Mother of Invention

High-handed corporate monopoly and high-minded national treasure, the American Telephone and Telegraph Company (AT&T) was a unique project of this country’s pragmatism and for decades the envy of the world in extending low-cost local telephone service.

At the heart of AT&T was its R&D unit, Bell Laboratories, the world’s greatest entity of its kind, and a giant manufacturing arm, Western Electric. Jon Gertner’s The Idea Factory is hardly the first book about AT&T history or the first about Bell Labs research. The company published a massive survey near its peak in 1977 (a second edition appeared in 1983), Engineering and Operations in the Bell System. On the topic of its pure science research, the physicist Jeremy Bernstein’s Three Degrees above Zero (1984) told the story of the discovery of the cosmic background radiation at the Holmdel branch. In 1997, Michael Riordan and Lillian Hoddeson published Crystal Fire: The Birth of the Information Age, an awardwinning history of the lab’s greatest technological triumph, the transistor.

The Idea Factory is still welcome. It is the first study of Bell Labs that puts its history in its full organizational, political, and administrative context. AT&T was a company striving to expand and maintain a privileged empire under a government that saw it alternatively as a trusted military/industrial partner and an anticompetitive threat. This ambiguous embrace, Gertner suggests, inadvertently encouraged a culture that combined a gifted and diverse workforce with a long-term outlook, creating the foundations of a new information economy, which in turn made radical changes in the charter of the parent company inevitable.

Gertner’s story is the interaction between three leaders of Bell Labs in its critical years—Mervin Kelly, Jim Fisk, and William Oliver Baker—and three of its greatest scientific minds: William Shockley, Claude Shannon, and John Pierce. Of all these men, Kelly may have been the most influential. When Bell Labs was established in 1925, many vital components of the telephone system needed radical upgrading if the network was to continue growing. Bell Labs had to develop and constantly improve its own equipment, such as the manufacture and sheathing of cables and wires. It had to invent and build its own testing apparatus. Even the leather belts of telephone line workers were rigorously studied and specified. The integration of R&D with Western Electric manufacturing ensured that challenges in supply chains and industrial processes could be met at early stages.

For an ultimately revolutionary organization, Bell Labs had remarkably conservative values: Equipment was designed, manufactured, and tested to last for 40 years. By today’s Wall Street and Silicon Valley standards, it was also remarkably egalitarian. The best-paid staff members earned no more than 10 times the annual wages of the lowest paid. Yet it was also far from a civilservice culture. Kelly’s crucial move after the end of World War II was a virtual coup, demoting experienced managers so that new groups led by younger researchers such as William Shockley could take charge, reducing at least one of the veterans to tears. Many of the materials needed did not exist yet. Shockley’s fixation on taking personal credit, so contrary to the collegial norms of the Labs, was also tolerated because of his ability to catalyze ideas developed by others.

The transistor also illustrated how the Bell system’s sensitive political position encouraged rapid diffusion of its inventions. In the midst of today’s patent wars, it’s striking to see how AT&T managers licensed what Gertner and others have considered the greatest invention of the 20th century. Ceremoniously distributing samples around the world, they made the technology available to all manufacturers for a license fee of $25,000, almost a token amount considering the development expenses and the profits to be made. It was the Bell system’s monopoly status that encouraged such generosity, justifying their view of their own company as a public resource rather than a selfish old-style trust.

The Bell system could not deploy the transistor without another revolutionary innovation that the Labs were developing at around the same time: the theory of information, which enabled engineers to ensure the highest volume and quality of transmission possible. It was one of the Labs’ most gifted researchers, Claude Shannon, who in 1948 published one of the most influential mathematical papers of all time, “A Mathematical Theory of Communication.” It was Shannon who coined the word and concept of a bit and who showed not only how ordinary communication was full of redundancy that could be suppressed for the sake of speed, but also how the addition of redundant information in the form of error-correcting codes could overcome glitches in transmission. Shannon’s work, theoretical as it was, also owed much to the Labs’ military contracts. He had been a major figure in the development of cryptography and in the automation of counterattacks against German missiles attacking London during World War II.

The Cold War was also a strategic boon for Bell Labs, helping deflect planned antitrust actions against the Western Electric monopoly on AT&T equipment supply. By performing vital work on the Distant Early Warning missile detection system and Nike missile systems and by managing Sandia Laboratories, the Bell system could continue arguing for a privileged legal position as a national strategic asset. One of Mervin Kelly’s main challenges, as he saw it, was to balance the expansion of civilian telecommunication service, and its role in prosperity and economic growth, with the company’s military role.

Not all of Bell Labs’ investments paid off. Although managed by one of the most brilliant members of the Labs, John Pierce, the satellite communication program was doomed when the federal government decided to create COMSAT in 1962, excluding AT&T. The Picturephone seemed to be the future of business and ultimately residential service, after Bell Labs’ success in reducing the costs of what were originally elite services, but the service never reached a critical mass of users. And AT&T was slow to recognize the value of fiber optics, originally developed in Europe, partly because AT&T was focused on long-distance and Picturephone transmissions, for which it believed hollow pipes (waveguides) were more practical. Gertner shows how the Labs, for all their maverick genius, were not immune to corporate inertia and the tyranny of sunk costs in older systems.

Still, Gertner may be too harsh in judging the Picturephone only as a fiasco. The historian Kenneth Lipartito has suggested that the fate of major inventions is subject not to an all-knowing market but to many contingencies. The Picturephone initiative might be regarded as a kind of proto-Web, designed for sharing images and documents as much as for face-to-face electronic conversations. Although the Picturephone may not have had a chance to become more than a niche product, the project was like many other failures in creating the conditions for ultimate successes. The real point is not that the Picturephone was a folly, but that even by the 1960s, the acceptance of technological innovation had become so complex that it could no longer be planned reliably, even by the best minds.

The paradox of Bell Labs in its original form was that its very success upset the sensitive equilibrium of its charter. The greater the scope of its products and services, the more would-be competitors could cry foul. As early as 1943, Kelly wrote to his colleagues that despite the Bell system’s conservative philosophy, “our basic technology is becoming increasingly similar to that of a high-value, annual model, highly competitive, young, vigorous and growing industry.” As the management guru Peter Drucker wrote in 1984, the applications of the Labs’ innovations were beyond the ability of any one company to realize. One scenario Drucker imagined beyond the breakup—an independent, self-supporting laboratory licensing patents to all companies—seemed too visionary even to Drucker. As Gertner notes, some of the Labs’ greatest breakthroughs were responses to challenges of telephone service, not developments of pure science. In fact, Bell Labs survived through repeated reorganization, retaining its ties to AT&T and the operating companies and enjoying spectacular prosperity as Lucent in the 1990s, before the collapse of the dotcom boom destroyed most of its market value.

Key omissions

As a history of some of Bell Labs’ greatest ideas and an analysis of the company’s strengths and shortcomings, The Idea Factory is the best all-around account yet. Yet it slights some individuals and innovations that today’s software industry considers among the Labs’ greatest contributions: the UNIX operating system and the C programming language, both credited in part to the late computer scientist Dennis Ritchie. Nor does it describe the remarkable work of the Labs in human factors, such as the remarkable succession of handsets designed in consultation with Henry Dreyfuss, which were so robust and comfortable to use that many are still in service. Dreyfuss’s designs for the Bell System, realized in collaboration with the Labs and Western Electric, remain a foundation of today’s human/electronic interfaces.

On balance, Gertner sees Bell Labs as part of a bygone business model that cannot and should not be revived in an age of rapid, consumer-oriented electronic product development and global supply chains. He cites a 1995 paper by a Bell Labs mathematician, Andrew Odlyzko, suggesting that dominating narrow market segments, not pursuing breakthrough inventions, was now the main way to corporate profits. Incremental improvements rather than giant steps were the way to go.

That remains an accurate description of the current research environment. But is this situation entirely inevitable? Breakthrough industrial research continues. IBM’s Watson Research Center’s artificial intelligence program, for example, is on a par with the tradition of the old Bell Labs. Even Toyota, then considered a conservative “fast follower” company, created the radically new Prius hybrid car in only four years in the mid-1990s. Gertner might have also have considered the vigorous academic debate on whether markets promote underinvestment in research, dating from Kenneth Arrow’s paper “Economic Welfare and the Allocation of Resources for Invention” 50 years ago. If government rules and constraints stimulated creativity at Bell Labs, isn’t it possible that changes in the tax code to shift rewards to longterm investment might benefit not only shareholders of individual companies but the broader world economy?

Yet in one respect, Gertner is right: The Labs were a unique creature of their times. As he points out, they concentrated a generation of brilliant, ambitious, often eccentric young men from the ranks of 1920s Midwestern rural and small-town tinkerers. During the Depression, the Labs were able to assemble a critical mass of sheer talent, especially because they paid up to twice academic salaries at a time when few universities were hiring. Hard times may even have accelerated researchers’ ideas. Gertner observes that when hours were reduced to save money, staff members used their extra time to audit courses taught by famous physicists at Columbia University, a subway ride from the Labs’ original location in lower Manhattan. Mervin Kelly and his successors also understood how to manage this diverse cohort and above all how to promote an ethic of generally unstinting cooperation, open doors, and what is now called mentoring. Its only parallel in academia was the Massachusetts Institute of Technology’s Building 20 after World War II. As Gertner observes, the design of the Holmdel laboratory, with its vast atrium and exterior corridors with views of the countryside taking the place of the more spontaneous interactions of the Murray Hill campus, was already beginning to undermine the ethos of the Labs. Universities, for all their vaunted interdisciplinary initiatives, are still organized around proudly independent departments.

The wonder of Bell Labs is that it could be so focused on carefully specified corporate goals in expanding its network, yet so open to chance encounters (in hiring, too) and fortunate accidents. The study of the organization’s past will thus remain an indispensable resource for thinking about the future of serendipity.