Making far out the norm: Or how to nurture loonshots

by Ashutosh Jogalekar

Vannevar Bush – loonshot pioneer (Picture credit- TIME magazine)

What makes a revolutionary scientific or technological breakthrough by an individual, an organization or even a country possible? In his thought provoking book “Loonshots: How to Nurture the Crazy Ideas that Win Wars, Cure Diseases and Transform Industries”, physicist and biotechnology entrepreneur Safi Bahcall dwells on the ideas, dynamics and human factors that have enabled a select few organizations and nations in history to rise above the fray and make contributions of lasting impact to modern society. Bahcall calls such seminal, unintuitive, sometimes vehemently opposed ideas “Loonshots”. Loonshots is a play on “moonshots” because the people who come up with these ideas are often regarded as crazy or anti-establishment, troublemakers who want to rattle the status quo.

Bahcall focuses on a handful of individuals and companies to illustrate the kind of unconventional, out of the box thinking that makes breakthrough discoveries possible. Among his favorite individuals are Vannevar Bush, Akira Endo and Edwin Land, and among his favorite organizations are Bell Labs and American Airlines. Each of these individuals or organizations possessed the kind of hardy spirit that’s necessary to till their own field, often against the advice of their peers and superiors. Each possessed the imagination to figure out how to think unconventionally or orthogonal to the conventional wisdom. And each courageously pushed ahead with their ideas, even in the face of contradictory or discouraging data.

Vannevar Bush was a canny Yankee of Puritan stock who, after pioneering computer technology and mentoring brilliant scientists like Claude Shannon at MIT, became Franklin Roosevelt’s science advisor. He saw the importance of setting aside dedicated groups of scientists to work on unconventional ideas without bureaucratic pressure during his time advising the government on new weapons during World War 1. One of Bush’s brainwaves was to shelter the development of radar, a technology that had been invented in the 1920s by two American scientists at the Naval Air Station named Leo Young and Hoyt Taylor. Young and Taylor found that the intensity of a transmitted radio wave was amplified at certain points in time when a ship passed in front of it; in other words, the amplification of this wave could detect the ship. Similarly one could detect a plane when it passed overhead. Young and Taylor’s idea was rejected as being too far out and of no practical use until World War 2 when its utility in detecting German fighter aircraft was proven. Thanks to Bush’s lobbying and prescience, FDR nominated him to head the Office of Scientific Research and Development (OSRD). In this capacity Bush led both the MIT radiation lab that developed radar and the atomic bomb project that sprawled across Los Alamos and other sites.

But perhaps another group worth noting in the development of radar was a group of misfit, eccentric meteorologists in Britain led by Robert Watson-Watts who started trying to amplify radio signals in the mid-1930s when they were caught in the cross fire (described well in CP Snow’s book “Science and Government“) between science administrators Frederick Lindemann and Henry Tizard. Lindemann was Churchill’s favorite scientist and wanted to divert resources from radar toward more fanciful projects like aerial mines and even a “death ray” that would concentrate a beam of electromagnetic radiation to perhaps blow an enemy plane out of the sky. At one point it seemed that the radar project would be shelved, but Churchill who chaired the committee to make decisions about key defense projects turned out to be prescient enough to appreciate what Watson-Watts and his colleagues were doing. Radar was given priority, and when 1940 and the invasion of Poland came along, it literally saved Britain from being invaded. Further work on both radar and the atomic bomb took place at the University of Birmingham where immigrants and refugees like Mark Oliphant, Rudolf Peierls and Otto Frisch developed the first estimates of critical mass and the cavity magnetron, a critical device that demonstrated radar’s proof of principle. These British breakthroughs laid the groundwork for American supremacy in radar and atomic weapons during the war and beyond.

What made breakthroughs such as those at Vannevar Bush’s OSRD possible? In Bahcall’s view, Bush’s great idea was to recognize the importance of both loonshots and “franchise” work. Franchise work would correspond to what we usually call the development phase in industry. It could consist of deployment of a software package or mass production of an automobile. Loonshots on the other hand are more in the discovery phase which is the realm of basic research. Bush’s genius was not just his recognition of both curiosity-driven basic loonshot research and franchise work but of the critical interplay between them. To describe this interplay Bahcall borrows two concepts from his physics background – phase separation and dynamic equilibrium. At what’s called a critical point, two phases of matter can productively co-exist. Phase separation means the protected individual workings of loonshot and franchise parts of R&D while dynamic equilibrium means the fruitful sharing of ideas between them. Bush was as comfortable nurturing the loonshot province of scientists like Taylor and Hoyt and Alfred Loomis – the gentleman millionaire inventor who built a laboratory for physics breakthroughs in his house in rural new York – as he was hobnobbing with military generals and company executives who could run with the loonshots and transform them into practical inventions. After the war, Bush wrote an important report called “Science the Endless Frontier” which urged government investment in basic research. This report played a key role in the creation of the NSF, the NIH and other science agencies that have funded American science during the last fifty years. These organizations have supported long-term basic research whose fruits have then been picked and leveraged into practical innovations by industries like the pharmaceutical and electronics industries. This constant interaction between pure and applied science, with pure science largely taking place in universities while applied science and engineering taking place in industry, have largely been responsible for America’s dominance in science and technology since the end of the Second World War.

Bahcall also draws an interesting distinction between what he calls “P-type” and “S-type” loonshots. P-type loonshots refer to pure product (P)-based solutions that are breakthroughs. S-type Loonshots refer to strategy (S)-based solutions that move the needle on an organization or industry capturing the market. One of the best examples of a P-type loonshot is Edwin Land’s Polaroid Corporation which kept on coming out with newer and newer products that transformed the world of photography, motion pictures and industrial engineering. Polaroid kept on innovating for a long time, but missed the digital revolution which quickly left it behind. A good example of a S-type loonshot, and one which Bahcall spends quite some time dissecting, is American Airlines’ trumping of Pan Am Airlines. For decades Pan Am Airlines had pioneered air travel, in significant part because its chief executive Juan Trippe kept on executing series after series of P-type Loonshots, exploiting new developments in jet engine technology and partnering with luminaries like Charles Lindbergh. It was largely due to Pan Am that air travel went from being short haul air taxi service to an intercontinental mode of transport. But as the 1970s dawned, Pan-Am started losing ground to American Airlines, not because American came up with radically new technology but because they executed a series of incremental but attractive S-type maneuvers like an electronic booking system that allowed customers to compare different flights and a rewards program. These were relatively mundane and boring developments, but they were enough to let American gradually capture a bigger and bigger share of the market until Pan Am finally went out of business in the 1990s. S-type maneuvers can involve incremental customer-centric innovations as well as marketing. My own view is that S-type loonshots can only save an organization for so long – witness the dry spell that American has found itself in recently – but it’s also true that they can help a business gain an edge over its rivals. The best inventors and entrepreneurs like Edison were highly adept at executing both kinds of loonshot. Another example of a S-type loonshot that could not save a company for too long is the aggressive marketing that Pfizer did in the 1990s to sell its blockbuster drugs like Viagra and Lipitor. The scheme worked exceedingly well, to an extent. But ultimately, marketing and sales cannot overcome the lack of new products, and Pfizer today finds itself quite low on the ladder of innovative drug companies.

Pfizer’s story of Lipitor, part of a class of drugs called statins that are the best selling drugs in history (bringing in an amazing $300 billion in annual revenue) is a good case study in individual loonshots. The first statin was discovered by Akira Endo, a Japanese scientist working in the pharmaceutical company Sankyo in the 1970s. Endo found out that a small molecule made by a fungus found on rice could inhibit cholesterol synthesis in living organisms and potentially prevent heart disease. Early tests in rats showed no effects, but Endo kept on going in the face of opposition and succeeded in showing that his molecule did dramatically lower cholesterol in chickens (today it’s well recognized – although not practically utilized enough – that animals like rats can be very poor models for testing drugs in humans). Later studies in dogs showed what appeared to be the mushrooming of tumors, but it was found that these tumors were an artifact, and two other scientists who had come up with the molecular mechanism by which the statins could reduce cholesterol production urged important pharmaceutical companies to continue their clinical trials. Those two scientists ultimately won the Nobel Prize; Akira Endo certainly deserves to win one as well. Endo’s loonshot example is emblematic of cases where loonshots can be missed because someone simply did not drill deeper into the causes of failure. In Endo’s case the problem was that rats were not a good model system for testing his drug in humans, and he recognized that. Another case that Bahcall notes is that of Facebook. Many early investors had backed away from funding Facebook because of the lack of success with other social networking platforms, but only Peter Thiel and his associates tried to understand why the platforms weren’t working. Once they did this they found that the platforms were failing to retain users not because of some intrinsic bias toward social networking but because of technical problems that Facebook promised to fix. $500,000 and a few years later, Thiel’s investment paid off spectacularly.

After describing how dynamic equilibrium, phase separation between loonshot groups and franchise groups and recognizing P-type and S-type innovations can help nurture loonshots, Bahcall ruminates on how exactly one can create an environment suitable for loonshots in modern organizations. He comes up with a formula consisting of a few key variables like management incentives, number of reports and equity shares to try to estimate a magic number above which the balance would tip from an organization being one that focuses on loonshot projects versus one that focuses on politics. Having been part of organizations that unduly and frustratingly focused on the latter myself, I share Bahcall’s eagerness to understand what exactly tips this balance. Intriguingly, the magic number he comes up with is 150. This is the same limiting number suggested by anthropologist Robin Dunbar for nurturing favorable interactions between small groups ranging from hunter gatherers to modern corporate teams. While this analysis is amusing and creative, I find it mostly to be inspired numerology; at the very least more organizations would need to be examined in order to substantiate it, and at worst it can be an arbitrary result that could also be generated by considering other factors. However, the validity or lack thereof of the formula should not detract from the fact that employees of an organization should be incentivized to grow based on their contribution to projects rather than their contribution to schmoozing with upper management.

Bahcall is also squarely of the opinion that there is such a thing as Too Big to Innovate. Small companies in his opinion have the agility to nurture loonshots in ways that big bureaucracies don’t. Having worked for small organizations myself I generally agree. And yet Vannevar Bush headed a giant organization to create innovation and Merck CEO Roy Vagelos also headed a giant organization that innovated so well that it was rated as the most admirable and innovative company of the 1980s. Clearly it’s true that leadership plays a role, as exemplified by Bush, Vagelos, Edwin Land and Steve Jobs. Bahcall notes Jobs as a good example of someone who, after he left Apple and recognized the potential of Pixar, made sure to nurture the interactions between creative groups while not micromanaging the product itself; in this case Jobs was smart enough to realize that he knew nothing about moviemaking, and the Pixar personnel who had worked with George Lucas and other pioneers of computer graphics in films knew better. Even after he came back to Apple in 1997, Jobs was careful enough not to alienate the franchise group headed by Tim Cook – something that he failed to do and suffered the consequences of at Apple in the 80s – while keeping the loonshot group headed by Jony Ive separate and thriving. Similarly, Mervin Kelly who led Bell Labs during its heyday knew that he needed to keep the communications engineers separate from the physicists and chemists while creating useful contact between the two. This kind of inspired hands-off management has been a signature of other loonshot pioneers – the directors of Bell Labs, the Cavendish Laboratories and the MRC Laboratory of Molecular Biology (which produced no less than 15 Nobel Laureates) all made sure to spend informal time with their teams but otherwise stayed out of their way to ensure innovation.

How far can you extend the idea of nurturing loonshots? In the last part of the book, Bahcall extends his analysis to understanding one of the cardinal questions of history, one that can be summarized in the phrase “Why the West Rules”. The British historian, Sinologist and polymath Joseph Needham asked this as the Needham Question – Why did the Industrial Revolution take place in England and not China? It is a fact that economically and technologically, China and India were more prosperous and advanced than Europe until the 16th century or so. An observer seeing the world during this time would almost certainly have predicted the dominance of these two civilizations in the future. And yet today the developed world mostly speaks English and relies on Newtonian mathematics and Darwinian evolution to understand the universe and on Edward Coke’s law to understand systems of government and society. Clearly something happened in Europe during the 17th and 18th century to catapult it ahead of China and light the spark that was the Industrial Revolution. We know that something as the Enlightenment, but Bahcall says that the real cause was the failure of China and India to separately nurture both loonshots and franchise enterprises. For instance, China made huge technological advances in navigation, weaponry, writing and textile manufacturing, but it never really nurtured curiosity-driven, pure scientific research of the kind encouraged by the English Enlightenment tradition. In the ecosystem of Chinese technological evolution, the development of calculus or geometry might have never taken place, and whatever use China had for basic science was only as the handmaiden of the applied sciences.

China’s focus on applied technology rather than basic science is borne out by the historical facts, but I think Bahcall ignores a few crucial other factors that might have likely played a role in the development of this mother of all loonshots; not surprisingly, as our circle of understanding of innovation grows from individuals to teams to entire countries and civilizations, more and more factors would play a role as potential explanatory elements. One of the most illuminating books I have read about why science developed in the West rather than in the East is Toby Huff’s marvelous book “The Rise of Early Modern Science”. In this volume, Huff provides an explanation that I personally find equally if not more important – an explanation based on religion and religious philosophy. According to Huff, through its basis in Greek philosophy, Christianity was unique in nurturing the kind of inquiry into nature as God’s creation that laid the foundations for the scientific and the industrial revolutions. Contrary to the popular view of Europe’s scientific decline during the so-called Dark Ages, Plato’s key works that encouraged the study of Nature as the work of God stayed alive in Europe’s libraries and universities. The Bible did not consider itself as the end of all knowledge but urged Christians to seek knowledge and God through study and reflection. Because of this emphasis on scholarship, and contrary again to popular belief, experimentation and theorizing were almost never forbidden in Europe throughout the Middle Ages; this in significant part explains why there were so many Jesuit scientists for instance.

In contrast, Islam and China had a much more self-contained view of their holy books and philosophy to consider open-ended exploration of nature: Confucianism was too caught up in a kind of mystical cosmogony, while Muslims found that the Quran proclaimed that almost all the answers that human beings would have about God, themselves and Nature would be found within its confines. The difference between open and closed systems in Europe versus the East were reflected in the kinds of education that young men received in Europe, China and Islamic countries; in universities steeped in a foundation of Greek philosophy in the former case and in Madrassas and the strict hierarchical bureaucracy of the various Chinese dynasties in the latter. This explanation for why science and technology grew in the West rather than the East does not contradict Bahcall’s explanation of loonshot-franchise separation, but the latter seems like a more proximate cause and I think the former goes deeper. It might potentially further explain why the most important science developed in England rather than in Spain, France, Italy or Holland when each one of these countries was at one point ascendant in world affairs: a good case can be made that the rise of the Church of England, the vigorous disputes between Catholic and Protestant doctrines (which led to the flight of Puritans to the New World) and the concomitant rise of parliamentary democracy in England before anywhere else played a huge role in encouraging the skeptical and open thinking that is the hallmark of the scientific method. This analysis also says that if true scientific innovation is to happen in the future in China or the Islamic countries, these countries should not just give free rein to their bright minds to explore the frontiers of science but must also nurture the kind of free-thinking, democratic social and political systems of which scientific franchises ultimately are only a benefit. Without this openness science may always stay stifled in these environments, and large-scale engineering implementation may be the limit of their creative development.

If the world is to see progress it needs to nurture the kind of far out thinking that leads to loonshots and breakthroughs. Sadly, places like Silicon Valley are increasingly generating not loonshots but deceptive views of what they call “disruptive innovation”, a buzzword that Bahcall and I both share skepticism about. What would ideally be the wisdom of crowds has instead turned into herd thinking, with massive amounts of talent, money and ultimately hype thrown at incremental advances which are often trivial in terms of social or even material advancement. Key areas like sanitation, climate change, poverty and neglected diseases are often ignored in favor of the latest startup coming up with a new way to instantly message your friends. Perhaps we are suffering from a disease of excess, and perhaps one of the messages to consider from the story of loonshots is that less – not less innovation per se but less fruitless innovation – might actually be more. Figuring out how to incentivize people to think that way might be the ultimate loonshot we are waiting for.

Like what you're reading? Don't keep it to yourself!
Share on Facebook
Facebook
Tweet about this on Twitter
Twitter
Share on Reddit
Reddit
Share on LinkedIn
Linkedin
Email this to someone
email