Fifteen Eighty Four

Why do we see the behaviors that we do in the world? This question has challenged many notable thinkers, including Darwin, Saussure, Wittgenstein, Lévi-Strauss, Durkheim, and many other past and recent thinkers. Their conclusions identified how things in the world – from species to thoughts to culture – rely on the interconnectivity of the systems in which they arise. Behaviors are a consequence of the environments in which they emerge and the historical trajectories that have brought them about.

Behavioral Network Science is my attempt to understand behavior in relation to the structures that give rise to it. That is, it attempts to explain why structure matters. Tackling topics like the emergence of language, memory in old age, creativity, conflict, and conspiracies, the book tries to make visible the structure that lies beneath the behaviors that make up our everyday lives. This blog is largely taken from the introduction to the book and attempts to make the argument, with a few choice examples, that structure is so fundamental to behavior, both inside and outside our heads, that we overlook it at our own peril.  Though Behavioral Network Science will bring interested readers up to speed on a wide variety of topics in psychology, computer science, politics, and language, it will also bring the dedicated reader face-to-face with the formalisms of network science and how they apply. Having said all of that, the premise on which the book is based is not new.

The biologist Theodosius Dobzhansky expressed this when he said, “Nothing in biology makes sense except in the light of evolution.” The linguist John Firth echoed it, saying, “You shall know a word by the company it keeps.” These are structural claims about the origins of life and meaning. Organisms arise from a marriage of ancestry and environment. The meanings of words depend on their relationships with other words. People aggregate into like-minded groups. Beliefs nestle among one another like a huddle of penguins.

One tool bringing this structural vision of life and meaning into focus is network science. By describing relationships that bind systems together, network science gives quantitative precision to these earlier ideas. As a ruler measures length, allowing us to compare human height with the Burj Khalifa, network science measures structure, letting us compare human relationship networks with the organization of their environments, the patterns of brain activation, and – more enchantingly – the architecture of thought. When we stop seeing isolated things and start focusing on relationships, we invite a structural view of behavior that can be compared across disciplines.

This cross-disciplinary insight is a catalyst behind complex systems science, which attempts to identify general principles that apply across domains. By “general principles,” I mean principles abstracted away from the particulars of the environment. When someone describes a scurrying procession of driver ants, searching for prey on the forest floor in the Gold Coast of Ghana, with the rejoicing voices of a group of improvisational singers in a nearby church, we have something that approximates literature. Understanding how the ants and the singers solve a similar coordination problem is a view from complex systems.

Both ants and singers navigate structured spaces – whether among the forest’s fractal underbrush or the conventions of song. For both, the quality of their navigation depends on the structure of their communication: who communicates with whom, when, and by what means. The structural features of communication they share apply to 1970s jazz fusion as well as they do to a murmur of starlings, the challenges of international politics, or the way people use what they know to learn what they don’t. This shared structure invites us to think about behavior using the tools of network science.  When we do, we have behavioral network science.

Viewing behavior in this way, phenomena become symptoms of the structural processes that govern them. The goal of the behavioral network scientist is to articulate the structure and processes of this government.

Why should we care about structure?

We should care about structure because it helps us understand how and why systems work like they do. That, in turn, helps us to anticipate what they are likely to do next – and what, if anything, we should do about it. For all the grand unified generality of complex systems, behavioral network science shines most brightly when it tells us about specific things. Here are a few examples.

Structure can tell us what kinds of systems we are dealing with. When Valdis Krebs mapped the network structure of the 9/11 terrorist cells – identifying each individual and the relationships among them – he observed something peculiar for a social network. The 19 hijackers had sparsely interacted. This is rare in social networks. Our friends’ friends tend to become our friends, close-knit groups form from like-minded individuals, and coordinated groups need coordinated communication. On the face of it, the structure of the terrorist cells lacked these features. But this was by design. Sparsely connected social networks make sense for rogue networks that must escape detection. Many criminal networks invoke similar behavioral immune systems, insulating themselves from compromise by limiting what individuals know about one another and how often they interact. This makes it harder for authorities to identify and compromise them.

Structure also tells us about the kinds of individuals needed to maintain social health. Social health is critical for winter-over crews at the Amundsen-Scott South Pole Station in the Antarctic, who are effectively trapped together in a cage for nine months. Jeffrey Johnson and colleagues found that the social health of the group went hand in hand with its social structure, depending critically on specific group roles. Certain people were linchpins of community health. One was the leader – but not just any leader. Successful leaders were expressive leaders focused on social cohesion. Expressive leaders know what needs to be done and get people to care enough about one another to do it cheerfully. Another key role was the ‘positive deviant,’ or the clown. Through their ability to defy social conventions, the clown acts as a broker between various teams (such as nonscientists and scientists), bringing individuals together who would otherwise remain apart.

Structure also helps predict where certain events will happen. Consider Juárez, on the USA-Mexico border, the murder capital of the world during the early 2000s. What makes Juárez so dangerous? We often explain such things with exhausted stereotypes or broad explanations like “drug cartels.” The latter is half right, but there are drug cartels in many peaceful places. What differs about Juárez is almost entirely structural – it is what it is because of its relationship with other places. According to Tom Wainwright, Juárez is a critical gateway for drug trafficking between South and North America. Those who control Juárez control access to a large part of the North American drug market. Juárez is a place with what network scientists call high betweenness: to go from one place to another, one must pass through it. During the 1990s, control over Juárez was contested, and the disputes that arose reflected the value of its betweenness.

These examples demonstrate how understanding network science principles can guide behavioral intuition. When we see certain behaviors, we can better hypothesize what gave rise to them. Just as a knowledge of Latin can help us infer the meaning of a new word, a knowledge of structure helps us make sense of the world’s rich behavioral complexity.

These examples also show how behavior brings network science to life. When we see an abstract network principle embodied in issues we already understand, the principle becomes approachable. It starts to become something we can recognize.

Structure and Process

Like a map, structure is a simplification. But when our understanding of structure works at its best, it helps us think about how structure and behavior interact. For sparsely connected terrorist cells, we can see how a disconnected structure prevents disruption, like getting one individual to rat out the others. But if the process were different – if police took down criminals like lions single out a stray from the herd – then the structure would be different. Criminals would crowd together like gazelles on the African savanna. Similarly, our mental health benefits from social relationships, as if it were contagious. Knowing this, it becomes clear why people who act as social connective tissue keep communities healthy. Finally, Juárez’s structural position matters because of where drugs originate and how they move. Legalizing drugs or otherwise changing the conditions by which they are trafficked would change the processes driving the drug market and change Juárez as a result. Structure lives and dies by its processes.

There are countless examples. US Route 66 is a structural bygone that fell into decline not because we changed physical space but because we changed how we moved through it. Similarly, our social networks have changed because online environments make it easier to find people who share our obsessions. If only the mind had a recent upgrade in it search technology.  The processes we use to search memory mean the aging mind can know more and be harder to search at the same time, giving rise to symptoms of so-called age-related cognitive decline. Behavior is full of examples like these. Creativity, child learning, opinion dynamics, conspiracy beliefs, and more result from structure and process. Structure helps us understand processes better, and vice versa. There are many books on complex systems.  They are invaluable illuminations on how reality’s rich abstractions hold the details of the world together. The emergence of life, the self-organization of a beehive, the interconnectedness of an urban center. There are many books on behavior and psychology as well, explaining why we do what we do in the flickering light of theory. The associative mind of a developing child, the game-theoretic rules of international conflict. Behavioral Network Science attempts to take these both as seriously as it can with the goal of making them each shine a little brighter.

Title: Behavioral Network Science

Author: Thomas T. Hills

ISBN: 9781108793339