What Would It Take to Recreate Bell Labs?

For most of the 20th century, AT&T was almost entirely responsible for building and operating America’s telephone infrastructure. It manufactured the phones and electrical equipment, laid hundreds of millions of miles of wire across the country, and built and operated the switchboards and exchanges that made it possible for anyone with a phone to call anyone else.

This huge network required billions of dollars worth of equipment: telephones, switches, cables, amplifiers, repeaters, and so on. All this equipment was built by AT&T’s manufacturing subsidiary, Western Electric. But it was designed and developed by AT&T’s research arm, Bell Telephone Laboratories, better known as Bell Labs.

To students of technological progress, Bell Labs is a giant. For decades, Bell Labs was considered not only the best industrial research lab in the world, but arguably the best research lab in the world, period. One Bell Lab alumnus described it as “a parallel organization to almost all the academic institutions put together.” Bell Labs not only developed new telephone equipment but performed novel scientific research, under the assumption that such research would ultimately result in improved communications technology.

Bell Labs is most famous for being the birthplace of the transistor, but that’s just one of dozens of major inventions and discoveries that originated there. Bell Labs also spawned: the silicon solar PV cell, the first active and passive communications satellites, the first videophone, the first cellular telephone system, the first fiber optic telephone cable, the quartz clock, Information Theory, Statistical Process Control, the UNIX computer operating system, and the discovery of Cosmic Microwave Background radiation. Many of Bell Labs' less famous inventions were among its most important: the discovery of compounds that could protect polyethylene from decomposing in sunlight isn’t typically mentioned on lists of Bell Labs’ most impressive achievements, but the patents for them were the most valuable that AT&T ever produced. Among Bell Labs’ awards are 10 Nobel Prizes, 5 Turing Awards (the highest honor in computing), and 5 Draper Prizes (the highest honor in engineering).¹ 36 Bell Labs staff members have been inducted into the Inventors Hall of Fame (though not all for work they did at Bell Labs).

Bell Labs still exists today, as a subsidiary of Nokia, but outside the name it has little relationship to the industrial research powerhouse of the 20th century. Following the breakup of AT&T, Bell Labs was gradually carved apart through constant mergers and spinoffs to different organizations, and it has never achieved the heights it did in the 1950s through the 1980s.

Unsurprisingly, there’s perennial interest in recreating a Bell Labs-style research lab that steadily churns out world-changing inventions and discoveries. In a recent newsletter post, Noah Smith argues that we need an “energy Bell Labs” that can push U.S. energy technology forward, and he’s far from the only one to dream of recreating a Bell Labs-style organization. The idea pops up repeatedly.

Unfortunately, the conditions that made Bell Labs so successful were highly historically contingent and not the sort of thing that could be deliberately recreated. Being a subsidiary of a government-sanctioned, vertically integrated monopoly gave Bell Labs a broad research scope and freedom to pursue long-term research projects unavailable to most other industrial labs. Prior discoveries in quantum mechanics provided a wealth of new phenomena that Bell Labs could harvest for new technology, and WWII both pushed technology forward across the board and turned Bell Labs into an organization poised to capitalize on it. In the end, Bell Labs was ultimately undermined by the very technologies that it had created. The world that Bell Labs thrived in no longer exists: to push technological progress forward, we'll need to understand both why Bell Labs worked and why it no longer could.

Origins of Bell Labs

For the first several decades of its existence, AT&T’s scientific and inventing efforts were modest. Though new technology was needed to improve and expand the Bell System, AT&T largely relied on acquiring outside patents and inventions and adapting them to its particular needs. In 1892, Hammond Hayes, the head of the Mechanical Department (which designed and tested new equipment) wrote that he had decided to abandon any “theoretical” efforts, and would focus the department on the “practical developments of instruments and apparatus." Any “theoretical” needs could be obtained by collaborating with students at MIT or Harvard. In 1906, Hayes noted that “every effort in the Department is being executed toward perfecting the engineering methods; no one is employed who as an inventor is capable of originating new apparatus of novel design.”

But starting in 1907, this technological stance began to shift. AT&T’s financial position had deteriorated as it battled with the thousands of independent phone companies that sprang up after its key patents expired. Between 1902 and 1907, AT&T’s debt tripled, and its stock price fell by more than 50%; it could no longer sell the bonds it needed to finance its expansion.

In response to AT&T’s dire financial straits, the company was taken over by a JP Morgan-led group of bankers, who installed Theodore Vail as president. Vail had been the first general manager of AT&T in 1878, and had shepherded it through its early difficulties with expansion and patent battles, but had left the company in 1887 due to disputes with its ownership.² Vail had previously worked at the U.S. Post Office and had absorbed its public service ethos. He believed AT&T should operate with an eye towards the long term, and that it should constantly expand and improve its system with new technology. In fact, Vail had initiated the research project for the first buried telephone cable, to address the problems caused by huge masses of overhead wires. The company’s owners, on the other hand, were far more interested in operating as profitably as possible, and wished to avoid unnecessary capital expenditures, technical improvements, or risky research and development endeavors. Unable to resolve this disagreement, Vail left the company in 1887. But when Vail returned to AT&T in 1907, he was given a nearly free hand to shape its operations as he saw fit.

Vail desired to bring the entire telephone industry under the control of AT&T. His articulated (and likely seriously held) reasons were that the telephone system was a natural monopoly, and that the public interest would be best served by a single, government-regulated company, captured by the oft-repeated motto “one system, one policy, universal service." Vail came to believe that universal service — connecting all telephones into a single network — required not only the steady hand of a single company ensuring each component worked reliably and seamlessly, but also constantly improving technology to overcome distance and the exponential increase in complexity caused by more phones connecting to the network. While prior to his arrival, AT&T had largely eschewed invention or theoretical research, by 1910 Vail could boast that AT&T had “extensive laboratories and experimental departments," and that it was generating its own fundamental innovations.

The first major success of Vail’s technology-focused strategy was solving the problem of transcontinental transmission. When Vail ascended to the presidency, a telephone signal could travel a maximum distance of roughly 1,800 miles; enough to connect New York to Denver, but no farther. The farther telephone signals were sent, the more they attenuated; without some sort of signal amplifier, connecting phones on the east and west coasts (and Vail’s dream of universal service) would be impossible.

To tackle this problem, AT&T’s John Carty (who had replaced Hayes as head of all AT&T research) and researcher Frank Jewett began a project to develop a telephone signal repeater in 1909. Over the following year, they investigated several potential methods of amplification, including magnetic and mechanical means. But when none of these proved suitable, Carty and Jewett changed strategies. Jewett, who had a PhD in physics from the University of Chicago, was familiar with physics research on the behavior of (recently discovered) electrons, and thought that such research might offer a path forward to creating a successful amplifier. Carty and Jewett hired physicist Harold Arnold in 1911, who began investigating electronic means of amplification.

Arnold’s initial studies on a potential device, the mercury vapor tube, proved fruitless, but in 1912 he was impressed by a demonstration of the audion, invented by Lee de Forest, which was capable of amplifying an electric signal. De Forest, like Edison, was at heart a tinkerer, and it soon became clear that he didn’t quite understand how his invention worked. But it nevertheless seemed promising, and Arnold spent the next several years building a research team to study its behavior and turn it into an effective telephone amplifier. By 1915, 45 people were working in AT&T’s research division, including 7 PhDs. That year, the first New York-San Francisco telephone line opened, powered by the new vacuum-tube-based telephone amplifier. For the first time, it appeared possible to achieve Vail’s goal of a universal telephone system. Science-based technological development had paved the way.

Bell Labs wasn’t technically formed until 1925, when AT&T was reorganized and several different research and engineering organizations were combined under a single roof. But by then Vail’s values — constant improvement of telephone service through science-based technological developments — had been woven deep into the fabric of the company. The year Bell Labs formed, researcher Clinton Davisson began the experiments that would win Bell Labs its first Nobel Prize. While Vail retired in 1919 and died the following year, he made sure to leave the company in the hands of leadership that believed in his credo of universal service, and that constant innovation was the path to get there. Vail conceived of the telephone network as “an ever-living organism” which required “unceasing effort, continually improving and upbuilding." Thanks to Vail’s efforts, it became an article of faith within AT&T that technical innovation would always be needed, and would be valuable even if it took many years to bear fruit.

The successes of Bell Labs

In its efforts to constantly improve telephone service, Bell Labs would pursue a wide range of scientific investigations; its research division had departments ranging from physics and chemistry to metallurgy, mathematics, and even physiology and psychology. Though its mandate was to improve communications technology, there were innumerable possible ways that might be achieved. Invention, noted Harold Arnold, “is not to be scheduled or coerced” - who knew what fields of science might yield important, practical developments?

But despite its unusual willingness to engage in fundamental scientific research, Bell Labs was still an industrial research lab at heart, tasked with developing products and technology for its parent company. While it took an expansive view of this mandate, investigating anything related to the field of communications and taking on long-term projects with uncertain payoffs, its focus was nevertheless on new technology of practical use to the Bell System. Eric Gilliam, in his essay on Bell Labs’ research culture, describes this as a “long leash but a narrow fence” — researchers had the freedom to pursue a variety of avenues that might be valuable to the Bell System, but there were mechanisms in place (such as an army of system engineers who kept abreast of both scientific developments and practical needs of the telephone system) to nudge them towards the most promising problems.

And while its scientific investigations and groundbreaking inventions — that is, its research — made Bell Labs famous, research was always a comparatively small part of Bell Labs’ efforts. Depending on the period, researchers made up only around 10-20% of Bell Labs’ headcount. The vast majority of efforts were always focused on the less glamorous but important work of product development: taking inventions and discoveries and turning them into manufacturable products, testing materials and components to make sure they functioned properly, and gradually improving AT&T’s telephone equipment and infrastructure. In his history of Bell labs, Jon Gertner describes this sort of work:

Some men at West Street specialized in experimenting on springs for switchboard keys, others in improving the metal within the springs. AT&T linemen bet with their lives on the integrity of the leather harnesses that kept them tethered at great heights – so Labs technicians established strength and standards for the two-inch leather belts…and improving the metal rivets and parts. Millions of soldered joints held the system together – so Labs engineers had to spend years investigating which fluxes and compounds were best for reinforcing anything from seams on sheet metal to lead joints to copper wires to brass casings… A Bell Labs engineer named Donald Quarles…wrote a long treatise entitled “Motion of Telephone Wires in the Wind.” His men made rigorous, multiyear tests on the proper spans (how far should the poles be spaced apart?), proper lashing (how tight should the wires be wired together?), proper vertical spacing between horizontal strings of wires… Many of the system’s most important cables, meanwhile, were not strung through the air but ran underground. For burying wire, the men in Chester had to develop new processes involving special tractors they invented and splicing techniques.

But despite the relatively small proportion of research and invention work in Bell Labs, it was nevertheless crucial. Only by constantly pushing the boundaries of technology forward could AT&T accommodate the relentless growth of the company and demand for service. By 1939 AT&T controlled 83% of all U.S. telephones, 98% of long-distance wires, and 100% of intercontinental radio telephone links. To meet demand, AT&T was adding 2 million miles of telephone wire each year. Bell Labs inventions like the negative feedback amplifier, the coaxial cable, and the crossbar switch made handling this growth possible.

Because its investigations were so wide-ranging, Bell Labs often found itself inventing technology outside of its telecommunications purview.³ Bell Labs created the first motion picture with synchronized sound (ie: “talkies”), and the first demonstration of color TV. For a time, AT&T tried to license these non-telephone inventions under a subsidiary, Electrical Research Products Incorporated (ERPI), but it faced constant criticism that it was using its monopoly power to muscle into other industries. AT&T eventually spun off ERPI in 1935 (the company still exists today as Altec Lansing), and adopted generous licensing terms for its patents. AT&T was so large that it existed right at the edge of what the government would allow to exist, and this tension of capitalizing on its inventions while avoiding antitrust efforts would plague the company throughout its life.

Though it had many successes in the first 25 years of its life, the crowning achievement of Bell Labs research (and its strategy of leveraging early-stage scientific research to create new products) is undoubtedly its development of the transistor, along with its various derivatives (the MOSFET, the solar PV cell) and associated manufacturing technologies (including crystal pulling, zone melting, and diffusion furnaces). The transistor is a classic case of Bell Labs’ strategy: wide research freedom, circumscribed by the requirement to produce things useful for the Bell System. The telephone network required enormous amounts of vacuum tubes and mechanical relays to act as switches, but these were far from ideal components. Vacuum tubes were delicate, power-hungry, and fragile, and mechanical relays were slow and prone to wearing out. Mervin Kelly, physicist and head of the Bell Labs vacuum tube department in the early 1930s (and later the president of Bell Labs), dreamed of replacing them with solid-state components with no moving parts. Advances in quantum mechanics, and novel materials known as semiconductors, suggested that such components might be possible. 

Bell Labs had studied semiconductors since the early 1930s; Walter Brattain, who would eventually share the Nobel Prize for inventing the transistor, was hired in 1929 and had begun to study an early semiconductor device called the copper oxide rectifier. A Depression hiring freeze stymied more serious semiconductor efforts until 1936, when Mervin Kelly (now Bell Labs’ director of research) was finally able to start building a more robust solid-state physics department and hired physicist William Shockley (the second of the three transistor inventors). While not giving Shockley any specific research tasks (indeed, the entire solid-state group had “unprecedented liberty to follow their own research noses as long as their work dovetailed with general company goals”), Kelly emphasized to Shockley the potential value of a solid-state component to replace tubes and mechanical relays.

The solid-state physicists continued their research over the next several years, studying the behavior of semiconductors and attempting to create a semiconductor amplifier. This research was interrupted by the war but resumed in 1945, the same year physicist John Bardeen was hired. Bardeen proved to be the catalyst the solid-state group needed, and over the next several years Bardeen, Brattain, and Shockley made progress in understanding semiconductor behavior. In December 1947, they unveiled their semiconductor amplifier: the transistor. By 1950, Western Electric was making 100 transistors a month for use in Bell System equipment. A few years later, in 1954, another Bell Labs solid-state research effort yielded the world’s first silicon solar PV cell.

The invention of the transistor was followed by a series of other inventions that turned it from a laboratory curiosity into a practically useful product: the bipolar junction transistor in 1948 was easier to produce and more reliable than Bardeen and Brattain’s point-contact transistor. Crystal pulling in 1950 and zone melting techniques in 1951 created the ultrapure silicon required for the silicon transistor, invented in 1954 (the first transistors used germanium). That year, diffusion furnaces were used to introduce microscopic amounts of impurities to create p-n junctions, and so on. 

Bell Labs had a solid reputation for research since Clinton Davisson’s 1937 Nobel Prize, but it was the transistor that cemented its status as the world’s leading industrial research laboratory, and possibly the world’s leading research lab. Bell Labs offered nearly all the freedom of an academic research environment, without the burdens of writing grants or teaching classes. Bell Labs researchers had access to far more technical resources (such as cutting-edge equipment) than their academic counterparts. Its prestige, reputation for excellence, and envious working environment allowed Bell Labs to acquire some of the most talented researchers in the world. Bell Labs Nobel Prize winner Horst Stormer noted that “Over a very long stretch of time, it was the best place in the world and it attracted — and attracts — the best people.” In his short memoir about Bell Labs, Michael Noll likewise noted that “it seemed everybody wanted to get a job there”. Of Bell Labs’ 10 Nobel Prizes, 8 came from researchers hired in the ‘50s, ‘60s, and ‘70s following the invention of the transistor.

The decline of Bell Labs

Bell Labs continued to create important inventions and discoveries right up until the 1982 court-ordered breakup of AT&T. Following the consent decree that required AT&T to divest its local telephone operating companies, there was much concern, inside and outside the company, about what would happen to Bell Labs.

Initially, there was optimism that Bell Labs’ seemingly unique, highly productive research environment might be maintained. In 1983, a year before divestiture, Bell Labs AI researcher Mitchell Marcus stated, “I have a feeling that we are returning to an older spirit at the Laboratories that has been very productive.” Though a portion of Bell Labs was carved off following divestiture to form Bellcore, most of Bell Labs remained intact, and it continued its tradition of world-class research. Indeed, research leading to 5 Nobel prizes occurred after the court mandated the breakup of AT&T.

But this optimism proved to be misplaced. Following the divestiture, Bell Labs was repeatedly passed from organization to organization and split into smaller parts. After the 1983 split that created Bellcore, Bell Labs remained part of AT&T until 1996, when Western Electric was spun off to create Lucent Technologies. Bell Labs was split again, with most of the organization (and the name) going to Lucent, and a portion staying with AT&T. Lucent, in turn, would later spin off two smaller companies (Avaya and Agere) which each took a portion of Bell Labs researchers. Lucent was then later acquired by Alcatel, shifting ownership of the labs yet again. In 2015, Nokia acquired Alcatel-Lucent, where Bell Labs continues to exist today as “Nokia Bell Labs." 

In addition to this organizational whiplash, Bell Labs gradually found itself under more financial pressure. Bell Labs was reorganized along business lines, with separate departments focused on specific customer and product types. Some research areas, such as economics and social psychology, were cut, and the remaining research areas found themselves increasingly pressured to focus on the immediate needs of the business. In the face of these changes, talented employees began to leave for academia or Silicon Valley.

Problems accelerated after Bell Labs was passed to Lucent. Lucent was a far smaller company than AT&T at its peak: AT&T had around 1 million employees at the end of the 1970s, while Lucent had just 140,000 employees when it was formed (and shrunk to just 35,000 employees by 2002). By comparison, Bell Labs employed around 25,000 people and 1,300 researchers at the end of the 1970s. Even after two decades of carve-offs and employee departures, the Labs was a far greater financial burden to Lucent than it had been to the largest corporation in America. Offices began to be closed, and layoffs or requested early retirements began for researchers. By 2002, there were only around 500 Bell Labs researchers left. Financial pressure continued: in 2005, a new Bell Labs president announced his intent to “align Bell Labs research with the company’s business activities and provide a much stronger impact on the bottom line.”

Amidst these difficulties, Bell Labs’ research efforts slowed. Though Nobel Prize-winning research continued to be performed in the 1980s and even 1990s, it was almost entirely done by researchers who had been hired prior to divestiture. Only one of Bell Labs’ 10 Nobel Prizes was done by an employee hired following the divestiture. A similar pattern exists for the winners of the Draper Prize and those inducted into the National Inventors Hall of Fame. Research seems to have been increasingly focused on software (one notable Bell Labs software researcher is machine learning expert Yann LeCun, who joined Bell Labs in 1988).

Today, Nokia Bell Labs employs around 750 people, with research efforts focused on “network fundamentals, automation, semiconductors, and AI." But while it trades on the name of its predecessor, for all intents and purposes the old Bell Labs is gone.

What made Bell Labs so successful?

While Bell Labs was far from the only major industrial research laboratory, no other lab could match the scale or scope of Bell Labs’ achievements. In part, Bell Labs’ success was due to getting the right “formula” for running a productive research organization, but it was also due to a variety of historical circumstances that couldn’t be deliberately recreated.

Most importantly, Bell Labs was a subsidiary of AT&T, a highly vertically integrated, government-sanctioned monopoly provider of telephone service, and by far the largest company in the world. This yielded several key benefits. Its size allowed AT&T to devote incredible resources to research and development: though Bell Labs was a huge research organization, it was a comparatively small fraction of AT&T’s revenue. And AT&T’s monopoly status allowed Bell Labs to undertake long-term projects that might take years to pay off. Even if a research effort took 10 or 20 years to bear fruit, the monopoly would still be there. Many of Bell Labs' most important technological advances — fiber optics, undersea telephone cables, electronic switching — took decades to develop.⁴

AT&T’s size also gave it a low bar for what constituted a valuable technical improvement. Even a tiny improvement that saved a few cents on a component or service would be large when multiplied by the enormous scale of the Bell System. This low bar made it far more likely a given research effort would be successful.

Similarly, AT&T’s vertical integration, where it designed, built, and operated all its own telephone equipment, gave it a very large research “surface area." The breadth of its offerings could justify a similarly broad set of research initiatives in a variety of different fields. Any given discovery would likely be useful somewhere in the Bell System. For instance, a test developed for depth perception by Bell Labs vision researcher Bela Julesz proved to be useful for quality control inspection of integrated circuits, by weeding out inspectors who lacked depth perception. 

Cultivating a broad swath of research activities and expertise also likely made those activities more productive. Other experts could (and did) discuss problems, share insights, and suggest possible approaches. The huge pool of knowledge made it easier to avoid bottlenecks and find solutions to problems encountered during development. Many Bell Labs alumni have noted the value of the huge amount of expertise immediately available to help solve problems:

One of the great features of Bell Labs was that there were so many experts in so many different fields within it. Furthermore, they were all accessible and glad to share their experience and knowledge with others. The openness of the intellectual atmosphere was a huge strength - Walter Brown

The size of AT&T also forced technical progress as a matter of survival. AT&T was so large that it sat right on the edge of what the government would allow to exist. Over its life, AT&T was constantly threatened with antitrust action or with government takeover, and only appeasements — such as agreeing to manage Sandia National Laboratories for the government — managed to stave them off. The most important appeasement was constant improvements to telephone service: only this way could AT&T continue to justify its existence. As Peter Temin and Louis Galambos note in “The Fall of the Bell System”:

Only its best behavior could shield the largest regulated monopoly in the world from attack–and even that might not be enough…This record of relentless technological improvement was the glue that held together AT&T’s various accommodations with the state and federal governments

Bell Labs also benefited from a strong sense of purpose: its mandate was to improve telephone and communication technology, to make AT&T’s offerings better and cheaper, and to accommodate the constantly rising demand for telephone service. Though Bell Labs gave researchers a large amount of freedom, the researchers nevertheless knew they were there to help build the Bell System, and the Labs had mechanisms in place to ensure research efforts were directed to the most important problems.

John Pierce, the force behind Bell Labs’ communication satellites, argued that “it's very important for laboratories to have some responsibility and some general goal”: labs operate best when there are people on the work that you’re doing. Pierce argues that Bell Labs and the other great labs of the 20th century “were really needed, and they rose to the need.” Not only did this clearly articulated, strongly felt purpose provide a powerful motivation, but it created a “problem-rich environment," a clear set of problems that can act as a guiding star. Shockley and other researchers in Bell Labs’ solid-state department weren’t assigned any particular research task, but they all knew the potential value and usefulness of a solid-state amplifier, and their efforts coalesced around producing one as their research revealed ways that it might be possible. This focus on turning scientific research into real, physical products also likely improved research efforts, by allowing rapid feedback and iteration; people familiar with “on-the-ground” needs could provide suggestions for in-progress efforts, and test experiments and devices could be quickly fabricated and tried out. The silicon solar PV cell, for instance, was a direct result of Bell Labs maintaining facilities for melting silicon and creating purified silicon ingots.

Bell Labs also benefited from chance historical circumstances, finding itself in the right place at the right time to make major discoveries and create important inventions. Academic research on quantum mechanics in the 1920s and 1930s yielded a huge collection of physical phenomena, and Hitler’s rise in Europe created a steady stream of talented European physicists immigrating to the U.S. As one of the few organizations able to hire physicists at the tail-end of the Depression, Bell Labs was able to acquire top-tier talent in a field that was ripe for exploitation. William Shockley, for instance, noted that job offers “were not just hanging around on trees”: though he applied for jobs at GE and RCA, only Bell Labs offered him a full-time position.

Similarly, AT&T and Bell Labs would grow enormously during WWII, under pressure to fulfill government contracts and develop new weapons. During the war, Bell Labs added several thousand staff and took on more than 1,000 government contracts to develop tank radio sets, enciphering machines, antiaircraft gun directors, and most of all radar sets. When the war ended, Bell Labs emerged as “the greatest invention factory in any field of science that the world had ever seen,” poised to capitalize on the huge backlog of scientific discoveries and technical progress that had taken place during the war.

More generally, over its life, Bell Labs probably accumulated inventions that would have been invented somewhere else in its absence. The many cases of simultaneous invention (the telephone itself being a classic example) show that an invention will often “appear” once the conditions are right and the necessary predecessor technologies are in place. Because Bell Labs was the largest developer of telecommunications equipment, and because the existence of AT&T (a near-monopsony purchaser of said equipment) may have discouraged outsiders to try and invent new telephone technology, many new telephone-related inventions ended up being created at Bell Labs by default.

And of course, Bell Labs managed to create a culture within the labs that was highly conducive to creating novel inventions and discoveries. Theodore Vail believed that science-based technological development was the key to AT&T’s success, and he inculcated these values so successfully within the company that they lasted for decades. Bell Labs researchers had many of the benefits of an academic research environment (freedom to pursue whatever research avenues they saw fit, ability to achieve academic prestige, little risk of termination, ability to devote long periods to speculative projects with unclear payoffs) without many of its drawbacks (needing to spend time teaching or applying for grants). And it offered many benefits, such as higher salaries and better equipment, that academic research environments couldn’t. It wasn’t uncommon that Bell Labs researchers would plan a few-year stay, then find the environment so hospitable that they remained their entire careers. As the reputation of the Labs increased, these benefits increasingly allowed it to attract top talent, which further accelerated its achievements. (For a more thorough look at the culture within Bell Labs, see these pieces by Eric Gilliam.)

Could you recreate these conditions today?

While some of the factors behind Bell Labs’ success seem possible to recreate (though not necessarily easily), others seem impossible, or at best contingent on things outside our control.

In principle, there’s no reason that you couldn’t create a culture like Bell Labs: goal-directed at a high level but with a lot of freedom in how that goal was pursued, highly interdisciplinary, focused on real, practical problems but open to pursuing more “fundamental” research to solve them, minimizing burdens on researchers and maximizing their flexibility. But in practice, such a culture is likely difficult to create deliberately. Some of the things that helped create it at Bell Labs, such as a huge initial success with a science-based technological development project, can’t be willed into existence. Likewise, having a goal attached to a lab is not the same thing as an overriding, urgent need that infuses its culture; the former can simply be dictated, but the latter is dependent on outside forces (like a constantly rising demand for a product) that can’t necessarily be controlled. Similarly, top talent and making a lab highly prestigious is possible in principle but not completely controllable: much of the Labs’ prestige, for instance, came from the invention of the transistor.

And these are likely the easiest parts of the problem. Bell Labs was made possible by a large-scale, vertically integrated telephone monopoly that allowed for an unusually long and wide research and development horizon for an industrial lab. Outside of those conditions (not likely to be repeated), funding a Bell Labs-style operation does not appear to be something most companies are willing to do. Even a company like Google, which spends billions on R&D and has displayed a willingness to fund speculative, longer-term moonshot projects like self-driving cars or life extension, doesn’t completely bite the Bell Labs bullet. Google’s Moonshot projects absorb billions in funding each year, but they tend to be organized as independent companies that raise money outside Google and get spun off when they seem promising enough. Bell Labs also took advantage of historical circumstances: discoveries in quantum mechanics yielded promising new phenomena, and WWII energized the organization while simultaneously creating scientific and technological progress that could later be capitalized on. These contingencies were the result of pure chance, not anything that could be controlled.

And even if all these conditions could be created (or were found to not be critical for success), Bell Labs might have been the product of a particular technological regime that we no longer live in. For much of its existence, AT&T was criticized for adopting technology too slowly (such as automatic dialing and electronic switching), but following WWII it began to face the opposite problem: technology it developed seemed to slip out of its fingers to be exploited by others. In the second half of its life, Bell Labs’ invention machine seemed to be increasingly out of its control. Much of the technology invented at Bell Labs, such as the solar PV cell, the CO2 laser, or the chemistry behind the glow stick, had little to no impact on the Bell System, and was developed and commercialized for other purposes. Other research, such as efforts to create computer-generated music, had little to no practical application at all.

Even technology that did find use in the Bell System, such as the transistor, often left the parent company behind. Bell Labs invented the transistor and many of the early related inventions, but by the 1960s the center of gravity in microelectronics development had begun to move elsewhere; Bell Labs continued to make important inventions (such as molecular-beam epitaxypitaxy for transistor fabrication), but many of the most important inventions, such as the integrated circuit and the microprocessor, would be invented elsewhere. And AT&T itself struggled to introduce the transistor into its system. Though Bell Labs had developed the transistor specifically to replace vacuum tubes and mechanical relays, the first electronic switching center wasn’t deployed until 1964, and the technology was still being rolled out in the 1980s.

Microwave transmission also quickly escaped AT&T’s control. In 1951 AT&T deployed the first microwave transmission telephone system, leveraging work on microwaves it had done during WWII. Less than 20 years later, other companies were challenging AT&T with their own microwave systems, one of which (MCI) would ultimately instigate the downfall of AT&T.

This loss of control was partially due to a 1956 agreement with the Justice Department that forced AT&T to license its patents, but this is at best a partial explanation. After all, AT&T had (eventually) maintained control over the development of its technology after its key patents expired in the 1890s, and after it adopted generous patent licensing terms in the 1930s.

Another reason for the loss of control was a shifting culture within Bell Labs that gave researchers more freedom, and placed less emphasis on eventually turning their work into practical results. Philip Anderson, Bell Labs researcher and winner of the 1977 Nobel Prize in physics, argues that Bell Labs was forced to adjust its culture to retain its most talented researchers. While previous work had to be justified for its possible relevance to the telephone system (though an expansive view of possible relevance was taken), by the 1970s and ‘80s that seemed to no longer be the case. In his memoir about the later years of Bell Labs, Narain Gehani describes the research department as unwilling to cooperate or justify its existence to the business side of the organization at all.

And finally, this loss of control also might simply be the nature of technology itself. In his memoir of his experiences at Bell Labs, Alan Chynoweth argues that the technology that Bell Labs developed ultimately created a world where something like Bell Labs no longer made sense:

Before the discovery of the transistor, telecommunications… was a rather specialized, even arcane branch of engineering. The transistor was to sow the seeds of change. Though it led to a flowering of solid-state science and technology and would eventually revolutionize telecommunications, it also, once it had evolved into integrated circuits and digital technologies, lent itself to revolutionizing almost every other area of electronic engineering, information technology, and product development. It became everyone’s technology, the genie was out of the bottle… nowadays each step along the chain is populated with numerous competing companies specializing in that step so that a company in the next step has a multiplicity of sources of technology from which to choose. In particular, it has made it relatively easy for new competitors to develop their own versions of telecommunications devices, systems, and services. Perhaps it can be said that with the discovery of the transistor, forces were set in motion that would eventually and inevitably lead to Bell Labs losing its unique status.

Jon Gertner likewise posits in The Idea Factory that even in the absence of a justice department lawsuit, the technology that Bell Labs spawned would have ultimately unwound the AT&T monopoly and thus the very thing that allowed Bell Labs to exist.

Bell Labs may thus not only be a product of unique historical circumstances, but unique technological circumstances: telecommunications technology that was best provided and controlled by a single, enormous monopoly, which in turn made Bell Labs possible. The question may not be “How do we recreate Bell Labs?” but “How best can we evolve the technological landscape that we’re faced with today?”

If you're interested in reading more about Bell Labs, a reading list of the best and most useful books and other resources I found on the subject is available here for paid subscribers.

1. Bell Labs has also accumulated three Emmys, two Grammys, and an Oscar.

2. Technically, Vail was general manager of American Bell Telephone; in the 1880s he became president of AT&T, which at the time was a subsidiary of American Bell devoted to long-distance service.

3. Bell Labs’ investigations were wide-ranging in part because it wasn’t known where a new useful technology might come from, but also because AT&T deliberately pursued broad investigations to defensively patent new technologies that might disrupt its monopoly.

4. In part, this was because AT&T, which was responsible for the functioning and reliability of the telephone network, was often conservative about introducing new technology into its system, waiting until its long-term performance was assured.