The history of science is largely the history of individual genius. From Galileo to Einstein, Isaac Newton to Charles Darwin, we tend to celebrate the breakthroughs achieved by a mind working by itself, seeing more reality than anyone has ever seen before.
It’s a romantic narrative. It’s also obsolete. As documented in a pair of Science papers by Stefan Wuchty, Benjamin Jones and Brian Uzzi, modern science is increasingly a team sport: more than 80 percent of science papers are now co-authored. These teams are also producing the most influential research, as papers with multiple authors are 6.3 times more likely to get at least 1000 citations. The era of the lone genius is over.
What’s causing the dramatic increase in scientific collaboration? One possibility is that the rise of teams is a response to the increasing complexity of modern science. To advance knowledge in the 21st century, one has to master an astonishing amount of information and experimental know-how; because we have discovered so much, it’s harder to discover something new. (In other words, the mysteries that remain often exceed the capabilities of the individual mind.) This means that the most important contributions now require collaboration, as people from different specialties work together to solve extremely difficult problems.
But this might not be the only reason scientists are working together more frequently. Another possibility is that the rise of teams is less about shifts in knowledge and more about the increasing ease of interacting with other researchers. It’s not about science getting hard. It’s about collaboration getting easy.
While it seems likely that both of these explanations are true—the trend is probably being driven by multiple factors—a new paper emphasizes the changes that have reduced the costs of academic collaboration. To do this, the economists Christian Catalini, Christian Fons-Rosen and Patrick Gaule looked at what happens to scientific teams after Southwest Airlines enters a metropolitan market. (On average, the entrance of Southwest leads to a roughly 20 percent reduction in fares and a 44 percent increase in passengers.) If these research partnerships are held back by practical obstacles—money, time, distance, etc.—then the arrival of Southwest should lead to a spike in teamwork.
That’s exactly what they found. According to the researchers, after Southwest begins a new route collaborations among scientists increase across every scientific discipline. (Physicists increase their collaborations by 26 percent, while biologists seem to really love cheap airfare: their collaborations increase by 85 percent.) To better understand these trends, and to rule out some possible confounds, Catalini et al. zoomed in on collaborations among chemists. They tracked the research produced by 819 pairs of chemists between 1993 and 2012. Once again, they found that the entry of Southwest into a new market leads to an approximately 30 percent spike in collaboration among chemists living near the new routes. What’s more, this trend towards teamwork showed no signs of existing before the arrival of the low-cost airline.
At first glance, it seems likely that these new collaborations triggered by Southwest will produce research of lower quality. After all, the fact that the scientists waited to work together until airfares were slightly cheaper suggests that they didn’t think their new partnership would create a lot of value. (A really enticing collaboration should have been worth a more expensive flight, especially since the arrival of Southwest didn’t significantly increase the number of direct routes.) But that isn’t what Catalini et al. found. Instead, they discovered that Southwest’s entry into a market led to an increase in higher quality publications, at least as measured by the number of citations. Taken together, these results suggest that cheaper air travel is not only redrawing the map of scientific collaboration, but fundamentally improving the quality of research.
There is one last fascinating implication of this dataset. The spread of Southwest paralleled the rise of the Internet, as it became far easier to communicate and collaborate using digital tools, such as email and Skype. In theory, these virtual interactions should make face-to-face conversations unnecessary. Why put up with the hassle of air travel when there’s Facetime? Why meet in person when there’s Google Docs? The Death of Distance and all that.
But this new paper is a reminder that face-to-face interactions are still uniquely valuable. I’ve written before about the research of Isaac Kohane, a professor at Harvard Medical School. A few years ago, he published a study that looked at the influence physical proximity on the quality of the research. He analyzed more than thirty-five thousand peer-reviewed papers, mapping the precise location of co-authors. Geography turned out to be a crucial variable: when coauthors were closer together, their papers tended to be of significantly higher quality. The best research was consistently produced when scientists were located within ten meters of each other, while the least cited papers tended to emerge from collaborators who were a kilometer or more apart.
Even in the 21st century, the best way to work together is to be together. The digital world is full of collaborative tools, but these tools are still not a substitute for meetings that take place in person.* That’s why we get on a plane.
Never change Southwest.
Catalini, Christian, Christian Fons-Rosen, and Patrick Gaulé. "Did cheaper flights change the geography of scientific collaboration?" SSRN Working Paper (2016).
* Consider a study that looked at the spread of Bitnet, a precursor to the internet. As one might expect, the computer network significantly increased collaboration among electrical engineers at connected universities. However, the boost in collaboration was far larger among engineers who were within driving distance of each other. Yet more evidence for the power of in-person interactions comes from a 2015 paper by Catalini, which looked at the relocation of scientists following the removal of asbestos from Paris Jussieu, the largest science university in France. He found that science labs that had been randomly relocated in the same area were 3.4 to 5 times more likely to collaborate. Meat space matters.