feature article
Subscribe Now

William Shockley

Dick Selwood reviews

391 San Antonio Road in Mountain View is about to be redeveloped.  Why should this interest you?  Because that is where, one might argue, Silicon Valley started when William Shockley founded his semiconductor company in 1955. To understand the man, Dick Selwood reviews Joel N. Shurkin’s biography, Broken Genius.

William Shockley won the Nobel Prize for inventing the transistor, founded Silicon Valley, was a virulent racist, and donated sperm to a sperm bank for the super intelligent. All of this is true – but often only up to a point.

William Shockley was born in 1910 in London, where his American father had moved to try to restore his fortunes. He grew up in Palo Alto and attended Caltech at the time when it was beginning to build its reputation. At MIT he worked on the electron configuration of atoms, with a PhD thesis on electron movement in sodium. He was immediately recruited to Bell Labs, then a hot bed of research and innovation.

At Bell Labs he was tasked with finding replacements for valves in the telephone switching system of AT&T. Initial work with copper-copper oxide rectifiers was moving to experiments with silicon when, in 1940, he was put to work on radar. At the same time, with a friend and colleague at Bell, James Fisk, he produced a study of the potential of atomic fission for generating energy, including a design for an atom bomb. This report appears to have been suppressed, possibly under pressure from the US government.

When the USA entered the war, Shockley was seconded to the US government, where he worked initially on anti-submarine warfare, calculating the most efficient ways for planes to intercept the U-boats that were attacking the North Atlantic convoys and the best way for convoys to avoid German bombers. He was promoted to become an “Expert Adviser to the Secretary of State” and set to analysing the cost benefit of bombing. Using the techniques of Operations Research, his team assessed whether the cost to the US in training aircrews, building airplanes and bombs, and losing a percentage of those on every raid, was justified by the cost to the German economy of the raids. (It probably was not.) He later worked on similar calculations for atomic bombs. His work was sufficiently important that he was awarded the National Medal of Merit, the highest civilian award, and he continued to act as a consultant to the government for the rest of his life.

After the war, he returned part time to Bell Labs while working as a member of the Joint Research and Development Board and lecturing part time at Princeton. Bell was still looking for a solid state telephone switch, and Shockley was given a multidisciplinary team of 34 men (all men!) and a budget of half a million dollars, big bucks for 1945, to look for one.

British physicist Alan Wilson probably coined the word ‘semi-conductor’ in 1931. By 1945 semi-conductor research was underway at several centres in the US, including Bell Labs, MIT and Purdue University. In particular Purdue had produced what were soon called n-type and p-type semiconductors in germanium using a variety of dopants.

Shockley created a theoretical model of a field effect amplifier using silicon and a pn junction, but it didn’t work in practice. After nearly a year in which the group had got nowhere, John Bardeen realised (in March 1946) that what stopped the amplifier working was a surface effect of trapped electrons, which Shockley had also described.

Working from this basis, Bardeen and Walter Brattain moved from silicon to germanium, using ingots produced at Purdue. By 16 December 1947, a point contact transistor gave a power gain of 4.5 and a fourfold voltage improvement. Although Shockley provided the theory and was Bardeen’s and Brattain’s manager, he was not directly involved. When he was told about the success, according to author Shurkin, he threw what can only be called a sulk, and, after a demonstration to senior management, he never returned to the transistor team.

The patents that Bell filed did not include Shockley’s name, creating further resentment. He was, however, included in all the publicity, perhaps after he threatened to undermine the patents.

Although Shockley was partially reconciled to Bardeen and Brattain when they attended the Nobel Prize ceremony together in 1956, his resentment apparently returned when he heard rumours that some nominations had included only Bardeen and Brattain. Since the meetings of the Nobel Prize Committee are secret, this will never be convincingly known.

Spurred on, perhaps by envy, what Shockley did produce was, in many ways, more important: he came up with a design that ‘employs at least three layers, having different impurity contents,’ an npn sandwich or the junction transistor. Bardeen and Brattain were not allowed near this work, and eventually Bardeen quit and went to the University of Illinois, where his work on superconductors won him a second Nobel, the first person to win two prizes in the same field.

Shockley felt that he was undervalued at Bell: certainly other people were promoted over him, in part, Shurkin feels, because of his very limited people skills. In the summer of 1955 he quit and set up Shockley Semiconductors in Mountain View, the next city to Palo Alto, with funding from Beckman Industries.

Among his first recruits were Gordon Moore and Bob Noyce. As part of his recruitment, Shockley used a New York testing company, which said that Moore and Noyce were both very bright but would never make good managers. Shockley himself was a terrible manager, even though he researched and published on the subject. He saw himself in competition with his staff, argued with them, put them down at meetings, and didn’t understand them. Eventually the senior staff mutinied, and the top eight left and set up a subsidiary of Fairchild. Over time they moved on again, founding, for example, National Semiconductor and Intel, so that by the early 1980s much of Silicon Valley was comprised of “Fairchildren” companies.

Shockley Semiconductors never made money, in part because of Shockley’s insistence on trying to design a four layer diode rather than concentrating on transistors. The company was sold firstly to Clevite and then to ITT, who lost all the staff after a decision to relocate to Florida.

Shockley continued to teach at Stanford, to provide consultancy to the Pentagon, and he resumed consultancy with Bell Labs. He also entered what was the most contentious period of his life.

In January 1965, at a conference that Shockley called ‘the turning point of my life,’ he preached that the human race was threatened by nuclear war, famine and ‘genetic deterioration of the human race…’. He went on to say that he had ‘reduced the problem of intelligence and genetics to the laws of probability and could see things that [his critics] could not.’ Later, in an interview, he said that African Americans as a group had a mean IQ 15% below the mean of whites. Families on welfare should be sterilised so that they did not continue to reproduce and have further generations on welfare, and he agreed that blacks had high incidence on welfare rolls.

The interview appeared in U.S. News and World Report in November 1965 and, when reprinted in a Stanford publication, was immediately repudiated by the Stanford genetics faculty as ‘pseudo-scientific justification for class and race prejudice.’ The faculty went on to attack his call for more research and said that he ignored the need to improve welfare and medical facilities.

The book shows how Shockley became obsessed with “the human quality problem” with his customary complete lack of self-doubt and complete indifference to other people’s views, which in the late 1960s were expressed vehemently. He began recording all his phone and personal conversations, indexing and storing them. Every piece of paper, even Fedex delivery receipts, family birthday cards and laundry lists, was filed, catalogued and indexed.

And at this point his personal life and his professional life merged totally. He married his first wife, Jean, with whom he had three children, while still in grad school. The marriage was stressful, and he later described Jean, in a Playboy interview, as intellectually inferior to him, and he said that his children were a regression.

When he left Bell, he left Jean and soon after married Emily Lanning, who became almost an extension of him and later joined him in the obsessive recording of his life.

She also, presumably, approved of his donating sperm to the Repository for Germinal choice, a sperm bank in San Diego that offered women sperm from men who felt that they were intelligent. Initially the founder, Robert Graham, wanted only Nobel Prize winners, but after attracting only Shockley and two others, he included men with high IQ scores. The scheme was not a great success and was closed, with the deposits destroyed, in 1999.

Some of those who have looked closely at Shockley, not just reacted to him and his views, have suggested that he suffered from Asperger’s Syndrome or from high level autism or even paranoid personality disorder, all syndromes that would account for his obsessions, complete lack of self-doubt and complete indifference to, or perhaps unawareness of, other people’s views and feelings.

Since many engineers can also be diagnosed as having low level Asperger’s (in some areas it is actually called Engineers Syndrome), was Shockley just an extreme version of the guy in the next cubicle, or do we, by extrapolating from the guy in the next cube to Shockley, move into a different domain?

It is easy to say Shockley was a racist, and his views were enthusiastically embraced by many who were indeed out and out racists. It is probably more accurate to say that he was an elitist; even in grad school he felt that the world should be ruled by an intellectual elite and that those of lesser intellects, of whatever race, should be discouraged from breeding. He was aware that the distribution of intelligence predicated by IQ tests meant that many individual African Americans would be more intelligent than the mass of other racial groups, but his critics, many of whom did not understand statistics, did not hear him say this.

It is possible to argue that, even without Shockley, the work going on in Bell and other labs would have produced transistors in the late 1940s. Maybe Silicon Valley, as we now know it, would not have appeared if he had not set up in Mountain View, but something similar might have appeared elsewhere. But what comes through in Shurkin’s book was that Shockley was a deeply difficult and unpleasant person. He turned on friends, who often had no idea what they had done to deserve that treatment, and by the end of his life he had virtually no friends left, except his wife: his children learned of his death only when obituaries appeared in newspapers.

Shurkin is worth reading on Shockley the man, even though the man as described here leaves an unpleasant aftertaste. When Shurkin is weak on the technology you should turn, as Shurkin did, to Michael Riordan and Lillian Hoddeson’s Crystal Fire. This tells the story of the discovery of the transistor and its evolution into the integrated circuit. If you are at all interested in the forces that formed our industry, read both.

Leave a Reply

featured blogs
Aug 19, 2018
Consumer demand for advanced driver assistance and infotainment features are on the rise, opening up a new market for advanced Automotive systems. Automotive Ethernet allows to support more complex computing needs with the use of an Ethernet-based network for connections betw...
Aug 18, 2018
Once upon a time, the Santa Clara Valley was called the Valley of Heart'€™s Delight; the main industry was growing prunes; and there were orchards filled with apricot and cherry trees all over the place. Then in 1955, a future Nobel Prize winner named William Shockley moved...
Aug 17, 2018
Samtec’s growing portfolio of high-performance Silicon-to-Silicon'„¢ Applications Solutions answer the design challenges of routing 56 Gbps signals through a system. However, finding the ideal solution in a single-click probably is an obstacle. Samtec last updated the...
Aug 16, 2018
All of the little details were squared up when the check-plots came out for "final" review. Those same preliminary files were shared with the fab and assembly units and, of course, the vendors have c...
Jul 30, 2018
As discussed in part 1 of this blog post, each instance of an Achronix Speedcore eFPGA in your ASIC or SoC design must be configured after the system powers up because Speedcore eFPGAs employ nonvolatile SRAM technology to store its configuration bits. The time required to pr...