Psych 101 is an occasional series on classic psychology research and how it informs the way we understand ourselves today.
Half a century ago, a wheat farmer in Kansas received in the mail a brown folder containing a set of instructions and the name of an assigned target: a Boston divinity school student named Alice.
Just four days later and hundreds of miles away, Alice was on a sidewalk in Cambridge when something surprising happened.
“Alice,” said one of her instructors, approaching her and holding the same brown folder. “This is for you.”
The wheat farmer had followed his instructions. One of hundreds of participants in an ambitious scientific experiment, he had been asked to try to convey the folder to Alice by giving it to someone in his social circle who might be more likely to know her, who then took up the same task. The wheat farmer had given the package to an Episcopalian minister in his hometown, who then mailed it to a colleague in Boston, where it soon reached its target.
The resulting connection — a rapidly created chain of acquaintances joining two perfect strangers in just a few links — was the first indication that Stanley Milgram, the social psychologist who designed the experiment, was on his way to a moving discovery: It really is a small world after all. It’s a finding that feels just as relevant now as it was then, and it’s one worth revisiting in what seems like an increasingly divided era.
It’s known as the small-world experiment. In it, packages were sent to hundreds of participants like the wheat farmer, as Milgram tried to determine just how many degrees of separation exist between any two people. In the 51 years since he published his results in 1967 (the same year he took over the social psychology doctoral program at City University of New York), the answer Milgram came up with — six — has become a commonplace truism and a Kevin Bacon–flavored parlor game. “It was really the first thing to experimentally demonstrate a phenomenon which is one of the most important properties of the social network of the world, which is that we’re all just a few steps from each other,” said Jon Kleinberg, a Cornell University computer scientist who studies networks.
Milgram’s experiment also helped to launch the field of network theory, leading to insights into other important features of an evermore connected world. Today, Milgram’s study can help us understand how diseases can tear through a continent in a matter of days, how a seemingly isolated financial tremor can send markets tumbling across the world, and how fake news from Russian trolls can go viral and transform an American election overnight.
Stanley Milgram, who died in 1984, is more widely remembered for his controversial obedience experiment, in which participants were ordered to administer what they believed were increasingly intense electric shocks to people who were begging them to stop. The 1961 experiment found a majority of participants would follow orders until the maximum voltage was reached even as the people receiving the “shocks” yelled and protested and finally went silent.
Milgram was already notorious for this experiment when he arrived in New York City in the late ’60s and he was an important hire for CUNY, said Herbert Saltzstein, a psychology professor there who overlapped with Milgram. “This was a big deal for them, absolutely,” Saltzstein said. Milgram’s presence may have raised the profile of the department and attracted students, Saltzstein said. “He was probably the best known social psychologist at CUNY, one of the best known in the country, maybe in the world.”
Milgram had a knack for designing experiments to upend conventional wisdom. Few suspected so many people would willingly electrocute someone, and few predicted the small-world experiment would determine that fewer than ten degrees separated pairs of random Americans. When Milgram asked people beforehand to predict what number he would come up with, the answers ranged from 100 to 1000 to “it’s impossible.”
“I think the reason it made a splash is that it was so different from what people thought,” said Mark Granovetter, a professor at Stanford University and the author of what’s believed to be the world’s most cited sociology paper, on the spread of information in social networks via weak ties. “I think Milgram’s was the first piece of research to capture the popular and scientific imagination at the same time.”
Some thinkers, however, had been quietly wondering if apparently unconnected people might in fact be linked. The idea of six degrees of separation is sometimes traced to a 1929 essay by the Hungarian writer Frigyes Karinthy. And Milgram’s work was preceded by some calculations by political scientist Ithiel de Sola Pool and mathematician Manfred Kochen who in the 1950s estimated a greater than 50-percent chance that any two people could be linked by two intermediate acquaintances.
But Milgram was the first to test the small-world idea with a real-world experiment. Funded by a $680 grant, Milgram mailed out brown folders, first to participants in Kansas and, in a later experiment, in Nebraska. Each package contained the name of and basic information about a target person, a roster that participants were asked to add their names to as they came into possession of the folder, and a packet of postcards to be returned to Milgram at each step so he could track folder progress. Participants were asked to give the folder to someone with whom they were on a first-name basis.
The attrition rate was high: Lots of people simply didn’t follow through and forward the package. But the folders that reached their target did so in between two and ten steps, with a median of five intermediaries.
Milgram wrote up his small-world experiment results in the debut issue of Psychology Today, in May 1967. Accompanied by folksy illustrations, the article began with an anecdote of two strangers finding they have a friend in common, and ended with a tidy kumbaya summation: “While many studies in social science show how the individual is alienated and cut off from the rest of society, this study demonstrates that, in some sense, we are all bound together in a tightly knit social fabric.”
It is compelling to think that we are all joined as one human family in a way that’s not necessarily obvious in our day-to-day lives. And so it’s not necessarily surprising that a study that seemed to prove it, presented in a glossy magazine and written for mass appeal, would have a lasting impact. (Milgram also published his findings in a more scholarly fashion two years later.)
But what really launched the small-world idea into the mainstream was the 1990 play — and subsequent film — Six Degrees of Separation, by John Guare. “I read somewhere that everybody on this planet is separated by only six other people,” Stockard Channing says in the 1993 movie. “Six degrees of separation between us and everyone else on this planet. The president of the United States, a gondolier in Venice, just fill in the names.”
Guare told the New York Times that he remembered reading about Milgram’s experiment in Psychology Today and although Milgram never mentioned the phrase “six degrees of separation,” Guare’s pithy summation of his findings stuck … to Kevin Bacon.
In 1994, three college students at Albright College in Pennsylvania were watching movies in a snowstorm when they had a moment of clarity: actor Kevin Bacon is the center of the entertainment universe. Bacon had appeared in so many films that he could be connected to seemingly any actor in fewer than six steps. Anthony Hopkins, for instance, has a “Bacon number” of two, since he was in Fracture with Ryan Gosling, who co-starred in Crazy, Stupid, Love with Kevin Bacon.
The three students’ dorm-room epiphany earned them a spot on The Jon Stewart Show on MTV, and they even created a board game based on the concept. Kevin Bacon initially resisted becoming a meme, but eventually embraced it, starting a charity called Six Degrees. (It’s worth noting that Bacon, despite his reputation, is NOT in fact the center of the Hollywood universe. That honor, according to an analysis by the website Oracle Of Bacon, belongs to Eric Roberts, Julia’s brother and Emma’s father.)
By the late 1990s, the idea of six degrees of separation had permeated the culture, and it found its way back into the world of science.
“I was on the phone with my dad one night,” said Duncan Watts, a mathematician and network theorist and author of Six Degrees: The Science of a Connected Age. “And he said, ‘Did you know that everyone is just six handshakes away from the president of the United States?’ I said, ‘That sounds like a math problem.’”
Watts, along with mathematician Steven Strogatz, began looking at the way networks function, and came up with a model that explains the small-world phenomenon. They used power grids, the neural network of a nematode worm, and the connections between Hollywood actors to demonstrate in 1998 that just a few long-range connections added to any network can shrink the whole system.
For instance, Strogatz has lots of local connections in Ithaca where he’s a professor at Cornell University. But he also plays chess on the internet and has gotten to know a player in Holland. “We’re now on a first-name basis. So now I’m connected to Holland,” Strogatz said. Just a single long-range connection exponentially reduces the world: In just a few more degrees of separation, everyone in New York is linked to everyone in Holland.
“It’s just in the nature of these networks that they have to be small,” he said. “Just the tiniest sprinkling of hubs or shortcuts makes the world small.”
Milgram’s small-world experiment has not been without its detractors. Most notably, Judith Kleinfeld, a psychology professor at the University of Fairbanks in Alaska, has objected that Milgram cherry-picked both his participants and his results, failed to show that the small-world phenomenon transcends race and class, and ignored the fact that many of the chain-letters in his experiment never reached the intended target.
But Watts said that while there may have been flaws in Milgram’s study, his results have been confirmed by subsequent studies. Watts did a version of Milgram’s small-world study via email in the early 2000s with 60,000 participants and 18 different targets drawn from an array of professions in 13 countries (a Norwegian veterinarian, an Australian policeman, etc.) The result: Participants were connected in a median of five to seven steps.
“We’ve done a lot of checking. We have a lot more data. The result as stated by Milgram is solid,” Watts said.
In a subsequent 2009 meta-analysis, Watts looked at 162,328 “small-world” message chains and found that roughly half could be completed in six or seven steps. And in 2011, a study by Facebook found that its users were separated by an average of 4.57 degrees.
There’s sometimes confusion between studies about whether “degrees” refers to intermediary acquaintances or to the links between them, Watts said. And different results are obtained by analyzing a data set and finding the shortest routes using perfect information, versus asking people to muddle through and find their own routes by instinct — a distinction referred to as a topological versus algorithmic approach.
But it doesn’t matter too much what the exact result is; what’s most significant is that all the studies end up with a remarkably small number of degrees of separation, Watts said. “The point is that it’s not 900,000. Or even a thousand, or a hundred. It’s this number less than ten and bigger than two.”
While Milgram had concluded his small-world study by invoking a sense of universal togetherness, network theorists have since pointed out that there are downsides to living in such a connected world — if we don’t stay mindful of the fact that we’re all in this together.
Consider the AIDS epidemic, Watts said. “Why did people not care about it? It was far away.” For many Americans, AIDS was something happening in Africa or to gay men or intravenous drug users, and people failed to appreciate it until it became impossible to ignore.
A danger of a small world, Watts said, is that even though we’re now globally connected, we evolved in small tribes and are thus doomed to fail to fully consider anything beyond our immediate social circle. We care about our friends, and we kind of care about our friends’ friends, but “anything more than two degrees is just some random person,” he said.
“As humans, we’re not really good at appreciating the consequences of exponential growth,” said Kleinberg. “We’re not aware that the actions we take are rippling outward with a kind of frightening velocity.”
“Think about fake news. Think about influencing our election. Think about what’s called today the weaponizing of social media,” said Strogatz. People can now remotely sabotage centrifuges half a world away, he added. “We’re playing this very wild social experiment with ourselves where we have now become this global network. We don’t know what we’re doing.”
People may be oblivious to it’s full impact in their day-to-day lives, “but the network is still there. And if the network is only six degrees, you need to pay attention,” Watts said. “That’s ultimately Milgram’s insight.”