by Matthew Syed
However, the Geniuses are not very social. They only have one friend in their network they can learn from. The Networkers, on the other hand, have ten friends, making them ten times more social. Now, after everyone has tried to invent the bow and arrow for themselves, and then tried to learn from their friends, from whom they have, say, a 50 per cent chance of learning in each encounter, in which population will the innovation be more common?
The answer is counterintuitive. Among the Geniuses, only 18 per cent of people will have the innovation. Half of those will have figured it out on their own. Among the Networkers, on the other hand, 99.9 per cent will have the innovation. Only 0.1 per cent will have solved it on their own, but the rest will have learned it from friends. And each of these will now have an opportunity to improve on the innovation, transmitting insights back into their networks.39 The result is clear – and is corroborated by field data, lab experiments and dozens of historical examples. As Henrich puts it:
‘If you want to have cool technology, it is better to be social than smart.’40
V
Route 128 is a highway that starts at Norfolk County in the south, winds up past the western suburbs of Boston, before culminating, right on the coast, by Gloucester, a fishing town that was the setting for Rudyard Kipling’s famous book Captains Courageous.
When Jonathan Richman wrote ‘Roadrunner’, his song about Route 128, listed by Rolling Stone magazine as one of the greatest 500 songs ever recorded, the highway encompassed what many believed was an enduring economic miracle. In 1975, the technology complex employed tens of thousands of workers, and boasted six of the ten largest tech companies in the world.41 Wang Laboratories, Prime and Data General were giants of the sector. At its height, Digital Equipment Corporation boasted a workforce of 140,000, the second biggest employer in the state. The western section of the route was dubbed America’s Technology Highway. Time magazine called it ‘the Massachusetts Miracle’.
The Santa Clara Valley, on the other hand, was an agricultural region over three thousand miles away on the west coast of America, largely devoted to apricot farming. These were juicy and fragrant, but they were rather a long way from chips and semiconductors. Most of the local industry was based upon small-scale food processing and distribution. One historian said: ‘there wasn’t much going on’. The Santa Clara region started to change in 1956 when William Shockley, a physicist and inventor, moved from an unsuccessful stint at Raytheon, a Massachusetts firm with an interest in transistors, to Mountain View, a small town towards the southern end of the peninsula. Over time, this led to a growing concentration of firms in the Valley, including the Fairchild Semiconductor Company.
By the 1970s, the Santa Clara Valley had spawned a soubriquet of its own, ‘Silicon Valley’, but it remained very much in the shadow of the Massachusetts miracle. The Boston firms had the classic economic advantages. Land and office space costs were significantly lower, as were the wages and salaries of workers, engineers and managers.42 There were other differences, too. The Boston firms were buttoned-up. They wore jackets and ties. The Silicon Valley rebels were more laid-back, preferring jeans and T-shirts. They had different ways of talking, and different terminology. Yet, these were the superficial differences. The key contrast consisted in the structure of networks, and the dynamics of information spillover. And this would prove utterly decisive.
The Route 128 firms had scale. They made chips and boards, monitors and frames, all internally. They even made disk drives. This vertical integration made sense economically. It meant that they had impressive efficiency in production. But this integration had another, less remarked upon consequence (a consequence it didn’t need to have). These large firms became socially isolated. Gordon Bell, a vice president of Digital Equipment Corporation, said: ‘DEC was a large entity that operated as an island in the regional economy.’ Glenn Rifkin and George Harrar, biographers of Ken Olsen, a co-founder of DEC, described the company as ‘a sociological unit, a world unto itself’.43 AnnaLee Saxenian, a sociologist who covered the tech wars in her classic book Regional Advantage, writes: ‘Route 128’s enterprises adopted autarkic practices.’
As the firms became isolated, they also became fiercely proprietorial. Wang hired private detectives to protect its ideas and property. People socialised only with people from their own firm. There were precious few forums or conferences that brought engineers together. ‘Practices of secrecy govern relations between firms and their customers, suppliers and competitors,’ Saxenian writes.44 Another said: ‘The walls got thicker and thicker, and higher and higher.’
The desire for secrecy made sense, in its own terms. The executives didn’t want other firms stealing their ideas. But it also represented a trade-off, unspoken but profound. By severing their engineers from the broader network, they inadvertently stifled the interplay of diverse insights, fusing, recombining, jumping forward in unpredictable ways: the complex dance of innovation. Route 128, then, was characterised by what network theorists call a ‘vertical’ dynamic. Ideas flowed within these hierarchical organisations, but not outwards. ‘Information on technologies remained trapped within the boundaries of individual corporations, rather than diffusing to other local firms and entrepreneurs . . .’ Saxenian wrote. There was very little horizontal transmission.
You could even sense the social isolation in the physical terrain along the highway, the way that the indigenous companies distanced themselves from each other. ‘The technology companies were scattered widely along the corridor and increasingly along the outer band . . . with miles of forest, lakes, and highway separating them. The Route 128 region was so expansive that DEC began to use helicopters to link its widely dispersed facilities.’45
On the surface, at least, Silicon Valley seemed less suited to the high-tech sector. The region didn’t enjoy tax benefits to help them catch up with Route 128, nor did they have additional state support in, say, defence spending. And, as already noted, costs were higher in land, office space and wages. And yet Silicon Valley had something more powerful, an ingredient that rarely finds its way into conventional economic textbooks. You get a sense of this ingredient by reading Tom Wolfe in a famous essay on the Valley:
Every year there was some place, the Wagon Wheel, Chez Yvonne, Rickey’s, the Roundhouse, where members of this esoteric fraternity, the young men and women of the semiconductor industry, would head after work to have a drink and gossip and brag and trade war stories about phase jitters, phantom circuits, bubble memories, pulse trains, bounceless contacts, burst modes, leapfrog tests, p-n junctions, sleeping-sickness modes, slow-death episodes, RAMs, NAKs, MOSes, PCMs, PROMs, PROM blowers, PROM burners, PROM blasters, and teramagnitudes, meaning multiples of a million millions.46
In Silicon Valley, people socialised, ideas fizzed around, giving them a chance to meet and mate, to recombine, and to trigger yet new ideas. ‘There is a velocity of information here that is very high’, one observer said. ‘The region’s dense social networks and open labour markets encourage experimentation and entrepreneurship . . .’ Saxenian writes. ‘The standing joke was that if you couldn’t figure out your process problems, go down to the Wagon Wheel and ask somebody.’
This is what is sometimes called horizontal information flow: the kind that travels from engineer to engineer, firm to firm, spilling over all the time. Information didn’t merely circulate within institutions, but between institutions. Spaces like the Wagon Wheel were hubs of recombination, cauldrons bubbling with people with different perspectives and paradigms. Insiders on one topic or technology were outsiders on another and vice versa, creating vast diversity of thought.
One such space was the Homebrew Computer Club, started by enthusiastic hobbyists, which had its first meeting in a garage. The logic was contained in the very first newsletter, posted in March 1975. ‘Are you building your own computer? Terminal? TV Typewriter? I/O device? Or some other digital black-magic box? If so, you might like to come to a gathering of people with like-minded inte
rests. Exchange information, swap ideas, talk shop, help work on a project, whatever . . .’47 (That first meeting took place just a few blocks from where, a few decades earlier, two men called Bill Hewlett and David Packard started experimenting with electronic equipment in a different garage.)
The inaugural meeting set the stage for what would follow. Ideas bubbled up as if from a shaken soda can. There were only a few hundred personal computers in existence at the time, but as the conversation fizzed, the assembled group came up with dozens of ideas for potential home use: text editing, storage, games, educational uses. One person even suggested using a computer system to control home functions such as the alarm, heating and sprinkler system.
One of the attendees at that meeting was a bearded enthusiast in his mid-twenties. Shy and softly spoken, he listened as the discussion danced around the terrain of personal computing. He had built his own processors, played around with chips, but he was now surrounded by a conversation that was the sociological equivalent of being plugged into thirty new brains, each with its own insights, diverse perspectives, specialist information, and rebel ideas.
As they discussed the Altair 8800, the first personal computer, which was sold to hobbyists in a build-it-yourself kit, he was intrigued. He had never seen one before. Then a datasheet from the 8800 was handed out, which fired his mind. ‘It was a meeting that grabbed my attention for life,’ he later said. ‘It was a Eureka moment for me . . . I took this data sheet home and was shocked to find that the microprocessor had gotten to the point of being a complete processor of the type I’d designed over and over in high school. That night the full image of the Apple I popped in my head.’48
The hobbyist’s name was Steve Wozniak. Thirteen months later, he would start the Apple Corporation from the two ideas that fused in his head that night (could there be a more exquisite example of a rebel combination?). His co-founder was another attendee at the Homebrew: Steve Jobs.
It is symptomatic that forums for idea exchange – whether restaurants, cafes or organically created clubs – were conspicuous by their absence along Route 128. There was no demand. Jeffrey Kalb, who worked in Massachusetts in minicomputing before moving to Silicon Valley, said: ‘I was not aware of similar meeting spots in Route 128. There may have been a lunch spot in Hudson or Marlboro, but there was nothing of the magnitude of Silicon Valley hangouts.’49 Route 128 companies didn’t neglect these things as an act of deliberate self-sabotage. They were creative and smart, but had not made an essential conceptual leap. Innovation is not just about creativity, it is also about connections. They were somewhat like the Geniuses in the thought experiment. They had originality, but lacked sociality. Diversity existed but was not exploited. The companies, like Tasmania, were islands separated by high waters. As Saxenian writes: ‘The networking and collaborative practices that typified Silicon Valley never became part of the mainstream business culture of Route 128, and the region’s new management models only partially departed from traditional corporate practices.’
In 1957, fifteen years before ‘Roadrunner’ was penned, Route 128 employed more than twice the number of workers in the tech sector than Silicon Valley. They had established companies such as Sylvania, Clevite, CBS-Hytron and Raytheon. These firms accounted for a third of the entire nation’s transmitting and special-purpose receiving tubes and a quarter of all solid state devices. In 1987, fifteen years after the song was written, this gap was reversed. Now, Silicon Valley employed more than three times as many tech workers as Route 128. By 2000, the isolated corporate-islands of Boston had all disappeared, somewhat like the technology on Tasmania.
We should note that competition between firms (even insular firms) is a form of information discovery at the level of the system. When institutions go head to head, we find out which ideas work and which do not. Companies with poor ideas go bankrupt, the successful firms are copied, and the system adapts. Well-functioning markets are a powerful engine of growth and contribute to the expansion of the collective brain. What the analysis of this chapter reveals, however, is the danger when information gets trapped within institutional boundaries. This is bad both for the system, because it evolves more slowly, and for the institutions themselves, which struggle to innovate.
Consider, too, that the fissures in the network along Route 128 both contributed to insularity, and were exacerbated by it. This was a dangerous synergy. The more people retreated into their silos, the more they perceived new ideas not as opportunities but as threats.
Mitch Kapor, the founder of Lotus Development Corp., talks of a ‘bizarre’ meeting with Ken Olsen, the CEO of DEC, where the latter seemed incapable of grasping the significance of personal computers. Kapor said:
Some of my most sobering moments . . . were seeing how those guys weren’t getting it, and were dooming themselves. Olsen had himself designed the case for the DEC personal computer, and he was banging on it and showing me how solid it was. I was going, ‘What planet am I on? This has nothing to do [with it].’ But in his world, when computers were on factory floors and so on, they needed to be robust. That really mattered. This didn’t matter at all.50
As for Silicon Valley, the region was forging ahead, insiders and outsiders colliding, diverse concepts recombining, a process bolstered by the geographical topology. Unlike the widely scattered islands of Route 128, the firms in the Valley ‘clustered in close proximity to one another in a dense industrial concentration’. The net result was a maelstrom of high velocity information flow. As Larry Jordan, an executive at Integrated Device Technology, said in a seminal interview in 1990: ‘There is a unique atmosphere here that continually revitalises itself by virtue of the fact that today’s collective understandings are informed by yesterday’s frustrations and modified by tomorrow’s recombinations . . . Learning occurs through these recombinations. No other geographical region creates recombination so effectively with so little disruption. The entire industrial fabric is strengthened by this process.’51
VI
The science of recombination presents us with a compelling vision. Innovation is about breaking down walls. Some walls are good, of course. Most of us value privacy. Most companies need to protect intellectual property. Most institutions need specialists who, in turn, need space to do their job. But we often get the balance wrong, leaning towards insularity, not because we don’t value the insights of people who think differently from ourselves, but because we underestimate their significance. This is another aspect of homophily. We are comfortable in our own silos, our own categories, our own conceptual milieus.
This is true even of science itself. Too often, academics speak to academics in the same discipline. This is fine, but only to a point. When historians only talk to historians, and economists to economists, they undermine their capacity to understand the very phenomena they seek to explain. Much of this book is taken from the work of academics with bracing outsider perspectives, working in multidisciplinary groups, replete with diversity in gender and ethnicity; academics who are enriching our comprehension of the world.
Ironically, some of these thinkers struggle to get published in scientific journals. The reason is that parts of academia have become like conceptual islands, comprised of self-referential peer-reviewing groups who struggle with anything outside the paradigm. There is also an insufficient recognition that much of the greatest science is recombinant. Successful scientists are not just those with deep knowledge of their own terrain, but those who have the imagination to peer out into the broader constellation, looking for meaningful cross-pollination. This is how they discover rebel combinations.
Given what we have learned, it is perhaps unsurprising that network theory is moving centre stage in multiple fields. The mathematical aspects of the subject were formalised by Euler in the eighteenth century, but the basic ideas are intuitive to grasp and practical to apply. Take building design, where architects are now curating spaces that maximise the scope for connections. Instead of closed-off cubicles and walled offices,
the idea is to bring people away from their desks, to create areas where people feel encouraged to mingle, make chance encounters and engage with outsider perspectives.
One leader who grasped these truths intuitively was Steve Jobs. When he was designing the building for Pixar, the animation company he bought from George Lucas in 1986, he made the decision to create just one set of toilets. These were in the atrium, meaning that people had to traipse across from all over the building. It seemed inefficient, but it forced people out of their usual niches, and led to a symphony of chance encounters. ‘Everybody has to run into each other,’ Jobs said.
Or take Building 20 at the Massachusetts Institute of Technology. This was not a beautiful structure. According to one essay on the building, it was ‘hastily constructed of plywood. It leaked. It had bad acoustics and was poorly lit, inadequately ventilated, very confusing to navigate (even for people who had been working there for years) and was scorching in the summer and freezing in the winter.’52 And yet it nevertheless spawned astonishing innovation including the construction of the world’s first atomic clock, the development of modern linguistics, one of the earliest atomic particle accelerators, stop-action photography and more. Jerome Lettvin, the cognitive scientist, called it ‘the womb of the institute’.
Why was the building so conducive to innovation? The lack of a formal structure caused scientists from different subject areas to bump into each other. Amar Bose, for example, started hanging out in the acoustics lab, just down the hall from his own department, while taking a break from his dissertation. He would later invent an innovative, wedge-shaped speaker (as well as the Bose Corporation).