity. What makes the difference? It turns out that while ethnicity and language are important, what really matters is bureaucracy. This is clear in the varying fates of the independent states that emerged as Europe’s overseas empires fell apart after the second world war. According to the mythology of nationalism, all they needed was a territory, a flag, a national government and UN recognition. In fact what they really needed was complex bureaucracy. Some former colonies that had one became stable democracies, notably India. Others did not, especially those such as the former Belgian Congo, whose colonial rulers had merely extracted resources. Many of these became dictatorships, which require a much simpler bureaucracy than democracies. Dictatorships exacerbate ethnic strife because their institutions do not promote citizens’ identification with the nation. In such situations, people fall back on trusted alliances based on kinship, which readily elicit Dunbar-like loyalties. Insecure governments allied to ethnic groups favour their own, while grievances among the disfavoured groups grow – and the resulting conflict can be fierce. Recent research confirms that the problem is not ethnic diversity itself, but not enough official inclusiveness. Countries with little historic ethnic diversity are now having to learn that on the fly, as people migrate to find jobs within a globalised economy. How that pans out may depend on whether people self-segregate. Humans like being around people like themselves, and ethnic enclaves can be the result. Jennifer Neal of Michigan State University in East Lansing has used agent-based modelling to look at the effect of this in city neighbourhoods. Her work suggests that enclaves promote social cohesion, but at the cost of decreasing tolerance between groups. Small enclaves in close proximity may be the solution. But at what scale? Bar-Yam says communities where people are well mixed – such as in peaceable Singapore, where enclaves are actively discouraged – tend not to have ethnic strife. Larger enclaves can also foster stability. Using mathematical models to correlate the size of enclaves with the incidences of ethnic strife in India, Switzerland and the former Yugoslavia, he found that enclaves 56 kilometres or more wide make for peaceful coexistence – especially if they are separated by natural geographical barriers, Switzerland’s 26 cantons, for example, which have different languages and religions, meet Bar-Yam’s spatial stability test – except one. A French-speaking enclave in German-speaking Berne experienced the only major unrest in recent Swiss history. It was resolved by making it a separate canton, Jura, which meets the criteria. Again, though, ethnicity and language are only part of the story. Lars-Erik Cederman of the Swiss Federal Institute of Technology in Zurich argues that Swiss cantons have achieved peace not by geographical adjustment of frontiers, but by political arrangements giving cantons considerable autonomy and a part in collective decisions. Similarly, using a recently compiled database to analyse civil wars since 1960, Cederman finds that strife is indeed more likely in countries that are more ethnically diverse. But careful analysis confirms that trouble arises not from diversity alone, but when certain groups are systematically excluded from power. Governments with ethnicity-based politics were especially vulnerable. The US set up just such a government in Iraq after the 2003 invasion. Exclusion of Sunni by Shiites led to insurgents declaring a Sunni state in occupied territory in Iraq and Syria. True to nation-state mythology, it rejects the colonial boundaries of Iraq and Syria, as they force dissimilar “nations” together. Yet the solution cannot be imposing ethnic uniformity. Historically, so-called ethnic cleansing has been uniquely bloody, and “national” uniformity is no guarantee of harmony. In any case, there is no good definition of an ethnic group. Many people’s ethnicities are mixed and change with the political weather: the numbers who claimed to be German in the Czech Sudetenland territory annexed by Hitler changed dramatically before and after the war. Russian claims to Russian-speakers in eastern Ukraine now may be equally flimsy. Both Bar-Yam’s and Cederman’s research suggests one answer to diversity within nation states: devolve power to local communities, as multicultural states such as Belgium and Canada have done. “We need a conception of the state as a place where multiple affiliations and languages and religions may be safe and flourish,” says Slattery. “That is the ideal Tanzania has embraced and it seems to be working reasonably well.” Tanzania has more than 120 ethnic groups and about 100 languages. In the end, what may matter more than ethnicity, language or religion is economic scale. The scale needed to prosper may have changed with technology – tiny Estonia is a high-tech winner – but a small state may still not pack enough economic power to compete. That is one reason why Estonia is such an enthusiastic member of the European Union. After the devastating wars in the 20th century, European countries tried to prevent further war by integrating their basic industries. That project, which became the European Union, now primarily offers member states profitable economies of scale, through manufacturing and selling in the world’s largest single market. What the EU fails to inspire is nationalist-style allegiance – which Maleševic thinks nowadays relies on the “banal” nationalism of sport, anthems, TV news programmes, even song contests. That means Europeans’ allegiances are no longer identified with the political unit that handles much of their government. Ironically, says Jan Zielonka of the University of Oxford, the EU has saved Europe’s nation states, which are now too small to compete individually. The call by nationalist parties to “take back power from Brussels”, he argues, would lead to weaker countries, not stronger ones. He sees a different problem. Nation states grew out of the complex hierarchies of the industrial revolution. The EU adds another layer of hierarchy – but without enough underlying integration to wield decisive power. It lacks both of Maleševic’s necessary conditions: nationalist ideology and pervasive integrating bureaucracy. Even so, the EU may point the way to what a post-nation-state world will look like. Zielonka agrees that further integration of Europe’s governing systems is needed as economies become more interdependent. But he says Europe’s often-paralysed hierarchy cannot achieve this. Instead he sees the replacement of hierarchy by networks of cities, regions and even non-governmental organisations. Sound familiar? Proponents call it neo-medievalism. “The future structure and exercise of political power will resemble the medieval model more than the Westphalian one,” Zielonka says. “The latter is about concentration of power, sovereignty and clear-cut identity.” Neo-medievalism, on the other hand, means overlapping authorities, divided sovereignty, multiple identities and governing institutions, and fuzzy borders. “The future exercise of power will resemble the medieval model” Anne-Marie Slaughter of Princeton University, a former US assistant secretary of state, also sees hierarchies giving way to global networks primarily of experts and bureaucrats from nation states. For example, governments now work more through flexible networks such as the G7 (or 8, or 20) to manage global problems than through the UN hierarchy. Ian Goldin, head of the Oxford Martin School at the University of Oxford, which analyses global problems, thinks such networks must emerge. He believes existing institutions such as UN agencies and the World Bank are structurally unable to deal with problems that emerge from global interrelatedness, such as economic instability, pandemics, climate change and cybersecurity – partly because they are hierarchies of member states which themselves cannot deal with these global problems. He quotes Slaughter: “Networked problems require a networked response.” Again, the underlying behaviour of systems and the limits of the human brain explain why. Bar-Yam notes that in any hierarchy, the person at the top has to be able to get their head around the whole system. When systems are too complex for one human mind to grasp, he argues that they must evolve from hierarchies into networks where no one person is in charge. Where does this leave nation states? “They remain the main containers of power in the world,” says Breuilly. And we need their power to maintain the personal security that has permitted human violence to decline to all-time lows. Moreover, says Dani Rodrik of Princeton’s Institute for Advanced Study, the very globalised economy that is allowing these networks to emerge needs something or somebody to write and enforce the rules. Nation states are currently the only entities powerful enough to do this. Yet their limitations are clear, both in solving global problems and resolving local conflicts. One solution may be to pay more attention to the scale of government. Known as subsidiarity, this is a basic principle of the EU: the idea that government should act at the level where it is most effective, with local government for local problems and higher powers at higher scales. There is empirical evidence that it works: social and ecological systems can be better governed when their users self-organise than when they are run by outside leaders. However, it is hard to see how our political system can evolve coherently in that direction. Nation states could get in the way of both devolution to local control and networking to achieve global goals. With climate change, it is arguable that they already have. There is an alternative to evolving towards a globalised world of interlocking networks, neo-medieval or not, and that is collapse. “Most hierarchical systems tend to become top-heavy, expensive and incapable of responding to change,” says Marten Scheffer of Wageningen University in the Netherlands. “The resulting tension may be released through partial collapse.” For nation states, that could mean anything from the renewed pre-eminence of cities to Iraq-style anarchy. An uncertain prospect, but there is an upside. Collapse, say some, is the creative destruction that allows new structures to emerge. Like it or not, our societies may already be undergoing this transition. We cannot yet imagine there are no countries. But recognising that they were temporary solutions to specific historical situations can only help us manage a transition to whatever we need next. Whether or not our nations endure, the structures through which we govern our affairs are due for a change. Time to start imagining. Leader: “In our world beyond nations, the future is medieval” This article appeared in print under the headline “Imagine there’s no countries…”

Yeakel J.D.,Simon Fraser University | Moore J.W.,Simon Fraser University | Guimaraes P.R.,University of Sao Paulo | de Aguiar M.A.M.,University of Campinas | de Aguiar M.A.M.,New England Complex Systems Institute
Ecology Letters | Year: 2014

Spatial structure in landscapes impacts population stability. Two linked components of stability have large consequences for persistence: first, statistical stability as the lack of temporal fluctuations; second, synchronisation as an aspect of dynamic stability, which erodes metapopulation rescue effects. Here, we determine the influence of river network structure on the stability of riverine metapopulations. We introduce an approach that converts river networks to metapopulation networks, and analytically show how fluctuation magnitude is influenced by interaction structure. We show that river metapopulation complexity (in terms of branching prevalence) has nonlinear dampening effects on population fluctuations, and can also buffer against synchronisation. We conclude by showing that river transects generally increase synchronisation, while the spatial scale of interaction has nonlinear effects on synchronised dynamics. Our results indicate that this dual stability - conferred by fluctuation and synchronisation dampening - emerges from interaction structure in rivers, and this may strongly influence the persistence of river metapopulations. © 2013 John Wiley & Sons Ltd/CNRS.

De Aguiar M.A.M.,New England Complex Systems Institute | De Aguiar M.A.M.,University of Campinas | Bar-Yam Y.,New England Complex Systems Institute
Physical Review E - Statistical, Nonlinear, and Soft Matter Physics | Year: 2011

In population genetics, the Moran model describes the neutral evolution of a biallelic gene in a population of haploid individuals subjected to mutations. We show in this paper that this model can be mapped into an influence dynamical process on networks subjected to external influences. The panmictic case considered by Moran corresponds to fully connected networks and can be completely solved in terms of hypergeometric functions. Other types of networks correspond to structured populations, for which approximate solutions are also available. This approach to the classic Moran model leads to a relation between regular networks based on spatial grids and the mechanism of isolation by distance. We discuss the consequences of this connection for topopatric speciation and the theory of neutral speciation and biodiversity. We show that the effect of mutations in structured populations, where individuals can mate only with neighbors, is greatly enhanced with respect to the panmictic case. If mating is further constrained by genetic proximity between individuals, a balance of opposing tendencies takes place: increasing diversity promoted by enhanced effective mutations versus decreasing diversity promoted by similarity between mates. Resolution of large enough opposing tendencies occurs through speciation via pattern formation. We derive an explicit expression that indicates when speciation is possible involving the parameters characterizing the population. We also show that the time to speciation is greatly reduced in comparison with the panmictic case. © 2011 American Physical Society.

Widener M.J.,State University of New York at Buffalo | Widener M.J.,New England Complex Systems Institute | Metcalf S.S.,State University of New York at Buffalo | Bar-Yam Y.,New England Complex Systems Institute
American Journal of Preventive Medicine | Year: 2011

Background: Low-income, urban populations' limited access to healthy foods is often pointed to as a key barrier to improving nutrition. Although much has been written on identifying urban "food deserts," little has been done to examine how the food environment changes over the course of 1 year. Purpose: This study was designed to dynamically describe the urban food environment as a means to identify when at-risk neighborhoods are without access to healthy food. Methods: Demographic and road data of Buffalo NY from the 2000 U.S. Census, a 2010 listing of city supermarkets, and 2011 government records of the time and location of urban farmers' markets are mapped. Road network distances from block groups to supermarkets and farmers' markets are calculated. A computer simulation, written in 2011, examines the market closest to each block group for 52 weeks. Results: The average distance to markets with produce from block groups with poverty levels in the top 10th percentile is greater than that across all block groups during winter and spring months. However, during the farmers' market season, the same impoverished block groups are on average closer to markets when compared to all block groups. Conclusions: Including the temporal dimension in an analysis of healthy food access generates a more complex picture of urban food-desert locations. The implications are that spatiotemporal factors should be used to inform appropriate interventions for creating an equitable food environment. © 2011 American Journal of Preventive Medicine.

Herdagdelen A.,University of Trento | Herdagdelen A.,New England Complex Systems Institute | Baroni M.,University of Trento
Journal of the American Society for Information Science and Technology | Year: 2011

We extracted gender-specific actions from text corpora and Twitter, and compared them with stereotypical expectations of people. We used Open Mind Common Sense (OMCS), a common sense knowledge repository, to focus on actions that are pertinent to common sense and daily life of humans. We use the gender information of Twitter users and web-corpus-based pronoun/name gender heuristics to compute the gender bias of the actions. With high recall, we obtained a Spearman correlation of 0.47 between corpus-based predictions and a human gold standard, and an area under the ROC curve of 0.76 when predicting the polarity of the gold standard. We conclude that it is feasible to use natural text (and a Twitter-derived corpus in particular) in order to augment common sense repositories with the stereotypical gender expectations of actions. We also present a dataset of 441 common sense actions with human judges' ratings on whether the action is typically/slightly masculine/feminine (or neutral), and another larger dataset of 21,442 actions automatically rated by the methods we investigate in this study. © 2011 ASIS&T.

Braha D.,New England Complex Systems Institute | Braha D.,University of Massachusetts Dartmouth | Braha D.,Massachusetts Institute of Technology
PLoS ONE | Year: 2012

Civil unrest is a powerful form of collective human dynamics, which has led to major transitions of societies in modern history. The study of collective human dynamics, including collective aggression, has been the focus of much discussion in the context of modeling and identification of universal patterns of behavior. In contrast, the possibility that civil unrest activities, across countries and over long time periods, are governed by universal mechanisms has not been explored. Here, records of civil unrest of 170 countries during the period 1919-2008 are analyzed. It is demonstrated that the distributions of the number of unrest events per year are robustly reproduced by a nonlinear, spatially extended dynamical model, which reflects the spread of civil disorder between geographic regions connected through social and communication networks. The results also expose the similarity between global social instability and the dynamics of natural hazards and epidemics. © 2012 Dan Braha.

Hill S.A.,University of Toledo | Braha D.,New England Complex Systems Institute | Braha D.,University of Massachusetts Dartmouth
Physical Review E - Statistical, Nonlinear, and Soft Matter Physics | Year: 2010

The characterization of the "most connected" nodes in static or slowly evolving complex networks has helped in understanding and predicting the behavior of social, biological, and technological networked systems, including their robustness against failures, vulnerability to deliberate attacks, and diffusion properties. However, recent empirical research of large dynamic networks (characterized by irregular connections that evolve rapidly) has demonstrated that there is little continuity in degree centrality of nodes over time, even when their degree distributions follow a power law. This unexpected dynamic centrality suggests that the connections in these systems are not driven by preferential attachment or other known mechanisms. We present an approach to explain real-world dynamic networks and qualitatively reproduce these dynamic centrality phenomena. This approach is based on a dynamic preferential attachment mechanism, which exhibits a sharp transition from a base pure random walk scheme. © 2010 The American Physical Society.

News Article | November 19, 2016

There are 7.4 billion people on the planet – nearly three times as many as there were 60 years ago. The UN estimates that in another 60 years we will be approaching 11 billion. Others say that population will peak soon, then fall gradually as we hit resource limits. There is another possibility: that hitting those limits causes our surprisingly fragile civilisation to collapse, triggering a global die-off. Civilisation may appear robust, but is actually a juggling act. We keep all the balls in the air using densely coupled networks of manufacturing, trade, money, employment, food, water, transport, energy, technology, healthcare, geopolitics and law and order. Each network depends on all the others through many feedback loops. In other words, civilisation is an adaptive, complex system – and such systems are susceptible to catastrophic failure. Loss of any essential subsystem can cause the entire edifice to crash, says Yaneer Bar-Yam of the New England Complex Systems Institute in Cambridge, Massachusetts. Even small glitches are big trouble, as we saw in 2008 when local financial failures cascaded though coupled systems to cause the global crisis we are still feeling. With civilisation itself, we don’t know what subsystems, or how much loss, would be enough to get us in serious trouble. But we have some idea what might trigger collapse. If global warming, for example, causes methane release from frozen deposits, we could get positive feedback: further warming, further release and runaway rising temperatures. Farming systems would fail in the face of rapid changes in weather, pests and diseases. Millions would starve. Other major risks are

News Article | November 8, 2016

Roughly two-thirds of Americans believe the country is going in the "wrong direction," and Tuesday the country will vote for two of the least popular presidential candidates of all time. Both the left and the right say that the United States' government is ineffective. One potential reason for this? Human society is simply too complex for representative democracy to work. The United States probably shouldn't have a president at all, according to an analysis by mathematicians at the New England Complex Systems Institute. NECSI is a research organization that uses math cribbed from the study of physical and chemical systems—bear with me for a moment—and newly available giant data sets to explain how events in one part of the world might affect something seemingly unrelated in another part of the world. Most famously, the institute's director, Yaneer Bar-Yam, predicted the Arab Spring several weeks before it happened. He found that seemingly unrelated policy decisions—ethanol subsidies in the US and the deregulation of commodity markets worldwide—led to skyrocketing food prices in 2008 and 2011. It turns out that there is a very neat correlation between the United Nations food price index and unrest and rioting worldwide that no one but Bar-Yam had picked up. When considering our system of government, the link between these policies and unexpected global violence is an illustrative but hardly unique one: Bar-Yam was able to describe these cause-and-effect relationships in detail because he looking at very specific inputs and very specific outputs. He was zooming in on specific parts of the "system" that is human civilization in an attempt to explain one small but important part of the world. It is absurd, then, to believe that the concentration of power in one or a few individuals at the top of a hierarchical representative democracy will be able to make optimal decisions on a vast array of connected and complex issues that will certainly have sweeping and unintended ramifications on other parts of human civilization. "There's a natural process of increasing complexity in the world," Bar-Yam told me. "And we can recognize that at some point, that increase in complexity is going to run into the complexity of the individual. And at that point, hierarchical organizations will fail." "We were raised to believe that democracy, and even the democracy that we have, is a system that has somehow inherent good to it," he added. But it's not just democracy that fails. "Hierarchical organizations are failing in the response to decision-making challenges. And this is true whether we're talking about dictatorships, or communism that had very centralized control processes, and for representative democracies today. Representative democracies still focus power in one or few individuals. And that concentration of control and decision-making makes those systems ineffective." This idea of a quantifiable, measurable "complexity," refers to the difficulty of describing what the hell is going on in a system. And Bar-Yam says that human society is just like every other system. An individual human is made up of atoms, which make up cells, which make up organs, and so on. Describing the behavior of each individual atom is incredibly difficult; describing the behavior of organs is less difficult, and it's trivially easy to tell you that my organs are doing something inside me right now to allow me to type on a computer right now. Collective behaviors are inherently more "simple" than individual ones, in other words. Describing the behavior of atoms is more complex than describing the collective behavior of the many atoms that make up a human being. This analogy extends to humans living in society. Predicting the specific behavior of a car factory worker in his day to day life is much harder than predicting that he and a collective of other people will produce cars at the factory. "In human organizations, coordination occurs because individuals influence each other's' behavior," Bar-Yam wrote in a paper explaining this hypothesis. "A control hierarchy is designed to enable a single individual to control the collective behavior." Governance, then, is an attempt to organize the behavior of many individually complex humans (like the atoms above) into something simpler and more coherent. "During the time of ancient empires, large-scale human systems executed relatively simple behaviors, and individuals performed relatively simple individual tasks that were repeated by many individuals over time to have a large scale effect," he added. The relatively simple nature of the world at the time allowed one single person to be a master of all aspects of governance, in other words. The collective behavior of the whole of a city, town, or empire in early society was easily describable, because everyone was doing more or less the same thing. The issue here is that the sheer scale and interdependence of society has vastly increased since the days of ancient empires, increasing the overall complexity of society. To take this back to the biological analogy, it is as if society itself has evolved from being a very simple organism, such as a microbe to something much more complex, like a human (in all likelihood society will get much more complex—maybe a better analogy is something like a jellyfish right now). This is what you would expect—physics theory suggests that all systems increase in complexity over time. "We've become fundamentally confused about what the decisions are, and what their consequences are. And we can't make a connection between them" Technological advances during the industrial revolution allowed the automation of menial tasks and diversified the number of tasks human beings could perform. The industrial revolution led to advances in transportation and shipping that connected disparate parts of the world, and the internet, computers, and smartphones, of course, have served to intermingle nearly every corner of the world. "Human society" is now one gigantic, incredibly complex system or organism rather than many smaller, isolated, and simpler ones. This is how you end up with ethanol policies signed in America in the the late 1990s leading to widespread global unrest decades later. There are, of course, an unknowable number of decisions and events that have untold and difficult-to-predict effects on disparate parts of the world. A framework put forward by cybernetics pioneer Ross Ashby in the 1950s that served as the underlying basis for Bar-Yam's work suggests that organizations will begin to fail if the demands placed upon it exceed the complexity of the governance structure of that organization. When that governance structure concentrates power in one or a few people at the top, that means the demands placed on the structure can't be any more complex than one person can handle. In the case of a representative democracy, we are expecting a president—aided by advisors and Congress, of course—to ultimately make decisions in an environment that is far too complicated for any one person. Democracy as we know it is failing. "We cannot expect one individual to know how to respond to the challenges of the world today," Bar-Yam said. "So whether we talk about one candidate or another, the Democrats or Republicans, Clinton versus Trump. The real question ultimately is, will we be able to change the system?" "We've become fundamentally confused about what the decisions are, and what their consequences are. And we can't make a connection between them," he added. "And that's true about everybody, as well as about the decision-makers, the policymaker. They don't know what the effects will be of the decisions that they're making." Bar-Yam proposes a more laterally-organized system of governance in which tons of small teams specialize in certain policies, and then those teams work together to ultimately make decisions. "We end up with people who will say, 'I will do this, and things will be better.' And another person who will say, 'I will do this. And things will do better.' And we can't tell," he said. "Right now the danger is that we will choose strategies that will really cause a lot of destruction, before we've created the ability to make better decisions." When you vote Tuesday, don't vote for blowing up the system—Bar-Yam advocates for a gradual move to more lateral governance structures. But know that the person you're voting for will certainly be in over their head.

Loading New England Complex Systems Institute collaborators
Loading New England Complex Systems Institute collaborators