Rooted in Connection

Nicholas A. Christakis: I’ve spent the last twenty-five years studying complex social networks. They are intrinsically an interdependent structure. Each of us inherits our relatives, picks our friends and coworkers, and all of those people in turn are picking or inheriting other people as connections. We therefore proceed to assemble ourselves into this large, interdependent, beautiful, baroque, organic structure known as a social network.  

And your fate and life depends, to some material extent, on where you are in such a network. Anyone can understand this, nowadays, because of our recent experience with the pandemic. Whether you get COVID depends, for example, not just on your own actions, but the actions of your friends and the actions of your friends’ friends and the actions of your friends’ friends’ friends.  

Our fates are connected to the fates of others, including people we don’t even know. And this interdependence can be understood using a variety of scientific tools from biology, psychology, sociology, physics, and mathematics, because our social networks obey very particular (and scientifically intelligible) rules.  

Emergence is the idea that collections of things can have properties that are not intrinsic to the thing itself. The simplest example is carbon. You can take carbon atoms and combine them one way, and you get graphite, which is soft and dark. Or you can take the same carbon atoms and combine them another way, and you get diamond, which is hard and clear. There are two key ideas here. First, these properties of softness and darkness and hardness and clearness are not properties of the carbon atoms. They are properties of the collection of carbon atoms. And second, which properties you get depends on how you connect the carbon atoms to each other.  

That’s emergence—that these new properties arise because of complexity, because of interdependence, because of interconnection. It’s a magnificent phenomenon in nature, not just restricted to chemistry or biology, but also relevant to our social lives.  

Over the last twenty years in my lab, we’ve done a lot of studies of this, including large-scale experiments in online settings, as well as in public health settings around the world. Initially, we did observational research. We studied, using mathematical and statistical tools, naturally occurring systems. We found, for example, that your body size depends, to some extent, not only on your own choices and actions, but also the choices and actions and the body size of your friends, and your friends’ friends, and your friends’ friends’ friends. Also, we found that how happy you are can depend on where you are in the network and the state of happiness of other people, including not only the people you are directly connected to, but even people you don’t know and are only indirectly connected to.  

Those are examples of things that can spread by a process of social contagion within social networks. But it’s very difficult to know for sure that that is what’s happening when you use observational data. To really be certain, you have to do experiments.  

Beginning in 2010, we started doing experiments with tens of thousands of people online and in settings around the world, from Honduras to India to Uganda. For example, you can bring people into a laboratory setting, randomly assign them to groups of four strangers, and have them play something known as a public goods game. In such a game, each person has some money. If I give one dollar to the group of four, that one dollar is doubled and then it’s divided among the four people. The group as a whole has gained two dollars from my one dollar contribution, but then that two dollars is divided four ways and comes back to each person, so I only get back fifty cents. The group benefits, but I pay a price.  

Most people might expect that nobody will contribute. From an individual, selfish point of view, the smartest thing to do is give nothing and hope that everyone else contributes. Of course, if nobody contributes, the group collapses. But the best outcome is if everyone contributes maximally. And, indeed, most people are kind and do contribute, and they have some realistic expectation that others will reciprocate.  

In that first experiment, people played this public goods game across multiple rounds. A bell would ring and they would play with new strangers, and then ring again and they would play with new strangers. We were able to show that if Tom was kind to Dick in round one, Dick would be kind to Harry in round two, Harry would be kind to Betty in round three, and Betty would be kind to Susie in round four. So how Betty treated Susie depended on how Tom treated Dick, even though neither Betty nor Susie ever saw or interacted with Tom or Dick. We were able to experimentally demonstrate the social contagion of goodness, spreading through a network.  

In other experiments, we mapped the social networks of 24,000 people living in 176 isolated villages in Honduras, who we followed for eight years. We randomly chose certain people in the network, according to certain mathematical algorithms, and we taught them things relevant to maternal and child health, like breastfeeding, vaccination, and so on. We were able to artificially create cascades of the contagion of this desirable knowledge in these isolated villages, improving the health of everyone in the village, not just the subset of people selected to get the knowledge directly. We have also done similar work in India. 

In short, both observationally and experimentally, over the last twenty years, we have been able to show that social contagion is an important factor in our lives. How you act, what you know, and how you feel depends on your embeddedness in the vast fabric of humanity. 

In my laboratory, we’ve been able to show that genes don’t just affect the structure and function of our bodies and minds, but also the structure and function of our societies. In other words, our genes play a role in how we interconnect with each other. For example, it’s obvious that some people are born shy, and some people are born gregarious. People vary in their taste for friendship. But we have also shown that whether your friends are friends with each other depends not just on their genes but on your genes.  

If Tom, Dick, and Harry are in a room, whether Dick is friends with Harry, depends not just on Dick’s genes or on Harry’s genes, but on Tom’s genes. That’s really weird. And the reason is that people vary in their tendency to introduce their friends to each other. Natural selection has shaped our propensity to interconnect the world around us.  

Our genes oblige us to create these social worlds. We make social networks very deliberately and following very particular mathematical and biological rules. 

We have not only shown this using molecular biology but also anthropology. For example, the Hadza hunter-gatherers of Tanzania live like all humans did at the end of the Pleistocene. We found that their networks mathematically and structurally look just like ours. Despite the fact that in the intervening ten thousand years, humans have developed agriculture, cities, and telephones, the structure of social networks of people who live in the modernized world is indistinguishable from the structure the Hadza make.  

What’s amazing is that other scientists have also mapped the social networks of elephants and dolphins and other large social mammals. And the networks that elephants make are mathematically very similar to ours. Our last common ancestor with elephants is ninety million years ago. So elephants evolved the same structure of social networks independently of us.  

Many people have said about our work that it delivers a whack to free will, which is kind of true in a way. We have free will, in my view, but we may have less than we realize, once you take network effects seriously.  

First of all, the play of chance in our lives is huge. You have no control over your birth order or your parents’ educational status or where you were born. All of these things have huge impacts on what happens to you in your life and the choices you are free to make.  

Yet returning to network effects, on the one hand, our work adds to the list of things that constrain the range of opportunity for free will, because I’m suggesting, in an empirically documented way, that things like your happiness and your altruism and your childcare practices depend on people around you, even people you don’t even know.  

However, at the same time that our embeddedness in networks compromises your ability to act freely, it also ironically increases the importance of your acting freely. Because every time you make a positive change in your life—when you’re nicer to other people or when you quit smoking—you create these ripple effects around you that don’t just affect your friends, but also your friends’ friends and your friends’ friends’ friends. You could affect hundreds of other people when you choose to make a change.  

One of the first principles of social networks is that they magnify whatever they are seeded with. But networks are agnostic. They will magnify fascism and violence and hatred and misinformation. But also, equally, they will magnify love and kindness and happiness and ideas. Networks don’t on their own give rise to these phenomena. Some external force has to get the contagion going.  

I argue in my book Blueprint: The Evolutionary Origins of a Good Society that the benefits of a connected life outweigh the costs. But it’s important not to assume that there aren’t any costs, because mob action (such as lynchings and witch burnings) arises because of interconnection. Or market bubbles, which can arise because of interconnection and misinformation.  

I would say that social connection overall is good and is a net benefit, but it’s not an unmitigated good. There are costs that come with social connection as well.  

I think it’s important to be reflective and to think about how one is behaving and how one has come to believe certain things. Why do I actually hold this opinion? How can I be certain that this is true?  Why am I doing what I am doing? One of the easiest things you can do, in this regard, is to seek out people you disagree with and talk to them without judgment. A good way to test your beliefs is to talk to someone who has the opposite belief.  

We’ve done a lot of experiments in my lab showing that our scientific ideas regarding social contagion can be deliberately exploited to enhance awareness of beneficial public health practices; facilitate the flow of accurate information online; decrease misinformation, prejudice, and bigotry in social communications; and facilitate creativity in groups. There are ways pragmatically to exploit this knowledge to make the world better.