The battlefield experience that made me realize the danger of silos and isolated teams

“Chris, are you around?” The handheld radio on my desk crackled. “Roger,” I replied. “What’s up?” I recognized the voice. It belonged to a friend of mine from a civilian intelligence team whose headquarters were located just a few hundred meters away.

“Your team’s monkey . . .” the voice on the radio came back. “She just stole our laundry.”

“On it,” I replied with mock intensity. I got up and walked out of my office into the bleaching midday sun of Afghanistan.

It was 2004, and I was based on a small compound on the edge of Khowst—an arid city located on the country’s eastern border. It was my first deployment as a part of the Task Force, and the organization was still in the extremely nascent stages of changing its operational approach. Conversations about changing our internal culture were just starting to happen among our senior leadership, but these had not percolated down into tangible changes at our tactical levels.

Our compound sat on the edge of a long-abandoned Soviet airfield with rusted, decaying skeletons of aircraft from past wars scattered around its perimeter—some of which I could recognize, like the bulbous Mi-17 transport helicopter. My morning run, which consisted of laps around the mile-and-a-half outer loop of the base, wove between these wrecks, closely hugging the twelve-foot-high meshed HESCO barriers that lined our compound.

The monkey I pursued had been living in the compound for at least a year, passed down among the Task Force’s operators as we rotated deployments, and she’d become a mascot of sorts. Recently, perhaps bored with our area of the base, she’d gotten in the habit of running along the compound’s Tetris-block walls and stealing clothes off of the laundry lines of other units. It wasn’t uncommon for our intrabase radios to come alive with a complaint about some simian shenanigans.

Responding to the radio call, I started walking across the compound with a half-eaten banana in hand, intent on winning back the monkey’s attention. As I walked, I stared to the east, at the permafrost-tipped mountain range ever visible on the horizon. Beyond those peaks lay the Federally Administered Tribal Areas (FATA) of Pakistan.

The FATA was, and remains to this day, an ungovernable space on the Pakistani side of the border. It was there where pockets of anti-Coalition fighters sat, enjoying sanctuary in that rugged no-man’s-land separating these two ancient nations. Over there, opaque enemy networks found the freedom to execute attacks against us, dropping mortar shells in our general direction whenever they pleased. Their accuracy, thankfully, wasn’t great; but it seemed that they weren’t trying to strike us directly, but instead to remind us that they were safe, secure, and waiting. The outstation in Khowst hosted an eclectic mix of personnel, who, in accordance with the teachings of thinkers like Weber, Sloan, and Taylor, were compartmentalized across different teams, each beholden to a different narrative about our collective mission. As I walked through the compound and passed the various cordoned-off sections of the base that were each unit’s territory, I could see firsthand where one unit’s narrative ceased to apply and another’s began.

A few hundred yards away from my team’s annex, intelligence analysts from civilian agencies were working to develop a network of locally sourced informants, trying to get a better understanding of the membership of those FATA-based insurgent cells. At the far end of the airfield, a conventional unit—standard infantry soldiers—had been tasked with managing infrastructure reconstruction projects in nearby towns, and so were regularly patrolling outside our compound, interacting with the local populace. On another side of the base, a small team of Army Special Forces soldiers held another center of operations, while integrated across that entire mix was a yet-more-motley assortment of Afghan soldiers, early manifestations of the larger Afghan National Army (ANA) that the Coalition would try to build up in the near future.

The people who made up these teams were outstanding professionals, easy to relate to, and exceptionally dedicated to their jobs. We were all in the same conflict, sharing common facilities, and facing similar pressures to advance our efforts. To distant strategic leaders, these overlaps gave the impression that we were synergized. But despite appearances to the contrary, the units that shared the Khowst compound were poorly interconnected and uncommunicative on matters of true substance.

Unfortunately, it was difficult for us to see this, because the Task Force’s examinations of alignment tended to be of the “congressional hearing” style, where our organization’s leaders would ask high-level questions of their subordinates. If the answers matched what the examiners want to see, they often wouldn’t notice that empirical evidence pointed to an inconvenient truth. These kinds of examinations don’t succeed, because they are calibrated to measure vertical alignment only, and miss the evidence of horizontal misalignment altogether.

Consider the discussion that occurred in the Capitol in Washington, D.C., on January 13, 2010, among Goldman Sachs CEO Lloyd Blankfein, JP Morgan CEO Jamie Dimon, Morgan Stanley chairman John Mack, Bank of America CEO Brian Moynihan, and the Financial Crisis Inquiry Commission (FCIC) of the U.S. Congress.

The Commission, a ten-person council of congressional leaders, had organized this initial three-hour gathering to supplement its investigation into the factors contributing to the 2007–2008 global financial crisis. But the Republican vice president of the Commission, California representative Bill Thomas, wisely pointed out the limited utility of the hearing. As he argued, the hearing would not be able to see the “7/8ths of the iceberg underwater,” and he was right. Opening comments from those testifying scratched only the surface of factors that contributed to the financial crisis. This was no one’s fault but simply a by-product of how difficult it is, in a normal process review, to truly tease out the realities of a complex and interconnected system.

A congressional hearing–style test for connectivity among our teams in Afghanistan would have had similar results. Had the Task Force’s leaders administered such a hearing to the teams at Khowst, it would have appeared that our efforts were perfectly aligned. Senior solid-line officers would have asked broad questions of their assembled subordinates, who would have presented a unified front and given prepared answers supporting what senior leaders expected to hear: teams are aligned, sharing information effectively, and working together. But this would have been only half true: whereas our Druckerian metrics did in fact vertically align with the strategy of our parent organization, our horizontal alignment was ineffective. What would have prevented us from disclosing this in a congressional hearing–style test for alignment— similar to that of the FCIC—would be a mix of social pressure, frustration with the narrow questioning, and nervousness about disappointing solid-line leaders.

But had our leadership isolated personnel from one another and initiated a non-blaming, unassuming series of deeper questions, the truth would have quickly emerged. Less guarded, more spontaneous answers regarding the state of alignment among our teams would have been offered. These could have taken the form of open admissions on poor horizontal alignment (You know what? To be totally honest, sir, we don’t really have the communication or exposure we need with those guys) or honest revelations about the narratives that existed at the small-team levels in our organization (This is how we actually think through problem sets we encounter and who we try to go to in our organization for help in solving them) or frank finger-pointing (What baffles and frustrates us is that those guys don’t talk to us, seem to misunderstand our work, and won’t show up when we need them to). This approach would have found contradictory views, each of which was correct in the eyes of the person offering comment.

Much like the parable of the blind men and the elephant, our various elements were each seeing individual parts of a complex and constantly changing puzzle, which was always placing new and fresh demands on them to work closely with one another. Our intelligence teams were gleaning snapshots of the insurgency’s growing influence on criminal networks in both Afghanistan and Pakistan, while our conventional and Task Force units were getting a growing sense of these networks’ increasing ability to move across the border, and the experiences with Afghan soldiers on base were indicating a steady rise in the insurgency’s ability to disrupt our recruitment of local fighters.

The significance of any one of these activities might have been painfully obvious to the team monitoring it, but tying these smaller events into a fully contextualized picture, which could better inform the operations of all of our teams, required a level of personal relationship–based interconnection among us that didn’t exist and that our senior leadership couldn’t perceive the need for.

It wasn’t necessarily personal animosity that kept us apart. Many members of these distinct teams were friendly with one another. Rather, it was the impersonal bureaucratic stove-piping that prevented us from pooling our knowledge or interacting on a deeper level upon it.

In pursuit of our mascot, I finally arrived at the intelligence team’s wing of the base and spotted the primate perpetrator. With a half banana in hand, I moved to coax her off her perch and away from the precious clotheslines. She was sitting atop a tall wall that ringed an intelligence team’s workplace, and—hearing them moving about in the courtyard on the other side—I yelled across to the analysts, “Sorry, fellas.”

A cipher-locked door sat between us, a security feature that represented the broader cultural and operational divide that was metastasizing across the Task Force’s ordered body. Whereas the exterior of the Khowst base might have made it aesthetically appear to be a sprawling, open space behind high HESCO barriers, guard towers, and a well-fortified main entrance, a knowing eye walking the interior would instead recognize it as a series of bases within bases, defined by cipher locks, guarded doors, and thick walls. Protected and maintained behind these features were the tribal narratives of our different teams.

Eventually my efforts to lure the monkey away from the intelligence analysts’ annex worked. She hopped down from the wall and, taking the banana from my hand, perched nimbly on my shoulder. Now re-united, and safe from the intelligence team’s retribution, we headed back to the Task Force’s side of the compound. I had a conference call to dial in to.

“Chris on the line,” I said, after returning to my team’s courtyard, dropping the monkey off in our annex, and joining the call from our team room. My solid-line superior—an operations officer in Bagram, Afghanistan—was leading the discussion, and the topic at hand was a recent raid that we’d been forced to cancel the night prior.

The reason for this cancellation was that a conventional unit had, unbeknownst to us, established a temporary outpost in the same area where our operation’s target was located. We had identified this issue only just before our team had launched, when I’d put in a call to a contact in the conventional unit’s command chain to notify them of our pending mission—an oversight that would have been negated had our teams been better operationally integrated.

Once identified, it was immediately clear that the risk of our forces encountering each other, and the resulting potential for friendly fire, was too high. The conventional team was already established in the area, so we’d decided to stand down. In my mind it had been a successful deconfliction, a sign that the system was working.

“From now on, no more stovepipes,” announced an unexpected voice. It had the unmistakably southern drawl of a senior officer in our bureaucracy who was in command of all the Task Force’s units in Afghanistan. A colonel, he reported directly to McChrystal, who himself was less than a year into his time as the Task Force’s commanding general. It was the first time I’d heard that word—“stovepipe”— used in our organization, but this exchange would be a simple foreshadowing of what was to come.

This tactical slipup was not the type of issue that would normally gain the attention of the colonel overseeing operations around the entire country. But he was, I would understand in hindsight, already involved in deep discussions with McChrystal about the complexity of the fight. He was leveraging this after-action discussion as an opportunity to educate all of us on the direction things would be heading in soon.

Similar discussions were in their early stages across the Task Force. As the colonel spoke about collaboration with other organizations, about early synchronization of efforts versus last-minute deconfliction of actions, and about building truly effective relationships with other organizations, my thoughts drifted to the walls, cipher locks, and security procedures that still separated the various units within the Khowst com- pound. These were products of, and contributors to, the narrative gulfs that helped keep our organizations compartmentalized.

I realized just how far we had to go. There was no single person who bureaucratically owned this issue, no standalone order that would force us to collaborate. This would be a culture change, something that would take years—but this was a start.

This article is adapted and excerpted from ONE MISSION: How Leaders Build a Team of Teams, now on sale from Portfolio/Penguin Random House.

Read more: Gen. Stan McChrystal gives a roadmap for building your own team of teams.