The Inverse Law of Sanity
"Genl Wm T Sherman Insane" ran the headline of the November 1861 Cincinnati Chronicle. General William Tecumseh Sherman had gone "stark mad" and been removed from Union command in Kentucky; his peers, family, and staff all agreed that he suffered from paranoid delusions. On his way home to Ohio, Sherman said with a shrug, "In these times it is hard to say who are sane and who are insane."
He would reclaim his commission and go on to become a symbol of the Civil War's horror and a spokesman for psychological terror — the man history remembers for decimating Atlanta and scorching a trail through Georgia on his devastating "March to the Sea." He is an iconic figure in American history, yet few Americans know about an essential aspect of the man whose "scorched earth" strategy informed modern warfare from London, Dresden, and the Battle of the Bulge to Vietnam, Bosnia, and Iraq.
Historical evidence suggests that Sherman suffered from manicdepressive illness, or bipolar disorder — extreme shifts in a person's mood, energy, and ability to function. Someone need have only one manic episode to be diagnosed as manic- depressive; in fact, most people with the illness suffer mostly from depression. In addition to the Kentucky breakdown, Sherman apparently had at least four other major depressive episodes, the first at age twenty-seven, with symptoms of hopelessness, inertia, insomnia, and loss of appetite. He'd been
having trouble settling into a military career and feeling excessively controlled by his father-in-law. The second episode occurred around age thirty-seven, when Sherman was a struggling banker. Another followed a few years later, again involving financial hardship. Another, at age fifty-eight, thirteen years after the war, came after his oldest son, Tom, a deeply depressed and sometimes homeless man who ultimately died in an institution, refused to study law, as Sherman desired, and decided instead to become a Jesuit priest. (A paternal uncle of Sherman's also likely suffered from recurrent depression, a genetic link that supports this diagnosis.)
Sherman never admitted to a mental illness. In his Memoirs, published in 1875, he famously blamed others for his mistakes and finessed all questions about his mental health. Historians indulged his charitable self-image for more than a century. Only in 1995, with the work of historian Michael Fellman, were Sherman's moods more thoroughly documented. Retrospective psychiatric diagnosis is fraught with risk and never definitive. Yet this doesn't mean we shouldn't follow the documentary trail and, in Sherman's case, consider the likelihood that a man who caused so much suffering, suffered much himself.
Most of us make a basic and reasonable assumption about sanity: we think it produces good results, and we believe insanity is a problem. This book argues that in at least one vitally important circumstance insanity produces good results and sanity is a problem. In times of crisis, we are better off being led by mentally ill leaders than by mentally normal ones.
There are different kinds of leadership for different contexts. The non-crisis leader succeeds in ordinary times, but in times of crisis should be kept far away from the scepter of rule. As we'll see, the typical noncrisis leader is idealistic, a bit too optimistic about the world and himself; he is insensitive to suffering, having not suffered much himself. Often he comes from a privileged background and has not been tested by adversity; he thinks himself better than others and fails to see what he has in common with them. His past has served him well, and he seeks to preserve it; he doesn't acclimate well to novelty. We see the non- crisis leader all around us — the CEO, the department chief, your neighbor's boss, the bank president, the president. One more fact: he is quite mentally healthy. He has never suffered from depression or mania or psychosis. He has never seen a psychiatrist.
Aristotle first speculated about the link between genius and madness twenty- five hundred years ago, and at the height of the Romantic era the nineteenth- century Italian psychiatrist Cesare Lombroso defined that link forcefully, which we might translate as a simple equation: insanity = genius. He believed you can't have one without the other. In contrast, the statistician and founder of behavioral genetics, Francis Galton, took the opposing view, which we can summarize as: sanity = genius. Galton argued that intelligence — the strongest indicator of a healthy brain — produced genius. Both men saw genius as biological in origin, but one believed it arose from illness, the other from health.
These two views have seeped into Western culture, with most of us reflexively preferring Galton over Lombroso. In this book, I take Lombroso's side, with some qualifications. Throughout I trace a basic law that emerges from studying the relation of mental illness to leadership. One might call it the Inverse Law of Sanity: when times are good, when peace reigns, and the ship of state only needs to sail straight, mentally healthy people function well as our leaders. When our world is in tumult, mentally ill leaders function best.
Four key elements of some mental illnesses —mania and depression — appear to promote crisis leadership: realism, resilience, empathy, and creativity. These aren't just loosely defined character traits; they have specific psychiatric meanings, and have been extensively studied scientifically. I use these terms in their scientific, not their commonsense, meanings. Among these qualities, psychologists have studied creativity and empathy most, but resilience and realism are just as important for leadership and have also been examined in some detail by recent researchers. Of these four elements, all accompany depression, and two (creativity and resilience) can be found in manic illness.
Except for resilience, none are specific for other mental illnesses (like schizophrenia and anxiety disorders). Depression makes leaders more realistic and empathic, and mania makes them more creative and resilient. Depression can occur by itself, and can provide some of these benefits. When it occurs along with mania — bipolar disorder — even more leadership skills can ensue. In this book, I'll examine eight great political, military, and business leaders whose lives and work show various aspects of the link between leadership and madness: William Tecumseh Sherman, Ted Turner, Winston Churchill, Abraham Lincoln, Mahatma Gandhi, Martin Luther King Jr., Franklin D. Roosevelt, and John F. Kennedy. I also provide counterexamples of five mentally healthy "normal" leaders who failed in moments of crisis: Richard Nixon, George McClellan, Neville Chamberlain, and possibly George W. Bush and Tony Blair. These counterexamples are important: I am not just diagnosing illness everywhere; I see mental health in most of our leaders, and I see it as a potential impediment in times of crisis.
In the course of my research, it became clear to me that mental illness was even more influential in historical terms than I had first imagined. Several major Civil War leaders were mentally ill or abnormal: Lincoln and Sherman, as will be shown later, but also Ulysses S. Grant, the alcoholic; possibly Stonewall Jackson; even, according to some evidence of depression and a family history of mental illness, Robert E. Lee. All the major leaders of World War II can be shown, with reasonable evidence, to have been mentally ill or abnormal: Churchill, FDR, and Hitler, as we will see; as well as Stalin and Mussolini, each of whom had severe depressive episodes and probable manic episodes. Two key figures in the American civil rights movement, John Kennedy and Martin Luther King, were also mentally abnormal.
I believe these examples are more than coincidence, and more than a historical oddity. They suggest a relatively consistent pattern that, if true, has been largely ignored by historians and the public, but that may have in fact shaped the second half of the twentieth century more than any other single force. Once we start to see history through this lens, the reach and import of madness and leadership become hard to deny.
This is a book of psychology and of history; it sits at the long-disputed intersection of two different disciplines. But this book is not psychohistory. Psychohistory is a discredited discipline, and with reason. One need only read the book that started it all, written by the founder himself, Sigmund Freud's Woodrow Wilson, cowritten with the American politician (and one of Freud's patients) William Bullitt. There one finds passages like this:
[Wilson] carried great burdens during the war for a man whose arteries were in precarious condition; and, although he continued to be troubled as usual by nervous indigestion and sick headaches, he suffered no "breakdown." His Super- Ego, his Narcissism, his activity toward his father, his passivity to his father, and his reaction- formation against his passivity to his father were all provided with supremely satisfactory outlets by the war.
No wonder historians are allergic to psychological interpretation. The book was so weak psychologically that Freud's daughter and his closest disciples suppressed its publication, and when it finally appeared in 1967, they tried to argue that Freud wrote very little of it. For many historians, psychiatry and psychology are synonymous with psychoanalysis, and any psychological interpretation seems bound to end up in fruitless speculation about the early childhood traumas of historical figures. Indeed, until recently historians were correct. Psychiatry and psychology, in the United States, have long been infatuated with psychoanalysis.
Only in the last two decades has psychoanalysis been put in its proper place — not simply discarded, but no longer seen as necessary and sufficient in itself. (Imagine if all of economics was thought to be contained in Marxism; psychiatry was that dependent on psychoanalysis until recently.)
This psychoanalytic obsession has been replaced by a perspective on
mental illness that is scientifically and medically sound. This psychiatry,
stripped of its psychoanalytic faith, can be an extremely useful tool
The new psychiatry begins where modern medicine began, with the search for objective ways to diagnose illness. In internal medicine, doctors get a "case history" —a story of signs and symptoms and their course over time. Psychiatrists and historians do the same. Yet the internist has one resource that that historians and psychiatrists do not: pathology. Physicians have long disagreed with each other; one could diagnose a patient with a certain illness, and another could offer a quite different diagnosis, even given the same case history. But medicine changed dramatically when the pathologist could take a piece of tissue and determine which doctor's diagnosis was right. The doctors would discuss the case in an auditorium, with students watching, each providing
a rationale for a diagnosis. At the end of an hour's debate, the pathologist would stand up, put a slide under a microscope, and reveal the right answer.
Sometimes other tests are done: an analysis of blood chemistry, or an MRI scan of an organ. Yet sometimes these tests don't give a definitive answer; sometimes tests can even be wrong. And good doctors know that tests help us get to the right answer by adding to the evidence gathered in the case history; alone they are hardly foolproof ways to diagnose illness. Of course, tests for physical conditions are often conclusive, but the problem with psychiatry — and with history — is that there's no conclusive test. One can't prove that a patient has schizophrenia with a blood test or a brain scan; and if this is true with a living patient sitting in front of me, it is obviously so with a dead historical figure.
Yet medicine has long faced and solved this problem. Many illnesses outside of psychiatry can only be examined based on the case history — migraine, for example, and rheumatoid arthritis, and many forms of epilepsy. In these cases, doctors are in the same boat as are those who study mental illness — there's no definitive test. The solution comes from the field of clinical epidemiology, the same discipline that teased out the link between cigarette smoking and lung cancer. When there's no single proof, the solution is to obtain several independent sources of evidence. No single source is enough to prove a diagnosis, but all of them can converge to make a diagnosis likely.
Four specific lines of evidence have become standard in psychiatry: symptoms, genetics, course of illness, and treatment.
Symptoms are the most obvious source of evidence: most of us focus only on this evidence. Was Lincoln sad? That symptom could suggest depression, but of course one could be sad for other reasons. Symptoms are often nonspecific and thus not definitive by themselves.
Genetics are key to diagnosing mental illness, because the more severe conditions — manic- depressive illness in particular — run in families. Studies of identical twins show that bipolar disorder is about 85 percent genetic, and depression is about half genetic (The other half, in the case of depression, is environmental, which is why this source of evidence is also not enough on its own.)
Perhaps the least appreciated, and most useful, source of evidence is the course of illness. These ailments have characteristic patterns. Manicdepressive illness starts in young adulthood or earlier, the symptoms come and go (they're episodic, not constant), and they generally follow a specific pattern (for example, a depressive phase often immediately follows a manic episode). Depression tends to start somewhat later in life (in the thirties or after), and involves longer and fewer episodes over a lifetime. If someone has one of these conditions, the course of the symptoms over time is often the key to determining which one he has. An old psychiatric aphorism advises that "diagnosis is prognosis": time gives the right answer.
The fourth source of evidence is treatment. This evidence is less definitive than the rest for many reasons. Sometimes people never seek or get treatment, and until the last few decades, few effective treatments were available. Even now, drugs used for mental illnesses often are nonspecific; they can work for several different illnesses, and they can even affect behavior in people who aren't mentally ill. Sometimes, though, an unusual response can strongly indicate a particular diagnosis. For instance, antidepressants can cause mania in people with bipolar disorder, while they rarely do so in people without that illness.