Team of Teams: New Rules of Engagement for a Complex World

by Stanley McChrystal

Management models based on planning and predicting instead of resilient adaptation to changing circumstances are no longer suited to today’s challenges.

The decision to produce yet another book to help shape and lead complex organizations did not come easily. Shelves are crammed with works of varying value, and busy leaders can feel pummeled by contradictory advice from business gurus and management consultants. But the impact of the Task Force experience drove us to test the conclusions we’d reached, because the wider implications for almost all organizations were so serious.

It is important to state up front what this book is, and what it is not.

To understand the challenge, we’ll go to factory floors with Frederick Winslow Taylor and look back at the drive for efficiency that has marked the last 150 years, and how it has shaped our organizations and the men and women who lead and manage them.

Maybe also control

For most of history, war was about terrain, territory held, and geographic goals, and a map was the quintessential tool for seeing the problem and creating solutions.

We covered the bright white surfaces with multicolored words and drawings, erased, and then covered again. We did not draw static geographic features; we drew mutable relationships—the connections between things rather than the things themselves.

Years later, in 2010, I took a teaching appointment at Yale University, where I had the opportunity to reflect on these experiences amid a community of wickedly intelligent people. One interaction proved particularly fascinating. Dr. Kristina Talbert-Slagle, a brilliant immunologist who studies AIDS, came to see me, curious about whether the similarities she saw between infections in the human body and insurgencies in a state correlated with my observations. They did; while neither HIV nor AIDS kills anyone outright, the human body is weakened to the point where it is fatally vulnerable to otherwise nonthreatening infections. The environmental factors that weaken the host indirectly strengthen and empower attackers. In 2004 I lacked this immunologist vocabulary, but I began to realize that an organization’s fitness—like that of an organism—cannot be assessed in a vacuum; it is a product of compatibility with the surrounding environment. Understanding that environment would be the key to understanding why we were failing and AQI was winning. We may have had the best equipment and the best special operations units in the world, but we were not—as an organization—the best suited for that time and place. AQI was successful because the environment allowed it to be.

Interconnectedness and the ability to transmit information instantly can endow small groups with unprecedented influence: the garage band, the dorm-room start-up, the viral blogger, and the terrorist cell. The twenty-first

Like many soldiers, I enjoy studying military history. Even the most storied of battles can contain revelations with unexpected applications. In 2004, mired in a fight against the most elusive and intractable of enemies, I was fascinated by Adam Nicolson’s Seize the Fire: Heroism, Duty, and the Battle of Trafalgar, which chronicles British admiral Horatio Nelson’s daring face-off with a superior Franco-Spanish fleet.

effective. In the military, where we love abbreviations, we have a term for the one element in a situation that holds you back—a limfac (limiting factor).

The Al Qaeda in Iraq (AQI) that our Task Force confronted in 2004 looked on the surface like a traditional insurgency. But under the surface it operated unlike anything we had seen before. In place of a traditional hierarchy, it took the form of a dispersed network that proved devastatingly effective against our objectively more qualified force. AQI’s unorthodox structure allowed it to thrive in an operating environment that diverged radically from those we had traditionally faced: the twenty-first century is more connected, faster paced, and less predictable than previous eras. Though we encountered this shift on the battlefield, similar changes are affecting almost every sector of society. To win we had to change. Surprisingly, that change was less about tactics or new technology than it was about the internal architecture and culture of our force—in other words, our approach to management.

To achieve efficiency and predictability, armies have long dressed, drilled, and disciplined men into becoming interchangeable parts of a military machine.

This drew a hard-and-fast line between thought and action: managers did the thinking and planning, while workers executed.

Taylor’s statement is offensive and inaccurate, but he was right in pointing out that many people do things in inefficient ways, and that small inefficiencies multiplied at industrial scale reduce productivity.

The scope and intricacy of production processes had expanded to the point that they required planners and coordinators to ensure that all the pieces came together efficiently and effectively. The rise of managers as the thinkers who would devise such blueprints—and the reduction of workers to instruments of implementation—seemed to follow as a natural consequence. Reductionism lay at the heart of this drive for efficiency.

All of this is true, and also the fact that markets had become larger (the entire US) which necessitated scaling production.

In 1930, France began construction of its famed Maginot Line, named for the minister of war and World War I veteran André Maginot. Like millions of his countrymen, Maginot had seen firsthand the death, disease, and suffering of static trench warfare. As the Holloway Commission would do decades later, he applied “never again” thinking to avoid repeating the horrors of the past.

Our Task Force’s structure and culture of disciplined, stratified reductionism had its roots deep in military organizational history. This organizational culture is not unique to the military; since the Industrial Revolution, most industries have subscribed to management doctrines informed by or similar to Frederick Taylor’s “Scientific Management,” a system that is excellent for achieving highly efficient execution of known, repeatable processes at scale. We were realizing in 2004 that despite the success of this approach throughout the twentieth century, it had its limits. Like the Maginot Line, it was insufficient for tackling a new generation of threats. Efficiency is no longer enough.

FROM COMPLICATED TO COMPLEX

Getting a handle on the problem of the two Tareks begins with the story of an eclectic mathematician and meteorologist working at MIT in 1961. Edward Lorenz had been using then-cutting-edge computers to try to crack weather for about a year. Weather was a tricky problem. While events such as the return of Halley’s Comet could be precisely calculated decades in advance, and tides and eclipses had surrendered long ago to scientific prediction, weather remained elusive. Lorenz hoped that the new technology would enable him to find a similar level of clockwork determinism in the Earth’s climate. He ran computer simulations, created rudimentary graphics to better visualize the trends in his data, and hoped to discern some sort of pattern.

But weather is different. Lorenz’s tiny “rounding error” existed in a more interdependent and volatile environment than the void through which Halley’s Comet orbits. Tiny eddies of air can be influenced by an almost immeasurably small event—something like the fluttering of a butterfly’s wings—and these eddies can affect larger currents, which in turn alter the way cold and warm fronts build—a chain of events that can magnify the initial disturbance exponentially, thereby completely undermining attempts to make reliable predictions. Lorenz’s program had been correct. When, several years later, Lorenz presented a paper about his findings, he titled it “Does the Flap of a Butterfly’s Wings in Brazil Set Off a Tornado in Texas?” The phrase “the butterfly effect” entered the world.*

A more lighthearted example: When musician Dave Carroll’s guitar was broken by United Airlines baggage handlers, he spent nine months navigating the company’s telephone-directory maze of customer service representatives to no avail, so he wrote a song called “United Breaks Guitars” and posted the video on YouTube. Within one day the video had racked up 150,000 hits and Carroll received a phone call from an abashed director of customer solutions at United. Within three days the video had more than a million hits and United’s stock price fell 10 percent, costing shareholders $180 million in value—600,000 times the value of the guitar. Within a week, the song peaked as the number one download on iTunes, and the company made a public show of donating $3,000 (the cost of a new guitar) to the Thelonious Monk Institute of Jazz at Carroll’s request (the makers of his broken instrument, Taylor Guitars, sent him two for free after watching his video).

Many great thinkers—most notably the nineteenth-century polymath Henri Poincaré—have observed aspects of the phenomenon that we now call “complexity,” but the concept’s coming-out party was thrown in a 1948 paper in American Science. “Science and Complexity” by Warren Weaver clocked in at a mere eight pages and involved no original research—it was an essay on the nature and aims of scientific thought—but it has left an enduring mark. Weaver argued that science up through the 1800s had concerned itself with questions of “organized simplicity”: problems involving one or two variables, like the attraction between two magnets or the rotation of the Earth around the sun.* But, Weaver observed, this was not the way much of the real world worked. Living organisms, for instance, “are more likely to present situations in which a half-dozen, or even several dozen quantities are all varying simultaneously, and in subtly interconnected ways.” Such traits, he noted, are found in ecosystems, economies, and political systems. In other words, the real world is full of the knotted interdependencies of complexity, and science was not equipped to deal with this—indeed, science actively avoided these unpleasant truths, preferring to simplify things to fit the clockwork universe. Such efforts, Weaver maintained, are futile. You cannot force a square peg into a round hole, and you cannot force the complex to conform to rules meant for the merely complicated.

Complex systems are fickle and volatile, presenting a broad range of possible outcomes; the type and sheer number of interactions prevent us from making accurate predictions. As a result, treating an ecosystem as though it were a machine with predictable trajectories from input to output is a dangerous folly.

Likewise, economic systems—the products of complex knots of human factors—confound linear attempts at prediction and control. It is because of this complexity that economist and philosopher Friedrich Hayek argued against state-run economic planning. In his landmark essay “The Theory of Complex Phenomena,” he drew a distinction between “the degree of complexity characteristic of a peculiar kind of phenomenon” and “the degree of complexity to which, by a combination of elements, any kind of phenomena can be built up.” In other words, some systems are essentially complex (like the human brain, or society), whereas other systems (like a big machine, or a factory) might appear complex because they have a lot of moving parts, but are essentially complicated. He argued that national economies, unlike industrial production, could never be transformed into mechanical systems with reductionist solutions: their behavior results from the decision making of millions of people, and all these decisions influence one another, making it impossible to forecast how markets will move—as in a game of chess, there are just too many possibilities for a prescriptive instruction card. Butterfly effects in the economy, triggered by tiny initial disturbances, are common.

A predictive hubris, perhaps bred by centuries of success at applying Newtonian models to complicated problems, has fooled us into believing that with enough data and hard work, the complex riddles of economies can be decoded.

In November 2007, economists in the Survey of Professional Forecasters—examining some forty-five thousand economic-data series—foresaw less than a one-in-five-hundred chance of an economic meltdown as severe as the one that would begin one month later.

As we have crept toward the “many to many” environment of complexity, we have engineered increasingly complicated solutions: gifted managers have developed intricate protocols and organizational hierarchies to cover all likelihoods. The baseline belief that any problem can be known in its entirety has never faded. Anyone who has worked in business or government for a few decades can testify to the seemingly endless increases in rules and paperwork. Nowhere was this more visible than at the Pentagon, where the growth of the Department of Defense manifested itself in an ever-expanding set of codes and procedures.

It is hard, if not impossible, to draw a firm line separating the complicated from the complex.

It is helpful to frame things in terms of timescale: for our purposes, we can think of a phenomenon as exhibiting complexity over a given time frame if there are so many interactions that one cannot reasonably forecast the outputs based on the inputs. By this definition, weather would be complex over the time span of a day but not a month. This is why it is relevant not only that things have become more interconnected, but also that processes have become faster.

The inevitable outcome of this approach is perhaps best summarized by Henry Mintzberg, author of The Rise and Fall of Strategic Planning: “Setting oneself on a predetermined course in unknown waters is the perfect way to sail straight into an iceberg.”

The technological changes of recent decades have led to a more interdependent and fast-paced world. This creates a state of complexity. Complexity produces a fundamentally different situation from the complicated challenges of the past; complicated problems required great effort, but ultimately yielded to prediction. Complexity means that, in spite of our increased abilities to track and measure, the world has become, in many ways, vastly less predictable. This unpredictability is fundamentally incompatible with reductionist managerial models based around planning and prediction. The new environment demands a new approach.

In the Netherlands, people are coming to understand that “the early command and control approach was not working.” Other countries and organizations are now following suit, stepping away from predictability and focusing on increasing resilience instead.

As environmentalists David Salt and Brian Walker explain in their book Resilience Thinking, Humans are great optimizers. We look at everything around us, whether a cow, a house, or a share portfolio, and ask ourselves how we can manage it to get the best return. Our modus operandi is to break the things we’re managing down into its component parts and understand how each part functions and what inputs will yield the greatest outputs . . . [but] the more you optimize elements of a complex system of humans and nature for some specific goal, the more you diminish that system’s resilience. A drive for efficient optimal state outcome has the effect of making the total system more vulnerable to shocks and disturbances.

Just as AQI had watched and learned from us at the start of the war, we would have to swallow our pride and learn from them. The messy diagrams on our whiteboards were not glitches—they were glimpses into the future organization of adaptable teams. Soon our whiteboard bore the observation “It Takes a Network to Defeat a Network.” With that, we took the first step toward an entirely new conversation.

Prediction is not the only way to confront threats; developing resilience, learning how to reconfigure to confront the unknown, is a much more effective way to respond to a complex environment. Since the pursuit of efficiency can limit flexibility and resilience, the Task Force would have to pivot away from seeing efficiency as the managerial holy grail. To confront a constantly shifting threat in a complex setting, we would have to pursue adaptability. Our foe, AQI, appeared to achieve this adaptability by way of their networked structure, which could organically reconfigure with surprising agility and resilience. We realized that in order to prevail, our Task Force would need to become a true network.

What was an embarrassment for America was vindication for coaches worldwide who have spent years saying “There is no I in team”; it proved that teams can be either far less or far more than the sum of their rosters.

United Flight 173

Compare the tragedy of United Flight 173 to the story of US Airways Flight 1549—the plane that Captain Chesley Sullenberger ditched in the Hudson River in 2009. Shortly

A hit on a moving target from seventy-five feet is, without question, difficult. But in the subculture of military snipers, it is not particularly dazzling. In 1969, a legendary Marine sniper in Vietnam shot an enemy sniper from several hundred yards away. The shot, fired with less precise rounds and from a less powerful rifle than those used today, struck the hidden Vietcong fighter in the eye after traveling through his own scope. Adept marksmen can harness the wind to curve bullets around buildings and strike targets from a mile out. In recent years, coalition sharpshooters have struck targets in Iraq and Afghanistan from distances in excess of eight thousand feet. In the Olympics of sniping, what the SEALs lying on the back of the Bainbridge did on April 12 would not even qualify for competition.

Of those who quit, only about 10 percent could not keep up physically.

Harvard Business School teams expert Amy Edmondson explains, “Great teams consist of individuals who have learned to trust each other. Over time, they have discovered each other’s strengths and weaknesses, enabling them to play as a coordinated whole.” Without this trust, SEAL teams would just be a collection of fit soldiers.

While building trust gives teams the ability to reconfigure and “do the right thing,” it is also necessary to make sure that team members know what the right thing is. Team members must all work toward the same goal, and in volatile, complex environments that goal is changeable.

When managers talk about “alignment,” they usually mean people knowing what the interim goal is at their level (the production of three axles by 5:00 p.m.). A good manager will nestle these interim goals efficiently, linking them in a tight chain that leads to the desired outcome.

One can make contingency plans, but these can account for only a modest number of possibilities. A contingency plan is like a tree that branches at every variable outcome (if they fire when we arrive, choose path A, if not, choose path B). But when dozens of saplings shoot out from those branches every second, the possibilities become so overwhelmingly complex as to render complete contingency planning futile.

Their structure—not their plan—was their strategy.

Emergence, Steven Johnson

The field of “emergence” examines how complex patterns and forms can arise from a multiplicity of simple, low-level interactions. Emergence has been used as a paradigm for exploring everything from the crystalline beauty of a snowflake, to the explosive development of cities,* to the capricious behavior of economic markets.

Adam Smith’s “invisible hand” of the market—the notion that order best arises not from centralized design but through the decentralized interactivity of buyers and sellers—is an example of “emergence” avant la lettre. It stands in direct contrast to what Alfred Chandler dubbed the “visible hand” of management—the reductive planning that has dominated most organizations for the past century. Smith’s invisible hand, like the leaderless ant colony, illustrates the core insight of emergence as it relates to our study of teams: in situations defined by high levels of interaction, ingenious solutions can emerge in the absence of any single designer; prices can settle without a central planner; complex operations can be executed without a detailed plan. Johnson describes emergence as producing “unpredictable creativity,” and identifies the ingredients necessary to unleash such creativity as “connectedness and organization.” In other words, order can emerge from the bottom up, as opposed to being directed, with a plan, from the top down.

The competitive advantage of teams is their ability to think and act as a seamless unit (this is sometimes ...