WHAT MAKES US HAPPY?

Is there a formula—some mix of love, work, and psychological adaptation—for a good life?

By Joshua Wolf Shenk

For 72 years, researchers at Harvard have been examining this question, following 268 men who entered college in the late 1930s through war, career, marriage and divorce, parenthood and grandparenthood, and old age. Here, for the first time, a journalist gains access to the archive of one of the most comprehensive longitudinal studies in history. Its contents, as much literature as science, offer profound insight into the human condition—and into the brilliant, complex mind of the study’s longtime director, George Vaillant.

Last fall, I spent about a month in the file room of the Harvard Study of Adult Development, hoping to learn the secrets of the good life. The project is one of the longest-running—and probably the most exhaustive—longitudinal studies of mental and physical well-being in history. Begun in 1937 as a study of healthy, well-adjusted Harvard sophomores (all male), it has followed its subjects for more than 70 years.

Interviews:

Ben Bradlee: "I Haven't Been Unhappy in My Life"
The famed editor reflects on his education, career, and experiences as a member of the Grant Study.

Donald Cole: "I Have Always Thought Adaptation Was a Wonderful Thing."
A historian and prep school teacher reflects on his life and how it has been affected by his participation in the study.

From their days of bull sessions in Cambridge to their active duty in World War II, through marriages and divorces, professional advancement and collapse—and now well into retirement—the men have submitted to regular medical exams, taken psychological tests, returned questionnaires, and sat for interviews. The files holding the data are as thick as unabridged dictionaries. They sit in a wall of locked cabinets in an office suite behind Fenway Park in Boston, in a plain room with beige carpeting and fluorescent lights that is littered with the detritus of many decades of social-scientific inquiry: a pile of enormous spreadsheet data books; a 1970s-era typewriter; a Macintosh PowerBook, circa 1993. All that’s missing are the IBM punch cards used to analyze the data in the early days.

For 42 years, the psychiatrist George Vaillant has been the chief curator of these lives, the chief investigator of their experiences, and the chief analyst of their lessons. His own life has been so woven into the study—and the study has become such a creature of his mind—that neither can be understood without the other. As Vaillant nears retirement (he’s now 74), and the study survivors approach death—the roughly half still living are in their late 80s—it’s a good time to examine both, and to do so, I was granted unprecedented access to case files ordinarily restricted to researchers.

As a young man, Vaillant fell in love with the longitudinal method of research, which tracks relatively small samples over long periods of time (as in Michael Apted’s Seven Up! documentaries). In 1961, as a psychiatric resident at the Massachusetts Mental Health Center, Vaillant found himself intrigued by two patients with manic depression who had 25 years earlier been diagnosed as incurable schizophrenics. Vaillant asked around for other cases of remitted schizophrenia and pulled their charts. “These records hadn’t been assembled to do research,” Vaillant told me recently, “but it was contemporary, real-time information, with none of the errors you get from memory or the distortions you get when you narrate history from the vantage of the present.” In 1967, after similar work following up on heroin addicts, he discovered the Harvard Study, and his jaw dropped. “To be able to study lives in such depth, over so many decades,” he said, “it was like looking through the Mount Palomar telescope,” then the most powerful in the world. Soon after he began to work with the material, he found himself talking about the project to his psychoanalyst. Showing him the key that opened the study cabinets, Vaillant said, “I have the key to Fort Knox.”

Such bravado had defined the study from the start. Arlie Bock—a brusque, no-nonsense physician who grew up in Iowa and took over the health services at Harvard University in the 1930s—conceived the project with his patron, the department-store magnate W. T. Grant. Writing in September 1938, Bock declared that medical research paid too much attention to sick people; that dividing the body up into symptoms and diseases—and viewing it through the lenses of a hundred micro-specialties—could never shed light on the urgent question of how, on the whole, to live well. His study would draw on undergraduates who could “paddle their own canoe,” Bock said, and it would “attempt to analyze the forces that have produced normal young men.” He defined normal as “that combination of sentiments and physiological factors which in toto is commonly interpreted as successful living.”

Bock assembled a team that spanned medicine, physiology, anthropology, psychiatry, psychology, and social work, and was advised by such luminaries as the psychiatrist Adolf Meyer and the psychologist Henry Murray. Combing through health data, academic records, and recommendations from the Harvard dean, they chose 268 students—mostly from the classes of 1942, ’43, and ’44—and measured them from every conceivable angle and with every available scientific tool.

Exhaustive medical exams noted everything from major organ function, to the measure of lactic acid after five minutes on a treadmill, to the size of the “lip seam” and the hanging length of the scrotum. Using a new test called the electroencephalograph, the study measured the electrical activity in the brain, and sought to deduce character from the squiggles. During a home visit, a social worker took not only a boy’s history—when he stopped wetting his bed, how he learned about sex—but also extensive medical and social histories on his parents and extended family. The boys interpreted Rorschach inkblots, submitted handwriting samples for analysis, and talked extensively with psychiatrists. They stripped naked so that every dimension of their bodies could be measured for “anthropometric” analysis, a kind of whole-body phrenology based on the premise that stock character types could be seen from body proportions.

Inveighing against medicine’s tendency to think small and specialized, Bock made big promises. He told the Harvard Crimson in 1942 that his study of successful men was pitched at easing “the disharmony of the world at large.” One early Grant Study document compared its prospects to the accomplishments of Socrates, Galileo, and Pasteur. But in fact the study staff remained bound by their respective disciplines and by the kinds of narrow topics that yield academic journal papers. Titles from the study’s early years included “Resting-Pulse and Blood-Pressure Values in Relation to Physical Fitness in Young Men”; “Instruction Time in Certain Multiple Choice Tests”; and “Notes on Usage of Male Personal Names.” Perhaps the height of the study’s usefulness in its early days was to lend its methods to the military, for officer selection in World War II.

Most longitudinal studies die on the vine because funders expect results quickly. W. T. Grant was no exception. He held on for about a decade—allowing the staff to keep sending detailed annual questionnaires to the men, hold regular case conferences, and publish a flurry of papers and several books—before he stopped sending checks. By the late 1940s, the Rockefeller Foundation took an interest, funding a research anthropologist named Margaret Lantis, who visited every man she could track down (which was all but a few). But by the mid-1950s, the study was on life support. The staff, including Clark Heath, who had managed the study for Bock, scattered, and the project fell into the care of a lone Harvard Health Services psychologist, Charles McArthur. He kept it limping along—surveys dwindled to once every two years—in part by asking questions about smoking habits and cigarette-brand preferences, a nod to a new study patron, Philip Morris. One survey asked, “If you never smoked, why didn’t you?”

It was a far cry from Galileo.

But as Vaillant points out, longitudinal studies, like wines, improve with age. And as the Grant Study men entered middle age—they spent their 40s in the 1960s—many achieved dramatic success. Four members of the sample ran for the U.S. Senate. One served in a presidential Cabinet, and one was president. There was a best-selling novelist (not, Vaillant has revealed, Norman Mailer, Harvard class of ’43). But hidden amid the shimmering successes were darker hues. As early as 1948, 20 members of the group displayed severe psychiatric difficulties. By age 50, almost a third of the men had at one time or another met Vaillant’s criteria for mental illness. Underneath the tweed jackets of these Harvard elites beat troubled hearts. Arlie Bock didn’t get it. “They were normal when I picked them,” he told Vaillant in the 1960s. “It must have been the psychiatrists who screwed them up.”

In all Vaillant’s literature—and, by agreement, in this essay, too—the Grant Study men remain anonymous. (Even the numbers on the case studies have been changed.) A handful have publicly identified themselves—including Ben Bradlee, the longtime editor of The Washington Post, who opened his memoir, A Good Life, with his first trip to the study office. John F. Kennedy was a Grant Study man, too, though his files were long ago withdrawn from the study office and sealed until 2040. Ironically, it was the notation of that seal in the archive that allowed me to confirm JFK’s involvement, which has not been recognized publicly before now.

Of course, Kennedy—the heir to ruthless, ambitious privilege; the philanderer of “Camelot”; the paragon of casual wit and physical vigor who, backstage, suffered from debilitating illness—is no one’s idea of “normal.” And that’s the point. The study began in the spirit of laying lives out on a microscope slide. But it turned out that the lives were too big, too weird, too full of subtleties and contradictions to fit any easy conception of “successful living.” Arlie Bock had gone looking for binary conclusions—yeses and nos, dos and don’ts. But the enduring lessons would be paradoxical, not only on the substance of the men’s lives (the most inspiring triumphs were often studies in hardship) but also with respect to method: if it was to come to life, this cleaver-sharp science project would need the rounding influence of storytelling.

In George Vaillant, the Grant Study found its storyteller, and in the Grant Study, Vaillant found a set of data, and a series of texts, suited to his peculiar gifts. A tall man, with a gravelly voice, steel-gray hair, and eyes that can radiate great joy and deep sadness, Vaillant blends the regal bearing of his old-money ancestors, the emotional directness of his psychiatric colleagues, and a genial absentmindedness. (A colleague recalls one day in the 1980s when Vaillant came to the office in his slippers.)

As with many of the men he came to study, Vaillant’s gifts and talents were shaped by his needs and pains. Born in 1934, Vaillant grew up in what he described to me as “blessed circumstances”—living “during the Great Depression with a nurse, a maid, and a cook, but without anybody having so much money that you stared in dismay at the newspapers” as stock prices sagged. And his parents had a storybook romance. They met in Mexico City, where she was the daughter of a prominent expatriate American banker and he was a hotshot archaeologist working on pre-Columbian Aztec digs. When George was 2, he says, his father “gave up being Indiana Jones and became a suit,” first as a curator at the American Museum of Natural History in New York City and then as the director of the University Museum at the University of Pennsylvania in Philadelphia. He was an accomplished man who, his son says, showed little trace of doubt or depression. But one Sunday afternoon in 1945, at home in Devon, Pennsylvania, George Clapp Vaillant, then 44, went out into the yard after a nap. His wife found him by the pool, a revolver next to him and a fatal wound through the mouth. His elder son and namesake, the last to see his father alive, was 10 years old.

Immediately, a curtain of silence fell around the tragedy. “In WASP fashion,” Vaillant says, “it was handled with ‘Let’s get this put away as quickly as possible.’” His mother, Suzannah Beck Vaillant, picked up the children and took them to Arizona. “We never saw our house again,” says Henry Vaillant, George’s younger brother. “We never attended the memorial service. It was just kind of a complete cutoff.”

A few years later, their father’s 25th-reunion book, hardbound in red cloth, arrived in the mail from Harvard College. George spent days with it, spellbound by the photographs and words that showed college students morphing, over the course of a few paragraphs, into 47-year-olds. The seed of interest in longitudinal research had been planted; it germinated decades later in Vaillant’s psychiatric residency and then in the ultimate vein of data he discovered at Harvard. It was 1967, and the Grant Study men were beginning to return for their 25th college reunions. Vaillant was 33. He would spend the rest of his career—and expects to spend the rest of his life—following these men.

The range of his training and the complexity of his own character proved to be crucial to his research. After Harvard College (where he wrote for the Lampoon, the humor magazine, and studied history and literature), Harvard Medical School, and a residency at the Massachusetts Mental Health Center, Vaillant studied at the Boston Psychoanalytic Institute, which he calls a “temple” to Freud’s ideas. He learned the orthodoxy, which included a literary approach to human lives, bringing theory to bear through deep reading of individual cases. But he also had training in the rigors of data-driven experimental science, including a two-year fellowship at a Skinnerian laboratory, where he studied neurotransmitter levels in pigeons and monkeys. There he learned to use the behaviorist B. F. Skinner’s “cumulative behavioral recorder,” which collapses behaviors across minutes, hours, or days onto a chart to be inspected in a single sitting.

The undertones of psychoanalysis are tragic; Freud dismissed the very idea of “normality” as “an ideal fiction” and famously remarked that he hoped to transform “hysterical misery into common unhappiness.” The spirit of modern social science, by contrast, draws on a brash optimism that the secrets to life can be laid bare. Vaillant is an optimist marinated in tragedy, not just in his life experience, but in his taste. Above his desk hangs a letter from a group of his medical residents to their successors, advising them to prepare for Vaillant’s “obscure literary references” by reading Tennessee Williams’s The Glass Menagerie, Arthur Miller’s Death of a Salesman, and Henrik Ibsen’s A Doll’s House. Vaillant loves Dostoyevsky and Tolstoy, too, and the cartoons of the dark humorist Charles Addams, like the one where several Christmas carolers sing merrily at the Addams family doorstep, while Morticia, Lurch, and Gomez stand on the roof, ready to tip a vat of hot oil on their heads. When his children were small, Vaillant would read them a poem about a tribe of happy-go-lucky bears, who lived in a kind of Eden until a tribe of mangier, smarter bears came along and enslaved them. “I would weep at this story,” remembers his daughter Anne Vaillant. “Dad thought it was funny, and I think somehow it was helpful to him that I had such feelings about it. There was this sort of, ‘This is the way life is.’”

Yet, even as he takes pleasure in poking holes in an innocent idealism, Vaillant says his hopeful temperament is best summed up by the story of a father who on Christmas Eve puts into one son’s stocking a fine gold watch, and into another son’s, a pile of horse manure. The next morning, the first boy comes to his father and says glumly, “Dad, I just don’t know what I’ll do with this watch. It’s so fragile. It could break.” The other boy runs to him and says, “Daddy! Daddy! Santa left me a pony, if only I can just find it!”

The story gets to the heart of Vaillant’s angle on the Grant Study. His central question is not how much or how little trouble these men met, but rather precisely how—and to what effect—they responded to that trouble. His main interpretive lens has been the psychoanalytic metaphor of “adaptations,” or unconscious responses to pain, conflict, or uncertainty. Formalized by Anna Freud on the basis of her father’s work, adaptations (also called “defense mechanisms”) are unconscious thoughts and behaviors that you could say either shape or distort—depending on whether you approve or disapprove—a person’s reality.

Vaillant explains defenses as the mental equivalent of a basic biological process. When we cut ourselves, for example, our blood clots—a swift and involuntary response that maintains homeostasis. Similarly, when we encounter a challenge large or small—a mother’s death or a broken shoelace—our defenses float us through the emotional swamp. And just as clotting can save us from bleeding to death—or plug a coronary artery and lead to a heart attack—defenses can spell our redemption or ruin. Vaillant’s taxonomy ranks defenses from worst to best, in four categories.

At the bottom of the pile are the unhealthiest, or “psychotic,” adaptations—like paranoia, hallucination, or megalomania—which, while they can serve to make reality tolerable for the person employing them, seem crazy to anyone else. One level up are the “immature” adaptations, which include acting out, passive aggression, hypochondria, projection, and fantasy. These aren’t as isolating as psychotic adaptations, but they impede intimacy. “Neurotic” defenses are common in “normal” people. These include intellectualization (mutating the primal stuff of life into objects of formal thought); dissociation (intense, often brief, removal from one’s feelings); and repression, which, Vaillant says, can involve “seemingly inexplicable naïveté, memory lapse, or failure to acknowledge input from a selected sense organ.” The healthiest, or “mature,” adaptations include altruism, humor, anticipation (looking ahead and planning for future discomfort), suppression (a conscious decision to postpone attention to an impulse or conflict, to be addressed in good time), and sublimation (finding outlets for feelings, like putting aggression into sport, or lust into courtship).

In contrast to Anna Freud, who located the origins of defenses in the sexual conflicts of a child, Vaillant sees adaptations as arising organically from the pain of experience and playing out through the whole lifespan. Take his comparison of two Grant Study men, whom he named “David Goodhart” and “Carlton Tarrytown” in his first book on the study, Adaptation to Life, published in 1977. Both men grew up fearful and lonely. Goodhart was raised in a blue-collar family, had a bigoted, alcoholic father, and a mother he described as “very nervous, irritable, anxious, and a worrier.” Tarrytown was richer, and was raised in a wealthy suburb, but he also had an alcoholic father, and his mother was so depressed that he feared she would commit suicide. Goodhart went on to become a national leader on civil-rights issues—a master, Vaillant argued, of the “mature” defenses of sublimation and altruism. By his late 40s, staff researchers using independent ratings put Goodhart in the top fifth of the Grant Study in psychological adjustment. Tarrytown, meanwhile, was in the bottom fifth. A doctor who left a regular practice to work for the state, a three-time divorcé who anesthetized his pain with alcohol and sedatives, Tarrytown was, Vaillant said, a user of dissociation and projection—“neurotic” and “immature” defenses, respectively. After a relapse into drug abuse, Tarrytown killed himself at 53. Goodhart lived to 70. Though Vaillant says that the “dashing major” of midlife became a stolid and portly brigadier general, Goodhart’s obituaries still celebrated a hero of civil rights.

Most psychology preoccupies itself with mapping the heavens of health in sharp contrast to the underworld of illness. “Social anxiety disorder” is distinguished from shyness. Depression is defined as errors in cognition. Vaillant’s work, in contrast, creates a refreshing conversation about health and illness as weather patterns in a common space. “Much of what is labeled mental illness,” Vaillant writes, “simply reflects our ‘unwise’ deployment of defense mechanisms. If we use defenses well, we are deemed mentally healthy, conscientious, funny, creative, and altruistic. If we use them badly, the psychiatrist diagnoses us ill, our neighbors label us unpleasant, and society brands us immoral.”

This perspective is shaped by a long-term view. Whereas clinicians focus on treating a problem at any given time, Vaillant is more like a biographer, looking to make sense of a whole life—or, to take an even broader view, like an anthropologist or naturalist looking to capture an era. The good news, he argues, is that diseases—and people, too—have a “natural history.” After all, many of the “psychotic” adaptations are common in toddlers, and the “immature” adaptations are essential in later childhood, and they often fade with maturity. As adolescents, the Grant Study men were twice as likely to use immature defenses as mature ones, but in middle life they were four times as likely to use mature defenses—and the progress continued into old age. When they were between 50 and 75, Vaillant found, altruism and humor grew more prevalent, while all the immature defenses grew more rare.

This means that a glimpse of any one moment in a life can be deeply misleading. A man at 20 who appears the model of altruism may turn out to be a kind of emotional prodigy—or he may be ducking the kind of engagement with reality that his peers are both moving toward and defending against. And, on the other extreme, a man at 20 who appears impossibly wounded may turn out to be gestating toward maturity.

Such was the case, Vaillant argues, with “Dr. Godfrey Minot Camille,” a poetic and troubled young man who spent so much time at the Harvard infirmary complaining of vague symptoms that a college physician declared, “This boy is becoming a regular psychoneurotic.” He’d grown up in a frigid environment—he ate his meals alone until age 6—and spoke of his desolation with heartbreaking clarity. A member of the study staff advised him: “When you come to the end of your rope, tie a knot and hold on.” He replied: “But the knot was tied so long ago, and I have been hanging on tight for such a long time.” After graduating from medical school, he attempted suicide.

With the help of psychotherapy and with the passage of time, his hypochondria eased and he began to show “displacement,” the strategy of shifting preoccupations from a painful source to more neutral ground. When his sister died, he sent her autopsy report to the Grant Study office, with a cool note saying that he expected it would be “an item of news.” He reported another family death this way: “I received an inheritance from my mother.”

For Camille, such detached neutrality seemed to herald progress. At 35, he spent 14 months in a hospital for an infection and had what he described as a spiritual awakening. “Someone with a capital ‘S’ cared about me,” he wrote. Afterward, he bloomed as a psychiatrist, channeling his own needs into service. He said he liked the “distant closeness” of psychotherapy—and liked getting paid for it. As a child, he had fantasized about being a minister or physician. “Finally, at age forty, wish became behavior,” Vaillant wrote.

In his 2002 book, Aging Well, Vaillant returned to this man’s story, this time calling him “Ted Merton” to emphasize his spiritual development. (The men in Vaillant’s books always have florid pseudonyms—Horace Lamb, Frederick Lion, Bill Loman, etc.) In several vignettes in the book, Vaillant presents Merton as an exemplar of how mature adaptations are a real-life alchemy, a way of turning the dross of emotional crises, pain, and deprivation into the gold of human connection, accomplishment, and creativity. “Such mechanisms are analogous to the involuntary grace by which an oyster, coping with an irritating grain of sand, creates a pearl,” he writes. “Humans, too, when confronted with irritants, engage in unconscious but often creative behavior.”

But “creative” doesn’t equate to ease. At ages 55 and 60, Merton had severe depressions. In the first instance he was hospitalized. The second instance coincided with his second divorce, and “he lost not only his wife, his savings, and his job, but even his network of professional colleagues.” Going forth into the breach of life can deepen meaning, but also deepen wounds.

What allows people to work, and love, as they grow old? By the time the Grant Study men had entered retirement, Vaillant, who had then been following them for a quarter century, had identified seven major factors that predict healthy aging, both physically and psychologically.

Employing mature adaptations was one. The others were education, stable marriage, not smoking, not abusing alcohol, some exercise, and healthy weight. Of the 106 Harvard men who had five or six of these factors in their favor at age 50, half ended up at 80 as what Vaillant called “happy-well” and only 7.5 percent as “sad-sick.” Meanwhile, of the men who had three or fewer of the health factors at age 50, none ended up “happy-well” at 80. Even if they had been in adequate physical shape at 50, the men who had three or fewer protective factors were three times as likely to be dead at 80 as those with four or more factors.

What factors don’t matter? Vaillant identified some surprises. Cholesterol levels at age 50 have nothing to do with health in old age. While social ease correlates highly with good psychosocial adjustment in college and early adulthood, its significance diminishes over time. The predictive importance of childhood temperament also diminishes over time: shy, anxious kids tend to do poorly in young adulthood, but by age 70, are just as likely as the outgoing kids to be “happy-well.” Vaillant sums up: “If you follow lives long enough, the risk factors for healthy life adjustment change. There is an age to watch your cholesterol and an age to ignore it.”

The study has yielded some additional subtle surprises. Regular exercise in college predicted late-life mental health better than it did physical health. And depression turned out to be a major drain on physical health: of the men who were diagnosed with depression by age 50, more than 70 percent had died or were chronically ill by 63. More broadly, pessimists seemed to suffer physically in comparison with optimists, perhaps because they’re less likely to connect with others or care for themselves.

More than 80 percent of the Grant Study men served in World War II, a fact that allowed Vaillant to study the effect of combat. The men who survived heavy fighting developed more chronic physical illnesses and died sooner than those who saw little or no combat, he found. And “severity of trauma is the best predictor of who is likely to develop PTSD.” (This may sound obvious, but it countered the claim that post-traumatic stress disorder was just the manifestation of preexisting troubles.) He also found that personality traits assigned by the psychiatrists in the initial interviews largely predicted who would become Democrats (descriptions included “sensitive,” “cultural,” and “introspective”) and Republicans (“pragmatic” and “organized”).

Again and again, Vaillant has returned to his major preoccupations. One is alcoholism, which he found is probably the horse, and not the cart, of pathology. “People often say, ‘That poor man. His wife left him and he’s taken to drink,’” Vaillant says. “But when you look closely, you see that he’s begun to drink, and that has helped drive his wife away.” The horrors of drink so preoccupied Vaillant that he devoted a stand-alone study to it: The Natural History of Alcoholism.

Vaillant’s other main interest is the power of relationships. “It is social aptitude,” he writes, “not intellectual brilliance or parental social class, that leads to successful aging.” Warm connections are necessary—and if not found in a mother or father, they can come from siblings, uncles, friends, mentors. The men’s relationships at age 47, he found, predicted late-life adjustment better than any other variable, except defenses. Good sibling relationships seem especially powerful: 93 percent of the men who were thriving at age 65 had been close to a brother or sister when younger. In an interview in the March 2008 newsletter to the Grant Study subjects, Vaillant was asked, “What have you learned from the Grant Study men?” Vaillant’s response: “That the only thing that really matters in life are your relationships to other people.”

The authority of these findings stems in large part from the rarity of the source. Few longitudinal studies survive in good health for whole lifetimes, because funding runs dry and the participants drift away. Vaillant managed, drawing on federal grants and private gifts, to finance surveys every two years, physicals every five years, and interviews every 15 years. The original study social worker, Lewise Gregory Davies, helped him goad the subjects to stay in touch, but it wasn’t a hard sell. The Grant Study men saw themselves as part of an elite club.

Vaillant also dramatically expanded his scope by taking over a defunct study of juvenile delinquents in inner-city Boston, run by the criminologists Sheldon and Eleanor Glueck. Launched in 1939, the study had a control group of nondelinquent boys who grew up in similar circumstances—children of poor, mostly foreign-born parents, about half of whom lived in a home without a tub or a shower. In the 1970s, Vaillant and his staff tracked down most of these nondelinquent boys—it took years—so that today the Harvard Study of Adult Development consists of two cohorts, the “Grant men” and the “Glueck men.” Vaillant also arranged to interview a group of women from the legendary Stanford Terman study, which in the 1920s began to follow a group of high-IQ kids in California.

In contrast to the Grant data, the Glueck study data suggested that industriousness in childhood—as indicated by such things as whether the boys had part-time jobs, took on chores, or joined school clubs or sports teams—predicted adult mental health better than any other factor, including family cohesion and warm maternal relationships. “What we do,” Vaillant concluded, “affects how we feel just as much as how we feel affects what we do.”

Interestingly, while the Glueck men were 50 percent more likely to become dependent on alcohol than the Harvard men, the ones who did were more than twice as likely to eventually get sober. “The difference has nothing to do with treatment, intelligence, self-care, or having something to lose,” Vaillant told Harvard magazine. “It does have to do with hitting bottom. Someone sleeping under the elevated-train tracks can at some point recognize that he’s an alcoholic, but the guy getting stewed every night at a private club may not.”

But Vaillant has largely played down the distinctions among the samples. For example, while he allows that, in mortality rates, the inner-city men at age 68 to 70 resembled the Terman and Harvard cohorts at 78 to 80, he says that most of the difference can be explained by less education, more obesity, and greater abuse of alcohol and cigarettes. “When these four variables were controlled,” he writes, “their much lower parental social class, IQ, and current income were not important.” But of course those are awfully significant variables to “control.” Vaillant points out that at age 70, the inner-city men who graduated from college were just as healthy as the Harvard men. But only 29 Glueck men did finish college—about 6 percent of the sample.

Having survived so many eras, the Grant Study is a palimpsest of the modern history of medicine and psychology, each respective era’s methods and preoccupations inscribed atop the preceding ones. In the 1930s, Arlie Bock’s work was influenced by the movement called “constitutional medicine,” which started as a holistic reaction to the minimalism engendered by Pasteur and germ theory. Charles McArthur, who picked up the study in the mid-1950s, was principally interested in matching people to suitable careers through psychological testing—perfect for the Man in the Gray Flannel Suit era. Vaillant’s use of statistical technique to justify psychoanalytic claims reflected the mode of late-1960s academic psychiatry, and his work caught on in the 1970s as part of a trend emphasizing adult development. Gail Sheehy’s 1976 best seller, Passages, drew on the Grant Study, as well as on the research of Daniel Levinson, who went on to publish The Seasons of a Man’s Life. (Sheehy was sued for alleged plagiarism by another academic, Roger Gould, who later published his own take on adult development in Transformations; Gould’s case was settled out of court.)

As Freud was displaced by biological psychiatry and cognitive psychology—and the massive data sets and double-blind trials that became the industry standard—Vaillant’s work risked obsolescence. But in the late 1990s, a tide called “positive psychology” came in, and lifted his boat. Driven by a savvy, brilliant psychologist at the University of Pennsylvania named Martin Seligman, the movement to create a scientific study of the good life has spread wildly through academia and popular culture (dozens of books, a cover story in Time, attention from Oprah, etc.).

Vaillant became a kind of godfather to the field, and a champion of its message that psychology can improve ordinary lives, not just treat disease. But in many ways, his role in the movement is as provocateur. Last October, I watched him give a lecture to Seligman’s graduate students on the power of positive emotions—awe, love, compassion, gratitude, forgiveness, joy, hope, and trust (or faith). “The happiness books say, ‘Try happiness. You’ll like it a lot more than misery’—which is perfectly true,” he told them. But why, he asked, do people tell psychologists they’d cross the street to avoid someone who had given them a compliment the previous day?

In fact, Vaillant went on, positive emotions make us more vulnerable than negative ones. One reason is that they’re future-oriented. Fear and sadness have immediate payoffs—protecting us from attack or attracting resources at times of distress. Gratitude and joy, over time, will yield better health and deeper connections—but in the short term actually put us at risk. That’s because, while negative emotions tend to be insulating, positive emotions expose us to the common elements of rejection and heartbreak.

To illustrate his point, he told a story about one of his “prize” Grant Study men, a doctor and well-loved husband. “On his 70th birthday,” Vaillant said, “when he retired from the faculty of medicine, his wife got hold of his patient list and secretly wrote to many of his longest-running patients, ‘Would you write a letter of appreciation?’ And back came 100 single-spaced, desperately loving letters—often with pictures attached. And she put them in a lovely presentation box covered with Thai silk, and gave it to him.” Eight years later, Vaillant interviewed the man, who proudly pulled the box down from his shelf. “George, I don’t know what you’re going to make of this,” the man said, as he began to cry, “but I’ve never read it.” “It’s very hard,” Vaillant said, “for most of us to tolerate being loved.”

Vaillant brings a healthy dose of subtlety to a field that sometimes seems to glide past it. The bookstore shelves are lined with titles that have an almost messianic tone, as in Happier: Learn the Secrets to Daily Joy and Lasting Fulfillment. But what does it mean, really, to be happier? For 30 years, Denmark has topped international happiness surveys. But Danes are hardly a sanguine bunch. Ask an American how it’s going, and you will usually hear “Really good.” Ask a Dane, and you will hear “Det kunne være værre (It could be worse).” “Danes have consistently low (and indubitably realistic) expectations for the year to come,” a team of Danish scholars concluded. “Year after year they are pleasantly surprised to find that not everything is getting more rotten in the state of Denmark.”

Of course, happiness scientists have come up with all kinds of straightforward, and actionable, findings: that money does little to make us happier once our basic needs are met; that marriage and faith lead to happiness (or it could be that happy people are more likely to be married and spiritual); that temperamental “set points” for happiness—a predisposition to stay at a certain level of happiness—account for a large, but not overwhelming, percentage of our well-being. (Fifty percent, says Sonja Lyubomirsky in The How of Happiness. Circumstances account for 10 percent, and the other 40 percent is within our control.) But why do countries with the highest self-reports of subjective well-being also yield the most suicides? How is it that children are often found to be a source of “negative affect” (sadness, anger)—yet people identify children as their greatest source of pleasure?

The questions are unresolved, in large part because of method. The psychologist Ed Diener, at the University of Illinois, has helped lay the empirical foundation for positive psychology, drawing most recently on data from the Gallup World Poll, which interviewed a representative sample of 360,000 people from 145 countries. “You can say a lot of general things from these data that you could never say before,” Diener says. “But many of them are relatively shallow. People who go to church report more joy. But if you ask why, we don’t know. George has these small samples—and they’re Harvard men, my goodness, not so generalizable. Yet he has deep data, and he brings so many things together at once.”

Seligman describes Diener as the “engineer” of positive psychology, “trying to do better, more replicable, more transparent science.” Vaillant and his work, though, remind Seligman of the roots of psychology—the study of the soul. “To practice scientific psychology is to have as few premises as you can, to account for as much of the soul as you can get away with,” Seligman says. “Everyone in positive psychology who seeks to explain the mysteries of the psyche wants deeper stuff. George is the poet of this movement. He makes us aware that we’re yearning for deeper stuff.”

When Vaillant told me he was going to speak to Seligman’s class, he said his message would be from William Blake: “Joy and woe are woven fine.” Earlier in his career, he would use such occasions to demonstrate, with stories and data, the bright side of pain—how adaptations can allow us to turn dross into gold. Now he articulates the dark side of pleasure and connection—or, at least, the way that our most profound yearnings can arise from our most basic fears.

In Adaptation to Life, where you appeared as “Alan Poe,” Vaillant had admired your altruism and sublimation, and your eloquence, but worried you were “stalked by death, suicide and skid row.” You had written in retort, “Of course, the prognosis of death is a pretty sure bet … Hell, I could be dead by the time you get this letter. But if I am, let it be published … that—especially in the last five years—‘I sure squeezed that lemon!’”

Can the good life be accounted for with a set of rules? Can we even say who has a “good life” in any broad way? At times, Vaillant wears his lab coat and lays out his findings matter-of-factly. (“As a means of uncovering truth,” he wrote in Adaptation to Life, “the experimental method is superior to intuition.”) More often, he speaks from a literary and philosophical perspective. (In the same chapter, he wrote of the men, “Their lives were too human for science, too beautiful for numbers, too sad for diagnosis and too immortal for bound journals.) In one of my early conversations with him, he described the study files as hundreds of Brothers Karamazovs. Later, after taking a stab at answering several Big Questions I had asked him—Do people change? What does the study teach us about the good life?—he said to me, “Why don’t you tell me when you have time to come up to Boston and read one of these Russian novels?”

Indeed, the lives themselves—dramatic, pathetic, inspiring, exhausting—resonate on a frequency that no data set could tune to. The physical material—wispy sheets from carbon copies; ink from fountain pens—has a texture. You can hear the men’s voices, not only in their answers, but in their silences, as they stride through time both personal (masturbation reports give way to reports on children; career plans give way to retirement plans) and historical (did they vote for Dewey or Truman?; “What do you think about today’s student protesters, drug users, hippies, etc.?”). Secrets come out. One man did not acknowledge to himself until he reached his late 70s that he was gay. With this level of intimacy and depth, the lives do become worthy of Tolstoy or Dostoyevsky.

George Vaillant has not been just the principal reader of these novels. To a large extent, he is the author. He framed most of the questions; he conducted most of the interviews, which exist, not in recordings or transcripts, but only in his notes and interpretations. To explain the study, I needed to understand him, and how the themes from his life circled back to inform his work (and vice versa).

Strenuous defenses, I came to see, are no mere academic theme for Vaillant, who has molded his life story like so much clay. Consider the story of his father’s suicide and his own delight in going through the 25th-reunion book as a 13-year-old. When I asked Vaillant if the experience of paging through the book had been tinged with sadness, he said, “It was fascinating,” and went on to describe his awe and wonder at longitudinal studies. If he were observing his own case, Vaillant himself would probably call this “reaction formation”—responding to anxiety (pain at grasping a father’s violent departure) with an opposite tendency (joy at watching men, quite like him, develop through time).

But Vaillant’s sister, Joanna Settle, described their father’s death as the “North Star” essential for navigating her brother’s story. Henry Vaillant, George’s brother, agreed. “Since that time,” he said, “it was as if George wanted to do two things. He wanted to surpass our father, and he also wanted to find out who our father was.”

Considering the Harvard study through the lens of Vaillant’s adaptations, one wonders whether he looked to do both at once. Henry Vaillant says that their father was depressed and drinking heavily at the time of his suicide; afterward, he says, his mother propagated the “heroic myth” that their father—who had worked for the U.S. Embassy in wartime Peru and, at the time of his death, was set to join the Office of War Information—was a war casualty, undone by the pressure. Does this help explain George Vaillant’s deep interest in alcoholism, and in the psychological impact of combat?

“I sometimes wondered if another motivation for the study of these lives,” says Henry Vaillant, was “to learn how to live his own life right. As if by interviewing all these very successful people, he would get the knack. And of course in many ways, he has the knack.”

Indeed, Vaillant’s work is widely read and cited; he travels the world speaking to adoring audiences (“the leisure of the theoried class,” he calls it); his colleagues and students marvel at his capacity for empathy and connection. “George sees the best in people,” Martin Seligman says, “and he brings out the best in people.”

I saw this firsthand in Vaillant’s work with H’Sien Hayward, a second-year doctoral student in psychology at Harvard with a penetrating analytical mind and a big heart. Hayward has been paraplegic and bound to a wheelchair since a car accident at 16. She studies “post-traumatic growth,” the surprising beneficial changes that many people experience after pain or injury. She approached Vaillant on a lark—she never thought someone so famous would have time to advise her. She was shocked, she told me, to see that he insisted on talking about her ideas—and about the pains and hopes that gave rise to them. “The only way to keep it is to give it away,” he told her, articulating and enacting the essence of altruism.

The experience, Hayward said, was “transformative.” Frustrated by academic politics when she came to work with him, she told me, “I felt like a little bird with a broken wing, and he lifted me back up and mended me and made me fall back in love with behavioral science—using science to understand humans and all of their complexity.” Hayward came to consider Vaillant as “the embodiment of healthy aging—mentally, emotionally, and everything. He’s the person we’d all hope to end up to be.”

But Vaillant’s closest friends and family tell a very different story, of a man plagued by distance and strife in his relationships. “George is someone who holds things in,” says the psychiatrist James Barrett Jr., his oldest friend. “I don’t think he has many confidants. I would call George someone who has a problem with intimacy.”

Nowhere has Vaillant been more powerful and articulate than in describing the importance of intimacy and love. And nowhere has he struggled more deeply in his life. He had four children with his first wife, whom he divorced in 1970 after 15 years of marriage. He quickly got married again, to a young woman he had met while speaking in Australia. She came to the United States to help raise Vaillant’s children, including an autistic son. She and Vaillant also had a child of their own. During this time, his daughter Anne says, “he was jet-setting around the world and she was holding down the roof at home.”

But in the early 1990s, Vaillant left his second wife for a colleague at the study. After five tumultuous years, he and his third wife split, and he returned (“with his tail between his legs,” his brother says) to his second wife.

This protracted drama stirred up resentments on all sides—in the women involved, for obvious reasons, but among Vaillant’s children, too. “There was a civil war in the family,” Anne Vaillant says, “and everyone suffered.” And although she says there has been some “détente,” four of Vaillant’s five children have gone long periods without speaking to him. Vaillant himself describes his family as akin to King Lear’s, and himself as “a disconnected, narcissistic father.” It struck me that the kingdom has more than an ordinary share of woes.

Vaillant’s own work provides an uncanny description of his strengths and struggles. “On the bright side,” he has written, “reaction formation allows us to care for someone else when we wish to be cared for ourselves.” But in intimate relationships, he continued, the defense “rarely leads to happiness for either party.”

Yet Vaillant seems largely unaware of the way his defenses apply to his own case—even though he is aware of being unaware; he regularly told me that he would not be a good source of information about his own life, because of distortion. The Harvard data illustrate this phenomenon well. In 1946, for example, 34 percent of the Grant Study men who had served in World War II reported having come under enemy fire, and 25 percent said they had killed an enemy. In 1988, the first number climbed to 40 percent—and the second fell to about 14 percent. “As is well known,” Vaillant concluded, “with the passage of years, old wars become more adventurous and less dangerous.”

Distortions can clearly serve a protective function. In a test involving a set of pictures, older people tend to remember fewer distressing images (like snakes) and more pleasant ones (like Ferris wheels) than younger people. By giving a profound shape to aging, this tendency can make for a softer, rounder old age, but also a deluded one. One brilliant woman from the Stanford Terman study had been pre-med in college; when she was 30, a vocational survey identified medicine as the field most suitable for her. But her ambitions were squashed by gender bias and the Great Depression, and she ended up a housewife. How, the study staff asked her at age 78, had she managed the gap between her potential and her achievement? “I never knew I had any potential,” she answered. Had she ever thought of being a doctor? Never, she said.

At age 50, one Grant Study man declared, “God is dead and man is very much alive and has a wonderful future.” He had stopped going to church, he said, when he arrived at Harvard. But as a sophomore, he had reported going to mass four times a week. When Vaillant sent this—and several similar vignettes—to the man for his approval to publish them, the man wrote back, “George, you must have sent these to the wrong person.” Vaillant writes, “He could not believe that his college persona could have ever been him. Maturation makes liars of us all.”

When we discussed his marriages, Vaillant asked me to report simply that he had been married to his present wife for 40 years, which struck me not as a calculated deception but as a deeply worn habit of thought. Indeed, a few years ago, Anne told me, her father was looking over pictures of her wedding, and came across a picture of his third wife. He stood there puzzled for a time, and then finally asked Anne: “Who is that woman?” “I began to worry that he’d begun to have Alzheimer’s,” Anne says. “But I actually don’t think it’s an organic thing. I think it’s self-protection.” This is what Vaillant calls “repression,” and he’s been using it for a long time. “When I was younger, he would forget everything,” Anne says. “It was almost like he had his brain erased.”

Vaillant has passed along day-to-day management of the study to his colleague Robert Waldinger, a researcher and a psychoanalyst. As has always been necessary, Waldinger has kept this 72-year-old ship in the water by paying homage to the dominant model of health. Today, that means taking MRIs of the Grant and Glueck men, collecting DNA swabs—and asking for volunteers to donate their brains to the study. (Meanwhile, recent efficacy studies have restored some luster to psychoanalytic ideas, so the project still encompasses a range of approaches.)

Though Vaillant spends half the year in Australia, his wife’s native land, he is still deeply involved in the study, retains his title as co-director, and operates out of the study’s office when he’s in Boston. He also works the phones to keep track of the men’s lives—and their deaths. “I’m trying to reach [name deleted],” I overheard him say one day on the phone from the study’s office. He spoke loudly; I gathered the call was overseas. “Oh. I see,” he said after a pause. “Do you know of what cause?”

Recently, I asked Vaillant what happened when the men died. “I just got an e-mail this morning from one of the men’s sons,” he said, “that his father died this January. He would have been 89.” I asked him how it felt. He paused, and then said, “The answer to your question is not a pretty one—which is that when someone dies, I finally know what happened to them. And they go in a tidy place in the computer, and they are properly stuffed, and I’ve done my duty by them. Every now and then, there’s a sense of grief, and the sense of losing someone, but it’s usually pretty clinical. I’m usually callous with regard to death, from my father dying suddenly and unexpectedly.” He added, “I’m not a model of adult development.”

Vaillant’s confession reminded me of a poignant lesson from his work—that seeing a defense is easier than changing it. Only with patience and tenderness might a person surrender his barbed armor for a softer shield. Perhaps in this, I thought, lies the key to the good life—not rules to follow, nor problems to avoid, but an engaged humility, an earnest acceptance of life’s pains and promises. In his efforts to manifest this spirit, George Vaillant is, if not a model, then certainly a practiced guide. For all his love of science and its conclusions, he returns to stories and their questions. When I asked him if there was a death that had affected him, he mentioned Case No. 47—“Alan Poe”—an inspiring, tragic man, who left many lessons and many mysteries, who earnestly sought to “squeeze that lemon.”

Joshua Wolf Shenk is the author of Powers of Two: Seeking the Essence of Innovation in Creative Pairs, forthcoming from Eamon Dolan Books/Houghton Mifflin Harcourt.