Absent at the Creation; How one scientist made off with the biggest invention since the light bulb
By Ronald Kessler
The Washington Post Magazine
April 06, 1997, Pg. 16
At office parties William Shockley would perform magic tricks, pulling red balls, coins or flowers out of people's noses or ears. He would announce that he knew how to stop inflation, then take out a cigarette lighter and set fire to $ 1 bills. In front of his wife, he would endorse prostitution as a solution to marital boredom.
If Shockley had his quirks, he was also brilliant: He worked at Bell Telephone Laboratories, probably the world's preeminent industrial research lab. He was a physicist himself, and he supervised other physicists. He would, in time, receive the Nobel Prize. When the 50th anniversary of the transistor is celebrated this December, his name will surely be invoked as the father of that invention.
The anniversary is worth marking, because the transistor has changed daily life as broadly and profoundly as the light bulb or the telephone did before it. Baby boomers may remember the novelty of the pocket-size transistor radios introduced in the mid-1950s. They were just a harbinger of changes that would bring better televisions and tape recorders, plus pocket calculators, compact disc players and videocassette recorders. But transistors did more than make life richer for couch potatoes; they provided the basis for what Isaac Asimov termed "perhaps the most astonishing revolution of all the scientific revolutions that have taken place in human history."
Transistors made it easier to move electricity through transmission lines, sound through hearing aids, money through securities exchanges. Because of transistors, car engines start more efficiently, police and fire dispatch systems work more quickly, erratic heartbeats can be made more constant. Without transistors, missiles and spacecraft would be concepts out of Jules Verne, and the Computer Age -- the age of the modem, the database, the fax, e-mail, voice mail, the Internet -- would not be at hand.
"If we didn't know about nuclear energy and atom bombs, what difference would it make to everyday life? None," says Nick Holonyak Jr., a professor of engineering and physics at the University of Illinois and an expert on the development of the transistor. "But if we didn't have transistors, we would be in deep trouble."
Anyone who invented such a device might properly be as renowned as Alexander Graham Bell and Thomas Alva Edison. And William Bradford Shockley certainly has been given credit: At various times, the New York Times, Forbes, Business Week, the Economist and Scientific American, among many others, have named him as the sole inventor of the transistor. Such reference works as the Chronicle of America and Modern Scientists and Engineers do the same. Other reference works, including Collier's Encyclopedia, Webster's New World Encyclopedia, the Timetables of Science, the Cambridge Biographical Encyclopedia and the Encyclopedia of Science and Technology, list Shockley as the co-inventor.
The trouble is, credit has been given where it is not due. Shockley was indeed a gifted physicist, but his greater gift may have been for self-promotion. In the interim between the telephone and the transistor, scientific achievement became a great deal more complex. So did the machinery of fame. The careers of Bell and Edison inspired a cultural icon of the inventor as a solitary genius, a cerebral pioneer expanding the frontiers of human knowledge. What Shockley's career shows is that once those frontiers grew big enough, there was plenty of room for a man to poach his way into history.
In the mid-1930s, the American Telephone and Telegraph Co. had a problem: Demand was overtaking its technology. If telephone use continued to rise at its present pace, for instance, AT&T would have to employ half the people in the United States as long-distance operators. The company, which held a monopoly on local and long-distance telephone service, had to find a new switching system -- a new means of routing calls.
At that time, the state of electronic technology was defined by the vacuum tube, which had been developed 30 years before by an American electrical engineer named Lee De Forest. De Forest discovered that by placing an electrified wire grid across a stream of electrons in a vacuum, he could alter or manipulate the flow of electrons in a number of ways: The current could be interrupted, reduced, stepped up or stopped entirely. Because of this, vacuum tubes could pull faint electromagnetic signals out of the air, magnify them and convert them to sound or pictures. They made commercial radio and television possible.
Vacuum tubes could also act as electrical switches, but they had certain drawbacks. They wouldn't work until they warmed up. Once they did, they generated heat as high as 400 degrees. Because the filaments became so hot, they often broke. A black-and-white television set required two dozen vacuum tubes. The University of Pennsylvania's ENIAC computer, the world's first electronic computer, required 18,850 tubes. It weighed 30 tons, filled a 30-by-50-foot room, required extensive refrigeration, used tremendous amounts of power -- and had less speed and capacity than today's hand-held calculator. ENIAC also crashed a lot: Moths, attracted by the glowing tubes, kept gumming up the circuits. They were the original computer "bugs."
As the 1940s began, AT&T's switching system consisted of all those long-distance operators and, for local calls, hundreds of thousands of electromechanical switches -- devices that actually moved, on or off, in response to an electrical jolt. The company knew it needed a new system, but by 1945 had concluded that the vacuum tube would never do. The company turned to its research arm, Bell Telephone Laboratories.
Funded by the limitless reserves AT&T amassed as a result of its monopoly on phone service, Bell Labs could afford to engage in pure research that might or might not find some practical application in the telephone industry. "There was more academic freedom at Bell Labs' research department than at most universities," recalls Charles Kittel, a physicist who worked there from 1947 to 1951. "You could do what you wanted to. There was very little pressure to do anything practical."
Mervin J. Kelly, Bell Labs' director of research, believed that crystal semiconductors might hold the answer. Because of their atomic makeup, such crystals as germanium and silicon are normally poor carriers of electricity, but they can become conductive when they are "doped" with impurities and voltage is applied to them in particular ways. "Free" electrons -- those that are not attracted to and paired with other charges -- will wander toward or away from electrodes as the electrode charge is changed from positive to negative. This attraction-repulsion effect would allow a semiconductor to act as a switch that alternates between a conductive state and an insulating state.
Such a device would be a true breakthrough: The switching would take place not through a mechanism, but within the atomic structure of matter itself.
It was Kelly who assigned William Shockley to work on an alternative to the vacuum tube. Born in London in 1910, Shockley had graduated from the California Institute of Technology and, in 1936, earned a PhD in physics from the Massachusetts Institute of Technology. After finishing MIT, he went to work for Bell Labs. During World War II, he briefly helped direct the Navy's research on anti-submarine operations. But in 1945 he returned to Bell Labs' campus in Murray Hill, N.J., where he headed the solid-state physics group. He was skilled at explaining complex scientific concepts to lay people, he conveyed tremendous enthusiasm and he enjoyed the full support of the labs' management. Among his peers, however, his reputation was mixed.
"He was a very smart guy," says Charles H. Townes, another physicist who was at Bell Labs at the time and who later won the Nobel Prize for inventing the laser. "He understood everything except people. He was a very competitive person in general."
"Shockley was very quick mentally," says Conyers Herring, another Bell scientist at the time, whose intellectual prowess is widely esteemed by his fellow solid-state physicists. ". . . He was always a jump ahead of me, and it was difficult to persuade him of anything. He realized his own superiority. He always felt his own way of looking at things was better than anyone else's. Nine times out of 10 it was, but the 10th time got him in trouble because he didn't study the literature sufficiently carefully or didn't accept ideas from people who didn't explain themselves well enough to him."
As a rule, group heads at Bell Labs spent about half their time conducting their own research and the rest supervising 10 to 15 other scientists. Shockley, a theorist, designed what was then called a field-effect amplifier (and later called a transistor), but it did not work. As he went on to other projects, two senior physicists who worked for him, John Bardeen and Walter H. Brattain, took over.
Bardeen was, like Shockley, a theorist. He was born in 1908, in Madison, Wis., and proved to be a child prodigy, skipping from third grade to seventh. He earned undergraduate and master's degrees at the University of Wisconsin and a PhD from Princeton in 1936. Bardeen met Shockley while he was a fellow at Harvard and Shockley was at MIT. In fact, Shockley recruited him for Bell, where he was known as "whispering John" because of his soft voice and self-effacing manner. His greatest show of emotion was on the golf course, where he occasionally cursed after making a bad shot.
Brattain, a craggy descendant of farmers who had crossed the country in covered wagons, was an experimentalist -- he built apparatus that tested Shockley's and then Bardeen's ideas. Born to American parents in Amoy, China, in 1902, he received an undergraduate degree from Whitman College, a master's from the University of Oregon and a PhD from the University of Minnesota in 1929. He'd met Bardeen before Bardeen arrived at Bell Labs in 1945; his brother, R. Robert Brattain, also a physicist, had attended Princeton with Bardeen.
If Bardeen and Brattain were not the showmen that Shockley was, they also did not have his supreme ego. They listened to others and at times displayed a deeper understanding of the laws of physics. In Robert Brattain's telling, Bardeen "looked at some earlier work Shockley had done, and he decided he would make different assumptions . . . He went to my brother and asked if he could check this theory out with an experiment."
After trying several new approaches, Brattain and Bardeen developed what is known as the point-contact transistor. On the afternoon of December 16, 1947, the two men, using their new device, were able to amplify sound. As in many scientific breakthroughs, there was an element of serendipity. Trying to rectify a short circuit in the device, Brattain changed the positions of the electrodes. Suddenly, he detected a signal. Using a second, improved device, he could hear sound being amplified. Bardeen studied the arrangement and articulated why it worked.
By December 23, the two had improved the device and demonstrated it to Bell Labs management, including Shockley. The scientists recorded their discovery on page 708 of Bell Labs notebook No. 21780. The transistor had been invented. That is the anniversary the scientific community will celebrate this year.
One evening, Bardeen drove home and parked his car in his garage, just as he always did. His wife, Jane, was preparing dinner. As if discussing the weather, he said, "We discovered something today."
"They all saw right away that there was a Nobel Prize in this," recalls Kittel, who is now professor emeritus of physics at the University of California at Berkeley and a member of the National Academy of Sciences. "Shockley, Bardeen, Brattain and the organization saw that this was absolutely Nobel Prize material."
A colleague of Brattain's, John Pierce, coined the term for the new device by playing around with syllables from the names of existing electronic components. The transistor, unlike the vacuum tube, drew barely any power, generated little heat and required no time to warm up. It would last practically forever. While initially more expensive to manufacture than tubes, transistors would soon become far cheaper. Most important, as they were developed further and connected in what became known as integrated circuits, transistors performed more work while occupying millions of times less space.
But if the small community of Bell Labs scientists understood what Bardeen and Brattain's device portended, the larger world did not. When the lab announced its invention to the press, on June 30, 1948, few papers picked it up. The New York Times, in its issue of July 1, buried the news in the 10th and final item of a News of Radio column. It followed the announcement that one George Shackley's "Anthem for Brotherhood" would be performed the following Sunday on television station WPIX. It did not mention the two inventors, Bardeen and Brattain.
In the meantime, says Kittel, "Shockley meant to get in on the [Nobel] prize." In this he had a problem: He had been left out of the discovery. When the breakthrough occurred, Townes says, Shockley was in Europe. He was said to be working on the mechanical properties of solids. Still, he maneuvered to take credit for the new invention. It was he who dubbed December 1947 "the magic month."
"There was a change in [Shockley's] personality after the discovery of the transistor," says Herring, who later joined the faculty at Stanford University. "He realized he was extremely brilliant, and he looked forward to acquiring a great reputation. Here was the most important development to come out of the laboratory that he was not in on directly."
One of Shockley's first moves, according to what Bardeen told Nick Holonyak, was to apply his considerable institutional clout to persuade the Bell Labs patent attorney to include him on the transistor patent applications. But fearing that listing a bogus inventor would impair the lab's claim to the invention, the attorney refused. When the two patents were issued, on October 3, 1950, one was in Bardeen's and Brattain's names, the other in Bardeen's alone. The men assigned them to Bell Labs.
True, Shockley acknowledged early on that the original invention was not his. In the preface to his 1950 book on the subject, he wrote that the tome "owes its existence basically to the invention of the transistor by J. Bardeen and W. Brattain." In December 1950, Shockley inscribed Bardeen's copy of the book: "To John Bardeen. Who made a book like this a need."
Even before then, however, Shockley had launched a media offensive. As Bardeen and Brattain's supervisor, he designated himself Bell Labs' spokesman on transistors. He approved a nationwide publicity tour that featured himself as well as the other two. For the stock Bell Labs photos of the inventors, Shockley arranged to place himself in the center of an experiment with Bardeen and Brattain looking on, and the caption said that all three had invented the transistor. When one of the photos appeared in an electronics magazine, Holonyak recalls, Bardeen said, "Boy, Walter hates this picture." When Holonyak asked why, Bardeen shook his head, made a face and said, "That's Walter's apparatus and our experiment, and Bill didn't have anything to do with it.' "
Kittel believes that as time went on, Shockley became more aggressive in claiming credit for the invention. Articles based on interviews with Shockley, and Shockley's own writing as well, tended to minimize Bardeen and Brattain and emphasize Shockley's role, including his own initial failure. Claiming the transistor owed its invention to "creative-failure methodology," he told reporters grandly, "A basic truth that the history of the creation of the transistor reveals is that the foundations of transistor electronics were created by making errors and following hunches that failed to give what was expected."
Bell Labs was not about to correct the record. "Management was sold on Shockley," says Herring. "They could see how brilliant he was. He had this way of explaining things with simple language. Bardeen was also appreciated, but I don't think quite as much as he should have been."
Meanwhile, having failed at the patent attorney's office, Shockley redoubled his efforts in the lab. He knew that one way to get his name on the Nobel Prize would be to make a significant contribution of his own to transistor development. He also knew, from discussions with his subordinates Bardeen and Brattain, that the next improvement lay in what would become known as the junction transistor, a device that could be more reliably manufactured than its predecessor.
Shockley removed Bardeen and Brattain from the project and kept it for himself, consulting them only when he ran into obstacles.
"He drew a line saying he was going to be responsible for the junction transistor," recalls Kittel. "He told Bardeen and Brattain to stay on the point-contact transistor. Through monopolizing it for a couple of years, he was in control of the field. He had probably the strongest laboratory in the world to help him in problems that came up."
Within two years, he had developed the junction transistor himself. The junction transistor, along with the field-effect transistor made workable by new technology, is used in computers today.
By then, he had driven Bardeen out of Bell Labs.
In a memo to Mervin Kelly on July 24, 1951, Bardeen explained his decision to take a position at the University of Illinois at Urbana-Champaign. The memo -- which has not been published before -- was provided by Bardeen's son William A. Bardeen, a physicist at the Fermi National Accelerator Lab.
Bardeen wrote that he was "dissatisfied" because Shockley had allowed him and Brattain to work on the further development of the transistor "only as he thought of problems of his own that he wanted investigated experimentally . . . In short, he used the group largely to exploit his own ideas."
Since Bardeen, like Shockley, was a theorist, he could not contribute to the program "unless I wanted to work in direct competition with my supervisor, an intolerable situation."
Bardeen closed by saying he was open to having further discussions about the matter. Apparently, no one at Bell was interested.
"I think the management was kind of scared of [Shockley]," says Kittel, "possibly because they saw the transistor was the hottest thing they had, and Shockley was the only one on the research side who was really interested in the commercial applications of the transistor."
And now, having improved upon the original invention, Shockley was in a position to share a Nobel legitimately with Bardeen and Brattain. Indeed, Kittel was one of those who nominated all three men for the prize. He says that even though Bardeen and Brattain invented the transistor, and even though Shockley improved upon it only because he shoved the other scientists aside, Shockley still deserved the award for his work.
On December 10, 1956, all three men received the prize in physics from King Gustav VI of Sweden.
By then, only Brattain remained at Bell Labs. He spent the rest of his research career in Murray Hill, retiring in 1967 and returning to Whitman College in Washington state to teach. He died in 1987, at the age of 85.
Bardeen spent the rest of his career at the University of Illinois, where, for the theory of superconductivity, he won a second Nobel in 1972. He is the only physicist to have done so, and many scientists consider him in the same rank as Einstein and Fermi. He died in 1991, at the age of 82.
Today, the names Bardeen and Brattain mean about as much to the world as that of John F. Blondel. His contribution was a device for making holes in doughnuts.
Bell Labs, meanwhile, has kept its reputation, though not its corporate parent. Last year, AT&T spun off the labs and other operations into Lucent Technologies Inc. A spokesman confirmed the account of the transistor's invention given by Townes, Herring, Kittel, Holonyak and others -- and documented in the patent applications.
Shockley, for his part, rode a second wave of celebrity, though it could hardly have been as satisfying as the first.
In 1955, he left Bell Labs to create the Shockley Semiconductor Laboratory, a transistor-making unit within Beckman Instruments Inc. He enlisted some immensely talented employees, but because of the imperious way he treated them, most of them left -- and, in 1957, founded Fairchild Semiconductor Co., which in turn spawned Intel Corp. and dozens of other companies that are now known collectively as Silicon Valley. In 1960, Clevite Corp. acquired Shockley Semiconductor, and Shockley became a professor at Stanford University. He continued as a consultant to Bell Labs until 1975.
Much to the disgust of his colleagues, Shockley began espousing -- both in scientific papers and in interviews -- the theory that black people are genetically inferior to whites. He advocated paying people who had low IQs or genetically transmitted diseases to submit to sterilization.
Emmy Lanning Shockley, a psychiatric nurse whom Shockley married in 1955 after divorcing his first wife, says an article about a teenager who had been hired to blind a delicatessen owner precipitated her husband's interest in the subject. "This black teenager had an IQ of 70 or so. The mother had an IQ of 55. She had 12 or 15 children and couldn't remember the names of all her children. Dr. Shockley was shocked and said this sort of thing shouldn't be going on."
Shockley repulsed his peers at the National Academy of Sciences by trying to persuade them to take a stand on the issue. In 1981, after a black reporter for the Atlanta Constitution compared Shockley's theories to Hitler's, Shockley filed a libel suit. While he won, the award for damages was only $ 1.
In the last decade of his life, Shockley reached again for immortality. Robert K. Graham, a physicist who had amassed a fortune by developing shatterproof plastic eyeglass lenses, approached him about contributing to the Repository for Germinal Choice, a sperm bank for geniuses that Graham had founded in California. Shockley readily agreed. He began expressing open disappointment with his children from his first marriage. While his elder son, William, had dropped out of college, his daughter, Alison, had graduated from Radcliffe, and his younger son, Richard, had graduated with distinction from Stanford and earned a PhD in physics from the University of Southern California.
"The children regressed to the mean," says Emmy Shockley. "That means they don't measure up to the total intelligence of the parents."
"It was not a closely knit family," says Alison Iannelli, Shockley's daughter. William Shockley says he last talked with his father roughly 10 years before his death.
By the time that death occurred, in 1989, when Shockley was 79, he had taken to comparing himself to Galileo, Darwin and Einstein. Yet his final attempts to pass on his genius have failed: "So far as we know, we have no offspring from Shockley," Graham said in an interview.
That conversation came a few years before Graham's own death, on February 13 of this year. In its obituary for him, the New York Times noted that Graham had said that four Nobel laureates had donated to his repository. "But the only one who identified himself," the obituary said, "was Dr. William B. Shockley, inventor of the transistor.".
Ronald Kessler is a former reporter for The Post and the Wall Street Journal and the author of 11 books.