True Genius: The Life and Science of John Bardeen: The Only Winner of Two Nobel Prizes in Physics (2002)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

17 Epilogue: True Genius and How to Cultivate It F or centuries the question of genius has been a subject of mys- tery and debate. Such fundamental questions as “How does genius relate to creativity?,” “Is genius hereditary?,” or “Can genius be cultivated?” have been surprisingly difficult to answer. Even today there is little agreement in the vast literature that has grown up around such questions of creativity and genius. Yet within that contentious discussion are pockets of agreement, the beginnings of a modern, more realistic picture of genius. John Bardeen’s life and work offer a basis for discussing it. We need to be careful in presenting the results, for the research is incomplete. The picture is still hazy. And Bardeen is only one example. To gain a more complete account will require compari- sons across many cases. Even Bardeen’s work has not yet been fully mined with respect to questions of genius or creativity, for only in the last years of writing this book did we recognize that it could help to address such questions. Most scholars who work on creativity agree on three aspects of genius. First, there is a real and substantial distinction between a “true genius” who makes lasting contributions to culture and the fictional archetype of a genius, amply represented in popular cul- ture by such figures as Dr. Frankenstein or Will Hunting. In the group of scholars who had recognized this distinction by the early twentieth century are the biologist, eugenicist, and statistician 314 

Epilogue: True Genius and How to Cultivate It 315 Francis Galton, Charles Darwin’s first cousin, and the psychologist Lewis Terman, the father of intelligence quotient (IQ) testing. Galton had hoped to improve human creativity through selec- tive breeding that would enhance intelligence and other “good” characteristics. He initiated the systematic study of human creativity by developing statistical methods for evaluating indi- vidual differences. Terman built on Galton’s work, designing more sophisticated quantitative tests for studying creativity. An out- growth of his famous tests conducted at Stanford University in the early 1920s was the extensive study conducted by Terman’s student Catherine Cox on the future success of three hundred gifted chil- dren having superior IQ scores. A second general agreement among scholars of creativity is that genius and creativity must refer to a particular domain having a well-defined knowledge structure. Thus Mozart was a genius with respect to music, Newton a genius with respect to physics. This association with a domain brings genius and creativity into direct contact with culture and society, for a group of experts in the domain must agree that an achievement adds something new and important to it. If not, the idea or act can have no lasting impact. Such “gatekeepers” of knowledge are united by their accep- tance of the “paradigms” of their domain, a term that Thomas S. Kuhn introduced into the study of the history of science about four decades ago when he pointed out that in any given period scientific fields are characterized by a set of agreements about how knowl- edge is produced in that field. This description of a field in terms of its paradigms subsequently entered many fields of knowledge and allows us to define what a highly creative act is, namely one that creates, replaces, or restructures a field’s paradigms. The degree of restructuring can vary a great deal. Bardeen’s creativity in physics was far more limited in scope than that of Newton or Einstein, but it nevertheless brought significant paradigm shifts into the exist- ing structure of knowledge in solid-state physics, the domain in which Bardeen worked for almost his entire career. The third agreement in the field of creativity, not yet as wide- spread as the other two, describes a genius in terms of a profile, a configuration of heterogeneous features that support the highest levels of creative work in a field. Unlike the popular stereotype, with its familiar features (e.g., otherworldly, alone, charismatic, eccentric, unbalanced, self-taught, etc.), the profile of the true 

316 TRUE GENIUS geniuses who live and work in the real world includes features such as intelligence, passion, confidence, focus, perseverence, and the habit of breaking down problems into smaller parts. While many of the features are similar when we compare individuals—just as dif- ferent human noses or lips are similar to one another—the constel- lations they form are as individual as a human face. This notion of a genius profile idea is not a new one. Even Cox, whose work stressed genetic giftedness, conceded that “high intel- lectual traits” are not alone responsible for the eminence of her subjects. They achieve success “also by persistence of motive and effort, confidence in their abilities, and great strength or force of character.” What is new about the genius profile is the wide agree- ment the idea is coming to enjoy. As this conception offers fertile research opportunities, we will examine this profile in more detail, drawing on Bardeen’s life and science. Genius and exceptional creativity can be represented in three quali- tatively different dimensions: (1) an individual (or personal) dimension, which includes features such as talent and motivation that refer to the particular domain in which the person worked; (2) a methodological dimension, which includes the variety of problem-solving approaches such people employ in their domain; and (3) a contextual dimension, which includes environmental factors, such as family and education. These dimensions are not independent of one another. For example, Bardeen’s family strongly influenced his motivation and values. But examining the dimen- sions separately gives us the basis for studying them more closely. Most of the features in the personal dimension have been hotly debated over the years. For instance, while almost everyone acknowledges the importance of talent and intelligence, there is no general agreement about whether these traits are genetically deter- mined. By any accounting Bardeen was talented in the area of math- ematics; his outstanding ability in mathematics was apparent to his parents and teachers when he was a child. But it is not yet possible to say whether this talent derived solely, or even prima- rily, from genetics, as Galton, Terman, and Cox would have argued, or from his having been exposed to the field of mathematics over a long period of time. Today most scholars would agree that such a talent is probably 

Epilogue: True Genius and How to Cultivate It 317 based at least partially on genetic ability. Yet startling experiments have shown that people with rather ordinary ability in areas such as chess or music can be trained to achieve master-level perfor- mance by subjecting them to the right kind and amount of training over a long period of time, usually about ten years. There is indeed some evidence that Bardeen exposed himself massively to math- ematics from his young years, or perhaps as a result of his enthusi- asm for the subject. Talent appears to be based on both genetics and training. While early in the twentieth century Terman and others as- sumed that IQ was the necessary prerequisite for creativity, much of the current literature about creativity raises questions about the role of inherited intelligence, at least as measured by standard tests. Feynman, who was obviously highly creative, is said to have had an IQ of 129, which, while well above average, is not in the high range of most recognized geniuses. Some psychologists have sug- gested that creativity has more to do with having the right kind of intelligence for creative work in a particular domain. For instance, the kind of intelligence that Bardeen had made it possible for him to “see” the electrons in solids. In the same camp with those who believe that talent is largely a function of exposure are those who emphasize a person’s passion or motivation for work in a domain. The great teacher Confucius emphasized passion: “Only one who bursts with eagerness do I instruct; only one who bubbles with excitement, do I enlighten,” he wrote. The psychologist Teresa Amabile and her students insist that “intrinsic motivation” (a quality referred to even by members of the opposite camp, including Galton) fuels the massive invest- ment in learning about a given domain, an exposure they believe is the true basis for talent. From this point of view, self-help gurus who advise following one’s passions are not wrong, because passion inspires exposure and the drive for excellence, which, in turn, often sends a person looking for supportive environments. From the fact that motivation can be cultivated through positive reinforcement, it follows that talent can too. Although this argument opposes genetic determinism, it does not rule out the possibility that a genetic component figures importantly in an individual’s potential abilities. Passion is a compelling candidate for the source of creativity because it can be viewed as the source of most other qualities in 

318 TRUE GENIUS the individual dimension of the genius profile, such as the will to learn, perseverance, focus, confidence, or the drive to achieve mastery. Bardeen had strong passion for physics and mathematics, and he also possessed the complex of traits that one can expect passion to bring. He learned early on that persevering and main- taining his focus on difficult problems, even in the face of frustra- tion, often brings enormous rewards. Through his experience he learned that struggling with difficult mathematics problems builds the intellectual muscles and confidence for doing creative work in mathematics. He probably did the most to develop himself into a world-class physicist during his time at Harvard when he struggled almost constantly with problems that were then intractable. The confidence and the intellectual strength he developed as a physi- cist and mathematician also supported his ability to take measured risks. Bardeen’s intense focus on the problems he engaged with would sometimes unnerve his associates, especially when his attention wandered off into what appeared a meditative trance—sometimes right in the middle of a conversation. Schrieffer thought that at the times when Bardeen was on “cloud nine,” he was letting nature “flow into his being,” saturating his consciousness with informa- tion about the system he was studying. By coupling this intense focus with persistence and his years of training and experience, Bardeen nurtured his legendary “intuition” for physics. Bardeen’s communion with physics problems was not unlike the riveted attention a young child focuses on something of inter- est. Alison Gopnik and others have compared scientists to grown- up children. Howard Gardner has correlated the single-mindedness required in creative work with a retention of childish ways of focusing on how the world works, a notion that Einstein also touched on in his autobiography. Bardeen’s rapport with children may well have been related to his ability to attach his intense child- like focus to physics problems. Some research suggests that childhood traumas, even such dras- tic ones as the death of a parent, can actually benefit creativity. Bardeen was twelve when his mother, Althea, died. Rather than crushing him, confronting that painful experience appears to have enhanced his inner strength. Similarly, the marginalization he experienced in both grade school and high school may have helped him learn how to stand apart from the crowd. In times of emo- 

Epilogue: True Genius and How to Cultivate It 319 tional stress, Bardeen often turned to mathematics and physics, which offered controllable worlds in which to escape from the pain in his personal or social life. Sports offered other controllable worlds—as well as a multi- tude of additional benefits, including stress relief, physical exer- cise, social interaction, entertainment, and a model for teamwork. The sports that most appealed to Bardeen were ones in which he could practice mastery and engage in constructive competition. Bardeen was intensely competitive in both work and sports, but his kind of competition seems to have been driven less by a desire to win over another person than a wish to triumph over nature’s hardest problems. Schrieffer said that Bardeen was “in competition basically with himself or with nature.” Studies of the personal dimension of genius suggest that creative people often continue to work actively into their senior years. Bardeen was no exception. His passion or ability for research hardly diminished with age, although his mental flexibility appears to have declined somewhat. In his seventies Bardeen took on a prob- lem as difficult as any he had attempted earlier in life. His bold and controversial theory of charge density waves might have earned him a third Nobel Prize had it hit pay dirt. Despite his inability to convince his colleagues of the merits of his work on charge density waves, he remained confident that his theory was correct. He never lost his excitement for the problem. A colleague said, “He’s the only man I ever knew who in his later years was still actively excited, working like crazy about physics, and not just working but writing papers that were right at the forefront and controversial.” We find a similar pattern in Einstein’s struggles late in life to develop a unified field theory and in Dirac’s last theoretical work on the magnetic monopole. Attempts to describe the second, i.e., methodological, dimen- sion of the genius profile have benefited from extensive experimen- tal research conducted by cognitive psychologists. For over two decades their dominant metaphor for human problem solving has been information processing by computer. The pioneering work on scientific thinking by Herbert Simon and his collaborators has described problem solving as a search through a “problem space.” Among the many intriguing results of experimental studies in this field are those designed to explore the differences between novice and expert problem solvers. In exploring the differences in 

320 TRUE GENIUS the recall that master and novice chess players have of the arrange- ment of pieces on a chess board, psychologists have found evidence to support the theory that master players have at their disposal more complex (“chunked”) mental structures than do novice players. These structures were created over the course of their long periods of strenuous training and experience. There is reason to believe that such chunking of knowledge also occurs in the sciences. The development of the expert-level structures necessary for creative work appears to require about ten years of intense study and practice to develop. Thus highly creative individuals often produce major achievements about every ten years—as did Bardeen in his work on the transistor and superconductivity, which culmi- nated in 1947 and 1957. The resulting master-level cognitive appa- ratus makes the search through problem space much faster. According to this notion, then, the meaning of a person’s “intu- ition” in a domain is strongly correlated with the amount and intensity of their experience in their domains. Bardeen’s notebooks and letters document many problem- solving approaches that have been studied experimentally. A fre- quent method of Bardeen’s is one that psychologists sometimes call “problem decomposition,” another name for the methodology that Bardeen said he learned from Wigner. Bardeen’s scientific notes include many lists of “subproblems,” smaller problems to be solved along the way to solving larger problems. The subproblems he laid out in his study of superconductivity in 1951 included deriving the London equations for a multiply connected body and proving the current and effective mass theories for one-electron wave functions. Colleagues have reported that Bardeen gave every problem, no mat- ter how small, his full attention. While Wigner may have urged Bardeen to use this analytic approach, Althea Bardeen probably pre- conditioned him to decompose problems, just as she had encour- aged her students at the Dewey School to break down problems. Bardeen’s own references to breaking down problems conflate two kinds of decomposition, both of which he routinely attempted. In one approach he would simply break a larger problem down into smaller parts and, when possible, delegate responsibility for solving the parts to other members of his team. Thus in the push to complete the BCS theory, Bardeen asked Schrieffer to work on the thermodynamic properties and Cooper to handle electrodynamics, 

Epilogue: True Genius and How to Cultivate It 321 while he himself worked on transport and nonequilibrium properties. In the other kind of problem decomposition, Bardeen would reduce the problem to be solved to the simplest model that con- tained the basic features of the larger problem. In his words, “You reduce a problem to its bare essentials, so that it contains just as much of the physics as necessary.” Thus in 1952, in studying the coupling of electrons in a superconductor with the lattice vibra- tions, Bardeen asked Pines to examine the problem of the polaron, a simpler problem containing some of the important features of superconductivity. After that, he could add in more features of the larger problem. Bardeen was not concerned about elegance in his approach to problem solving. He preferred, as Seitz said, to “bully through” to the answer. He was usually willing to try a range of approaches. For instance, in the weeks before inventing the transistor, Bardeen and Brattain studied a series of systems different from one another in material, geometry, or an aspect of structure. To keep control, they tried to make only one or very few changes in moving from one system to the next. It was a kind of experimental tinkering, useful in the absence of a complete theory. Bardeen would also at times employ the theoretical counter- part of such experimental tinkering. A kind of “brainstorming,” such theoretical tinkering resembles techniques that were widely publicized during the 1950s and 1960s as a means for inspiring creativity. It was a time when creativity was being recognized as a tool useful in fighting the Cold War. In those days, creativity was often described as the result of “divergent” (or “lateral”) thinking, an alternative to “convergent” (logical) thinking. In this picture, creativity was possible when the mind moved off its “beaten track.” Such studies of divergent think- ing were officially sanctioned by the academic world. In 1950 the psychologist Joy P. Guilford, then president of the American Psychological Association, called for a scientific study of divergent thinking. He encouraged psychologists to work on identifying cognitive operations involved in creative problem solving. The resulting research motivated many variations on what came to be called the “psychometric approach,” for example, the work of Raymond Cattell in the 1960s and 1970s on fluid intelligence and that of Gardner in the 1980s on multiple intelligences. 

322 TRUE GENIUS The same basic approach was publicized by the creativity gurus of the 1950s and 1960s, such as Edward de Bono, whose seminars were widely attended by corporate executives. Some of the brain- storming (or “blockbusting”) exercises followed lines pioneered by Alex Osborn in the 1950s, with tasks such as listing as quickly as possible all words one can associate with some central element of a problem. While the fervor of the divergent thinking movement of the fifties and sixties has long since died down, we can recognize in the work of Bardeen and other creative people a tendency to engage in a kind of brainstorming reminiscent of de Bono and Osborn at moments in their research when they feel “stuck.” Consider the variety of approaches Bardeen and Brattain used twelve days before their great discovery of the transistor. They had achieved a struc- ture that yielded some current amplification, but they did not understand why. Struggling to advance, they tried many varia- tions—of material, geometry, or technique—carefully observing what improved or diminished the effect. A kind of brainstorming also showed up when Schrieffer tinkered in January 1957 with a range of theoretical ideas at the time he wrote down the expression for the BCS ground-state wave function. Countless examples of this approach can be found in the history of technology, for instance, in the time-pressured work on implosion during the Manhattan Project, in Fermilab’s building of the first large superconducting magnet accelerator, and in the Wright brothers’ experiments on flight. Such work—and much of Bardeen’s problem solving—is heavily grounded in experimentation. A trained engineer as well as a physicist, Bardeen believed that “experiment came first, experiment came second, experiment came third,” in the words of Pines. The theoretical physicist Alexei Alexeivich Abrikosov explained that, unlike many theorists who wish “to create laws and to impose [them] on nature,” Bardeen would always begin with the results from experiments. “He felt,” wrote Schrieffer, “that formalism could lead one astray, unless it was closely tied to experi- ment and physical intuition. He also felt that mathematics had not yet matured to the level required to attack straightforwardly many of the frontier problems.” On one occasion a vice-president at Sharp Corporation asked Bardeen about the other theorists with whom he worked. Bardeen responded, “I don’t generally work with theo- reticians. I like to examine the data myself.” 

Epilogue: True Genius and How to Cultivate It 323 Within experimental research programs are those rare moments when anomalies signal new insight. The physicist Leon Lederman called such moments “epiphanies” because they result in “the sudden understanding of something new and important, something beautiful.” Of course, epiphanies can only happen when anomalies are recognized and acted upon. In Bardeen’s work on the transistor, unexpected experimental results brought some of the most dramatic progress. For example, the first observation of a transistor effect in December 1947 was the result of accidental electrical shorting owing to the fact that a germanium oxide they thought was present had actually washed off. Germanium’s material prop- erty of not sustaining such an oxide changed the structure of the experiment in an unexpected way allowing holes to enter the in- version layer on the germanium slab. Bardeen and Brattain were brought to that realization when they noticed the most obvious consequence of this “accident,” that the current was flowing in the opposite direction to what they expected. There is ample data in the historical papers of Bardeen to fuel study of the roles of analogy and metaphor in his scientific think- ing. Finding the best way to represent a problem was a continuing concern for Bardeen. For instance, after Cooper’s suggestion of elec- tron pairing, Bardeen, Cooper, and Schrieffer were stuck for over a year because they did not know how to represent mathematically the situation in which many electron pairs overlap physically with many other pairs. At one point, following in the footsteps of Fritz London, Bardeen suggested changing what he referred to as the “language” of the representation, from ordinary three-dimensional position space to a space in which the electrons are specified by their momentum coordinates. Almost a year after BCS was invented, Schrieffer developed a clever dance analogy in an effort to explain the representation problem to colleagues and students. Comparing the electron pairs in a superconductor to couples danc- ing the Frug on a crowded floor allowed him to portray the nature of the interactions between the pairs. Psychologists describe analogy as a mapping from a more fa- miliar domain (the “base”) to the less familiar one being studied (the “target”). By explicitly conveying relationships from the base into the target domain, researchers, including Dedre Gentner, have explained how scientists can set the stage for developing a new picture. Bardeen’s entry on December 24, 1947, in his Bell Labs notebook, reproduced in part in Figure 17-1, illustrates how he and 

324 TRUE GENIUS FIGURE 8 Analogy of transistor to triode system (Bardeen notebook, Dec. 24, 1947. Brattain mapped well-studied features (grid and plate) of the familiar system of the triode onto the unknown semiconductor system they had built, in an attempt to understand how the semi- conductor system could in fact amplify. A deeper analysis of how such analogies brought Bardeen and other scientists to their major advances could shed new light on our understanding of creative discovery in the sciences. In using metaphor, one is also (as in the case of analogy) com- paring things that are not yet understood with things that are better understood, but the link between the two is more direct: one describes the unknown in terms of the known, even when the details don’t match up perfectly. For example, when Bardeen sug- gested that Gerald Pearson conduct a low-temperature experiment in which the electrons would be “frozen” into the surface states, he meant that the electrons act as though the crystal were frozen 

Epilogue: True Genius and How to Cultivate It 325 like ice. Drawing explicitly on a series of examples from the his- tory of physics, Arthur I. Miller has shown how metaphors can themselves act as models in research and thus become powerful instruments of discovery and invention. For cases in which the base and the target domains overlap, Bardeen would sometimes employ metaphors as bridging principles in which part of the known domain can point the way to under- standing an unknown domain of physics. For instance, he specified, in a manner resembling the way Niels Bohr used his Correspon- dence Principle, that the unknown superconducting energy states must correspond one to one with free electron states in the pres- ence of a weak residual interaction inducing superconductivity. Whenever possible, Bardeen distributed his work among the members of his team in such a way that coordinating their particu- lar experiences and talents could help him make a more effective attack on the problem at hand. Some of his best work occurred in collaborations in which the other members offered expertise that differed from his own. Bardeen’s and Brattain’s notebooks from 1946 and 1947 document their rich interplay in the context of the Bell Labs semiconductor group, which had been, in turn, structured in a way likely to motivate interdisciplinary collaboration. In later years Bardeen consciously built such interdisciplinary collaboration into his problem-solving methodology, for instance, adding Cooper to the superconductivity team to gain access to his expertise in quantum field theory. Bardeen’s network of collabora- tion also extended outwards from his immediate team into the larger physics community. In one crucial example, a phone call on May 15, 1950, from Bernard Serin offered Bardeen the clue “that electron-lattice interactions are important in determining super- conductivity.” Yet another strategy in Bardeen’s methodological profile was to limit the focus of his research to only a few questions. Psycholo- gists conceive of an expert as one who “knows a lot about a small number of things.” By staying in the same area of physics for many years, Bardeen developed deep expertise and a profound base of knowledge with which he could probe research problems in the domain of solids. Wigner, whose own professional interests shifted several times, told an interviewer that Bardeen “had one wonderful quality—he didn’t change the subject of his interest too often and therefore he could make very important contributions.” 

326 TRUE GENIUS Bardeen’s regular and systematic updating of his knowledge base should not be overlooked as an important and powerful meth- odology. Almost daily he took the time to visit the library and read publications of interest to his work. When starting a new problem, he always began with a literature review. By widening and deepen- ing his base of knowledge, such library research often gave Bardeen the advantage over other workers. These overlapping approaches to problem solving, clearly illus- trated in the documentation of Bardeen’s work, help us model one physicist’s creative methodology. To generalize, or perhaps to develop a typology of alternative models, it is crucial for historians, psychologists, and other scholars of creativity to study many cases. In the area of the history of science, this program of research is well underway in the research of scholars such as Nancy Nersessian, Ryan Tweney, Frank Sulloway, Howard Gruber, Arthur I. Miller, Gerald Holton, Keith Simonton, Dedre Gentner, and Pat Langley. But the community of scholars conducting studies of this kind is still too small to accomplish the larger task. The third dimension of the genius profile deals with context, the many environments, such as family or workplace, that can nurture the other dimensions, for instance, by instilling values, offering connections, and providing moral or financial support. Bardeen’s parents and grandfather all stressed learning, creativity, diligence, and the joy that comes from work in the interest of social betterment. John’s parents strongly supported the development of qualities useful to his education and emerging creativity. For example, they applauded his early tendency to “hang on” to problems stub- bornly. Bardeen’s colleagues often spoke of his “bulldog tenacity.” Bardeen’s mother also set the model of an independent person with the courage to follow her passion. The fact that she could at age twenty-five break from her family to study art cleared the way for John to make a similar move when, at age twenty-five, during the Depression, he quit his secure engineering post so that he could enter the graduate program at Princeton. An equally important model was set by his father, Charles, the academic scientist dedi- cated to solving problems for the public good. Bardeen’s grounded- ness in the problems of real materials and his commitment to industry and government had their basis in the progressive culture and ideals of his youth in Madison. “The greatest opportunity of all is to serve,” his grandfather had written to him. 

Epilogue: True Genius and How to Cultivate It 327 The fact that Bardeen’s parents intervened to support his aca- demic development impacted both his intellectual and personal development. He reflected in a later interview that, although he had been marginalized in his classes with children four or five years his senior, the fact that he had been skipped three grades had gener- ally benefited him. Not only was he challenged, especially in math, but, as he said, he had more time to concentrate on his studies. He considered it valuable to have avoided the boredom that gifted chil- dren often experience in school. In his later years, Bardeen often landed in settings that repro- duced his productive early learning environments. Similar in philosophy to the Dewey School, where Althea had taught before marrying Charles, Wisconsin’s University High was a progressive school aimed at supporting the creativity of its students. Charles credited Uni High with nurturing John’s mathematical talent. As Charles explained to his father, the Wisconsin Uni students were “allowed to go ahead in any line they take up as fast as they show ability and the hard and fast grade system does not exist.” The “graduate student’s paradise” that Bardeen had encountered at Princeton, the stimulating contexts of the Harvard Society of Fellows, the semiconductor subgroup at Bell Labs in 1945–1947, and the Department of Physics at the University of Illinois all resembled Uni High in supporting Bardeen’s creative work in physics. The Bardeen home was an important factor, too. An anchor offering stability and attachment to reality, home and family were critical, especially in times of stress or when Bardeen ventured off into unknown scientific territory. After John’s marriage in 1938, Jane Maxwell assumed the responsibility for maintaining this stability, even while she herself felt separated from her husband’s most passionate interests. Bardeen’s mentors played crucial roles in his creativity. Walter W. Hart, Bardeen’s seventh- and eighth-grade math teacher, fanned the flames of John’s interest in mathematics. When Bardeen was an undergraduate at Wisconsin, Paul Dirac, Peter Debye, and espe- cially John Van Vleck introduced him to modern physics in the period when quantum mechanics was born. Bardeen’s mentors opened doors on his behalf. Van Vleck wrote many letters of support, e.g., to Princeton, Harvard, Minnesota, and the University of Illinois. Consistent with Harriet Zuckerman’s 

328 TRUE GENIUS thesis about the attraction between “Nobel-bound” mentors and students is the fact that an unusually large number of Bardeen’s physics teachers later received Nobel prizes—Dirac, Debye, Van Vleck, Bridgman, and Wigner. Bardeen learned from them what it takes to do world-class research in physics. Although none of these teachers had yet won their Nobel prizes at the time Bardeen stud- ied with them, Bardeen and these mentors recognized each other, somehow. Bardeen studied with Wigner during the two or three years when Wigner was passionately interested in solids. In 1945 he worked at Bell Labs, by all accounts then one of the best places in the world to study semiconductors. In 1951, when Bardeen came to the University of Illinois, Seitz was building there one of the first large American solid-state departments in academia. How did it happen that Bardeen and other creative individuals so often worked in environments and with mentors who supported their creative work? Bardeen himself claimed that “accomplishments are a good bit of luck—being in the right place at the right time—and having the right associates.” He considered himself “lucky” to have been “on the ground floor of solid-state physics.” But serendipity cannot explain why Bardeen so often found himself in the right place at the right time to support his creativity. For any major invention to occur, a large number of contingencies need to mesh. How could such a highly improbable act occur twice for Bardeen? Why do such “epiphanies” typically occur more than once in the careers of other creative people, often with a frequency of about ten years? It appears that creative people actively seek or create the envi- ronments that support their intensely dedicated work in an area. They create their “historical junctures” by placing themselves in the “right place at the right time” and among the “right associ- ates,” whom they are able to recognize because of their own experi- ences, interests, and passions. Recall the care with which Bardeen selected his thesis advisor and made his various career moves. It seems clear that exceptional creativity is the product of many factors, some individual, some methodological, others contextual. Establishing these more rigorously and determining how they interact forms the central problem for the next stage of understand- ing genius and creativity. Much progress has been made, but the research is still at an early stage. 

Epilogue: True Genius and How to Cultivate It 329 Different approaches are needed, for each method has its advan- tages and its disadvantages. In experiments the parameters can be altered and controls are possible, but experiments can rarely be per- formed on the most creative individuals, nor can they usually run over the decade-long periods required for major creative work. Surveys and interviews allow people to be questioned about their work, but they are unreliable, biased, and, in the case of surveys, inflexible. Individuals do not know themselves all that was in- volved in their creative work. Computer simulations offer unlim- ited possibilities for analyzing a great variety of parameters, but so far at least computers are not able to handle nonrational variables. Videotapes of research in the making offer the researcher a ringside seat at the moment of creativity, but like experiments, their use is limited to short time periods. Much of what we know must come from biographies that ex- amine the lives and works of creative people. These too have dis- advantages, for biographies, and histories in general, are only as good as the sources they are based on. Creative people never record every step of their work and human memory distorts the path of discovery. Frans de Waal aptly wrote of scientific research, “In hind- sight, the path taken may look straight, running from ignorance to profound insight, but only because our memory for dead ends is so much worse than that of a rat in a maze.” Most of the legendary stories about individual acts of creativity—such as Fredrich August von Kekule’s famous dream of the snake biting its tail, which is said to have led to his discovery of the benzene ring, or Samuel Taylor Coleridge’s account of conceiving his poem Kubla Khan in an opium dream—have been disproved. Even the documents that people leave behind can be mislead- ing, because individuals have a tendency to shape the evidence so as to make their discovery paths appear straighter than they actu- ally were. Well-meaning family members, eager to shrink a massive collection of sources, often toss out the documentation of false starts. Even archivists occasionally destroy historical data, if only by rearranging papers for more convenient retrieval. The road ahead for scholars of genius and creativity must there- fore remain arduous. But the problems they face are not insur- mountable because their subjects are not the otherworldly savants that geniuses were once thought to be. They are real people, like John Bardeen, highly motivated to develop the elements of genius that exist potentially in all of us.