Part of a year-long series about complexity science by the Santa Fe Institute and The Christian Science Monitor. Read our other entries at breakthroughs.csmonitor.com.
During every US presidential primary season, we watch as the political fortunes of individual candidates rise and fall, seemingly without regard to whether a candidate has the skills, character, or ideological foundations to govern productively. To the casual observer, the process looks like unmitigated chaos.
Inevitably, this political circus results in nominees and, after the general election, a winner – someone who serves as our president for four years – even if we don’t all agree with the choice.
Messy, yes. But when you think about it, our political system is a remarkable outcome of human sociality. We intentionally create social chaos to achieve social order. What forces underlie this process?
Perhaps most important, does it – can it – result in wise choices?
How order emerges
The study of complex systems, like all of science, is a search for order. Traditionally, science seeks order by understanding the simplest parts of a system. How does a single gas particle behave given a certain temperature? Which gene in our DNA determines eye color? Scientists then try to develop theories that explain more general observations based on their detailed understanding of the individual parts.
Complex systems science is different. It seeks order by understanding how simple parts, interacting together and perhaps adapting to one another, create an entirely new whole. The collective outcomes of complex systems can be surprising because the parts often don’t add up as expected.
In his book “Wealth of Nations,” published in 1776, philosopher and economist Adam Smith noted how an individual selfishly seeking his own security “is in this, as in many other cases, led by an invisible hand to promote an end which was no part of his intention.”
Remarkably, while the study of economics has developed a sophisticated theoretical apparatus over the past 240 years, there are incredible gaps in economic theory that even today are filled by invoking “invisible hands” guiding the behaviors of economic agents.
Famously, the 2008 financial collapse – one of the most important economic events in the modern era – was widely unanticipated by economists. Equally as worrisome, once the crisis emerged, the prevailing theories yielded a dearth of prescriptive advice. It was as if a geologist happening upon the rim of the Grand Canyon was only able to proclaim: “Something happened here.”
The 2008 crisis embraced all of the seven deadly sins, from gluttonous banks to greedy mortgage brokers. But the biggest failure – our collective blind spot – was failing to appreciate the big, interconnected picture. While economists and policy makers were well equipped to understand and control the individual parts that contributed to the crisis, they were unable to comprehend how those parts added up to the whole.
In the case of the financial crisis, various positive feedback loops linked the different parts of the system, and the same forces that amplified the system on the way up (much to the delight and profit of all involved) accelerated its demise on the way down. The real danger was systemic.
Birds, brains, and bees
Over the last couple of decades, we have begun to develop new observations and techniques that allow us to understand how systems emerge from pieces. Good examples are often found in nature.
Flocks of starlings, known as murmurations, undulate in the sky like beautiful ghosts. Such seemingly coordinated collective behavior arises when individual birds follow a few simple rules: stay close, but not too close, to your neighbors and fly roughly in a similar direction and at a similar speed.
Likewise, the individual neurons in your brain respond in simple ways to signals from their local connections. There’s still a lot we don’t know about brains, but neurologists have demonstrated that decentralized mechanisms like signal averaging – essentially your neurons voting based on the preponderance of signals near them – set up a self-reinforcing feedback loop that probably has a lot to do with how we form thoughts.
For a wonderful example of decentralized decision-making in nature, consider a beehive. Queen bees, their royal titles notwithstanding, are little more than egg-laying machines. Indeed, no centralized hive governance system has ever been discovered.
At the heart of the complex system embodied by a beehive is the mystery of how a collection of many thousands of bees operates without apparent control, yet with an efficiency that would be the envy of most industries or governments.
The behavior of each bee in a colony is governed by a simple set of rules that, through interactions with the environment and the other bees, leads to an end that was no part of any individual bee’s intention, allowing the colony as a whole to thrive – as if by an invisible wing.
Put another way, bees are relatively dumb. Beehives are remarkably smart.
Making the best choice
One example of bees making smart decisions arises when they need to find a new home. Scout bees in the swarm are encoded with a simple exploratory behavior that motivates them to search for potential hive sites.
When a scout finds a potential site, she returns to the swarm and does a “waggle” dance – which looks a bit like human twerking – to advertize the new site’s location to the other scouts. Her dance will often motivate other scouts to check out the site and return to dance on their own.
Key to this process is a positive feedback loop in which the better the apparent quality of the new site to the scout, the longer she dances.
Over time, a variety of locations are explored, with the prospects of each site rising and falling as dancers come and go. Eventually, when a sufficient quorum of scouts forms at one of the locations, it becomes the final choice.
Remarkably, this process allows the bees to find a high-quality location in a timely manner without any kind of central authority, as it is only the local actions and observations of the individual scout bees that drive this system.
What’s most surprising is this: Both theoretical models and field experiments show that bees tend to make the best choice from among the available options. The positive feedback mechanism of scouting, communicating, and verifying in increasingly greater numbers causes seemingly better options to be explored more intensively. As long as the quorum is large enough, the vagaries of the initial search process diminish and a good final choice results.
Benefitting from buzz
Bees offer an interesting parallel to our political system, as well.
Candidates in presidential primaries are a lot like potential hive locations, and media attention, campaign donations, and even yard signs and bumper stickers constitute the political “dance.” As a candidate’s political fortunes rise or fall, he or she garners more or less attention and evaluation – essentially “buzz” – creating the necessary feedback loop.
As long as this feedback loop is tied roughly to the true quality of each candidate, as with the new beehive location, the apparent chaos of the system tends to result in a nicely ordered final choice.
Thus, while media attention driven by particular political ideologies or outsized monetary support may encumber the primary system, having multiple debates and requiring candidates to garner sufficient support from a series of delegate elections over time should allow the best candidate to emerge from among the pack.
Complex collective problems
Complexity science, a relatively recent arrival on the scientific landscape, offers a number of new ways for exploring what might be our last remaining scientific frontier: how unexpected system-wide phenomena emerge from parts.
We can, for example, use the ever-increasing power of the computer to form “agent-based” models of individual, computerized traders haggling in an artificial market to gain new insights into how prices arise (perhaps allowing us to wave goodbye to Mr. Smith’s invisible hand), or how bubbles form or markets collapse.
Understanding the structure of interconnected systems using network science allows us to analyze the various loan obligations made across the banking system and identify conditions in which a small number of failures can bring the entire system down.
Simple programs adapting to each other inside a computer give us insights into how cooperation can emerge in a social system, even when each individual would be better off by not cooperating.
Studying the complex social and agricultural rituals Balinese rice farmers have developed can give us insights into how they have been able to sustainably farm their rice terraces for centuries, even though water is chronically scarce enough to threaten a complete breakdown of the system.
Computational models showing how individual families’ (even mild) preferences to live in a neighborhood with families like them can quickly devolve into a highly segregated cityscape offer insights into wealth inequality, race relations, and urban planning.
We can study the apparent chaos of the beehive or anthill to gain a better understanding of the order that emerges in our governments – and when central control or decentralization makes the most sense.
Complexity is at the core of most of the major challenges confronting humanity – climate change, financial collapse, ecosystem survival, inequality, terrorism, and disease. If we understand complexity well enough, perhaps, the same complexity that creates these problems can help us choose a politician able enough to address them.
John H. Miller is a professor of economics and social science at Carnegie Mellon University and an external professor of the Santa Fe Institute. He is the author of the newly released book on complex systems: “A Crude Look at the Whole” (Basic Books).
Top photo: Republican presidential candidate and businessman Donald Trump speaks to supporters during a primary night rally, Tuesday, Feb. 9, 2016, in Manchester, N.H. (David Goldman/AP)