Range - David Epstein Page 0,114

goes across all those markets. So we specialized products, we specialized regulation, and the question is, ‘Who looks across those markets?’ The specialized approach to regulation missed systemic issues.”

In 2015, Casadevall showed that biomedical research funding rose exponentially over a recent thirty-five-year period, while discovery slowed down. Life expectancy in countries at the biomedical cutting edge, like the United Kingdom and the United States, recently declined after decades of improvement. The flu annually kills hundreds of thousands of people worldwide while humanity fights it with a cumbersomely produced vaccine from the 1940s. Casadevall’s mother is ninety-three, and on five medications that were available when he was a medical resident in the 1980s. “Two of them are older than I am,” he said, and two others are barely younger. “I cannot believe we can’t do better.” He paused for a moment, tilted his head, and leaned forward. “If you write an interdisciplinary grant proposal, it goes to people who are really, really specialized in A or B, and maybe if you’re lucky they have the capacity to see the connections at the interface of A and B,” he told me. “Everyone acknowledges that great progress is made at the interface, but who is there to defend the interface?”

The interface between specialties, and between creators with disparate backgrounds, has been studied, and it is worth defending.

* * *

• • •

When Northwestern and Stanford researchers analyzed the networks that give rise to creative triumph, they found what they deemed a “universal” setup. Whether they looked at research groups in economics or ecology, or the teams that write, compose, and produce Broadway musicals, thriving ecosystems had porous boundaries between teams.

In professional networks that acted as fertile soil for successful groups, individuals moved easily among teams, crossing organizational and disciplinary boundaries and finding new collaborators. Networks that spawned unsuccessful teams, conversely, were broken into small, isolated clusters in which the same people collaborated over and over. Efficient and comfortable, perhaps, but apparently not a creative engine. “The entire network looks different when you compare a successful team with an unsuccessful team,” according to Luís A. Nunes Amaral, a Northwestern physicist who studies networks. Amaral’s remark does not compare individual teams, but rather the larger ecosystems that foster the formation of successful teams.

The commercial fate of Broadway during any particular era, be it unusually prosperous or exceptionally flop-ridden, had less to do with specific famous names and more to do with whether collaborators mixed and matched vibrantly. The 1920s featured dozens of shows with Cole Porter, Irving Berlin, George Gershwin, Rodgers and Hammerstein (albeit not yet in collaboration), and also an unusually high overall flop rate of 90 percent for new shows. It was an era of stagnant teams, rife with repeat collaborations and scant boundary crossing.

New collaborations allow creators “to take ideas that are conventions in one area and bring them into a new area, where they’re suddenly seen as invention,” said sociologist Brian Uzzi, Amaral’s collaborator. Human creativity, he said, is basically an “import/export business of ideas.”

Uzzi documented an import/export trend that began in both the physical and social sciences in the 1970s, pre-internet: more successful teams tended to have more far-flung members. Teams that included members from different institutions were more likely to be successful than those that did not, and teams that included members based in different countries had an advantage as well.

Consistent with the import/export model, scientists who have worked abroad—whether or not they returned—are more likely to make a greater scientific impact than those who have not. The economists who documented that trend suggested one reason could be migrants’ “arbitrage” opportunities, the chance to take an idea from one market and bring it to another where it is more rare and valued.* It echoes Oliver Smithies’s advice to bring new skills to an old problem, or a new problem to old skills. The atypical combination of typical forms—say, hip-hop, a Broadway musical, and American historical biography—is not a strategy fluke of showbiz.

Uzzi and a team analyzed eighteen million papers from a variety of scientific domains to see whether atypical knowledge combinations mattered. If a particular paper cited other areas of research that rarely, if ever, appeared together, then it was classified as having used an atypical combination of knowledge. Most papers relied purely on conventional combinations of previous knowledge. That is, they cited work from other journals that often appeared together in other studies’ lists of references. The “hit” papers, those that over the next decade were used by