by Colin Powell
State maintains on its website background notes on every country in the world. The notes are put together by the embassies, but monitored and updated by department country and regional experts and by our public affairs office. Every few weeks, I checked the background notes list, which showed the date when each note had last been updated. More than once I found notes that hadn’t been updated in over a year. I fussed at the staff constantly to keep all our data current.
“But Mr. Secretary, we update quarterly,” my public affairs assistant secretary said, defensively and unwisely, one morning at staff meeting.
“Don’t tell me we only update our website once a quarter,” I said. “Walmart updates their entire information system whenever there’s a transaction at a Walmart checkout counter. If I wake up and see on television that a foreign leader has died and his replacement has been announced, I want that reflected on our website background note for that country by the time I get to the office. We may not always be able to beat Wikipedia or Google, but let’s try.”
Years ago I gave a speech to a large meeting of Walmart store managers and senior leadership. As I waited backstage before I spoke, the crowd was whipped up by corporate leaders to football-rally emotion. There came a huge cheer, followed by shouts and congratulations. I asked my host what was going on.
“They’ve just announced the latest sales report,” he told me.
Naïvely, I asked, “For the week, the month, or the quarter?”
“No, for yesterday,” he answered. “I could give it to you for this morning, if you like.”
I was surprised, but not shocked. I had seen it coming. Even before Google, Amazon, and the explosion of the Internet, big-box stores and supermarkets had realized that the technology revolution and the power and speed of information allowed them to move from a lunar world of calendar periods to a world of transactions.
Each transaction is flashed vertically and horizontally throughout the organization and triggers all kinds of actions. Inventory levels go down, profit is calculated, reorder formulas kick in, wholesalers and manufacturers are informed, replacement goods are assembled and loaded onto trucks, computers make projections. All of this happens in real time.
I set out to embed the same kind of mind-set in the State Department. Major change only works when followers realize your change has made their lives better and improved their productivity and performance. You only know you’ve succeeded in implementing change when your followers believe in the change and will pass their belief on to the next generation of followers. Real change has to outlive the change agents.
I never stopped pressing our people to increase their email use and update our databases with each transaction and not at the dictates of arbitrary calendar dates. Though I am long gone from the department, whenever I travel to a foreign country I send our ambassador a courtesy email from my personal account to let him know I am passing through and will be available for calls on leaders, as appropriate. I am proud to say I get very quick responses. Embassy desktops and laptops are not being used as paperweights.
In spite of my best efforts, I could never persuade my dear friend Igor Ivanov, then foreign minister of the Russian Federation, to get online. Igor’s intransigence gave me an irresistible opportunity to score points with Igor and my staff.
One day Igor called me from Moscow complaining about our UN delegation’s objections to a draft resolution his delegation was presenting in New York. Our delegation believed their resolution was inconsistent with a resolution passed by the UN some years earlier. He said we were totally wrong. I was unfamiliar with the earlier resolution, and didn’t have a clue what he was talking about. While I kept him talking, I pulled up the Google search box on my new computer and typed in the number of the earlier resolution. It came up in about a second. I let Igor ramble for a moment before interrupting him. “Igor, I am not sure you are right. If my memory serves me correctly, paragraph 2B(1) of that resolution suggests that you have misunderstood it.”
Silence. “Colin, are you sure?”
“Well, Igor,” I said, as I stared at the resolution text on my screen, “I can’t be positive, but perhaps you should have your staff take a look at it again.” It took his staff several hours to pull it all up. I was right; he was wrong. I loved it.
I never did win Igor over to computers. When he was in Washington, he often came to dinner. He always came with a gift.
Because Igor dressed exceptionally well, with a preference for blue Hermès ties, my gifts to him became Hermès ties. He was impressed. It must be lots of bother to get them, he told me. I took him downstairs to my home office and introduced him to the magic of online shopping. He watched skeptically as I ordered a Hermès tie for myself. It took about a minute.
He walked away shaking his head and muttering, “Nyet, nyet.”
Igor was no technophobe. He was a grandmaster of the other revolutionary technology of our time—the cell phone. We conducted some of the most important conversations I’ve ever had on cell phones in strange locations two continents and nine time zones away from Washington.
And there were no emails to be subpoenaed, discovered, or WikiLeaked. Hmmm, maybe Igor knew something I didn’t.
But even Igor couldn’t avoid the twenty-first century forever. Now that he has left office and is enjoying private life and successful business activities, Igor has caught up with other technologies and we email each other.
CHAPTER SEVENTEEN
Tell Me What You Know
You can’t make good decisions unless you have good information and can separate facts from opinion and speculation.
I have always been a glutton for information. I wanted an overflowing in-box, lots of people dropping in to chat, constant phone calls from the staff or trusted agents telling me what they heard and saw. Over the years I learned to read quickly to get to the essence of a paper; tossing aside filler, unnecessary adjectives and adverbs, puffery, and snake oil arguments. I took the same approach listening to oral presentations: “Just the facts, ma’am, just the facts,” an expression made iconic by Sergeant Joe Friday, the LAPD detective on the 1950s and ’60s television show Dragnet.
Facts are verified information that is then presented as objective reality. The rub here is the verified part. How do you verify verified? Facts are slippery, and so is verification. Today’s verification may not be tomorrow’s. It turns out that facts may not really be facts; they can change as the verification changes; they may only tell part of the story, not the whole story; or they may be so qualified by verifiers that they’re empty of information.
I’ve seen apparently verified facts go whoosh in the cold light of day. On March 19, 2003, the night before we launched the Gulf War, we were in the Oval Office receiving an eyes-on report from spies that Saddam Hussein was at Dora Farms, one of his palatial estates in Baghdad, which opened up the possibility that a successful attack there would decapitate the government. We bombed the place. The spies then reported they were sure they saw Hussein’s body being brought out. All wrong.
In Somalia in 1993, we were searching everywhere for the dictator Mohamed Aidid. Spies kept reporting that they had him located, but he was always gone by the time we raided the target. Spy information always has to be challenged. If the spy tells you exactly where the target is and we get it, the spy is out of a meal ticket.
The facts you are given may not add up to reveal the whole picture, but only squares on a paint-by-numbers canvas.
During the 1991 Gulf War, President Bush’s daily CIA briefer told the President that reports from General Norman Schwarzkopf were overestimating the numbers of Iraqi tanks and artillery being destroyed by our air attacks. CIA satellite photo analysts had come up with lower numbers. A huge bureaucratic battle broke out; Norm went ballistic; we set up a meeting in National Security Advisor Brent Scowcroft’s office to sort things out; and I worked to pry Norm off the ceiling of his Riyadh headquarters.
The truth was, the CIA satellite photo anal
ysts were not taking the whole picture of the battlefield into account; they were relying exclusively on narrow looking-down-a-soda-straw satellite images of the battlefield. Norm’s assessment relied on several sources—expensive satellites, inexpensive pilot eyes debriefings, and low-level aerial photos.
A pair of experts from the CIA headquarters at Langley, Virginia, attended the White House meeting: a satellite photo expert and a multisource expert, who gathered his facts from a broad spectrum of sources, not from a single, narrow soda straw. His picture of the battlefield, in other words, was very like the one Norm was seeing and that other multisource analysts at CIA, the Defense Intelligence Agency, the National Security Agency, and the National Reconnaissance Office were seeing. I laid out my explanation of Norm’s view from the field, and the multisource expert confirmed it: “Yes, that would be our assessment,” he said. Norm’s view prevailed.
Verified facts don’t always come pure, but with qualifiers. My warning radar always goes on alert when qualifiers are attached to facts. Qualifiers like: My best judgment . . . I think . . . As best I can tell . . . Usually reliable sources say . . . For the most part . . . We’ve been told . . . and the like. I don’t dismiss facts so qualified; but I’m cautious about taking them to the bank.
Don’t get me wrong. I don’t look down on intelligence gatherers, and I don’t mean to condemn any specific intelligence staff or the intelligence community. It’s a hard, stressful, vitally necessary job. During my career I’ve worked with intelligence agencies and experts of every kind, from a young lieutenant, battalion-level intelligence officer to all sixteen branches of the U.S. intelligence community. With rare exceptions, intelligence analysts do all they can to give you the information and facts you need to understand the enemy and the situation and come up with the best decision.
I found over the years that my intelligence staffs told the best story when I worked with them as they were putting it together. I questioned them constantly; I sent written analyses back, loaded with scribbles in the margins; I challenged them to defend their analyses. Staffs appreciated the challenge. They wanted to get the story right as much as I did.
Over time I developed for my intelligence staffs a set of four rules to ensure that we saw the process from the same perspective and to take off their shoulders some of the burden of accountability. The rules are simple; I’m told they hang in offices around the intelligence world:
• Tell me what you know.
• Tell me what you don’t know.
• Then tell me what you think.
• Always distinguish which from which.
What you know means you are reasonably sure that your facts are corroborated. At best, you know where they came from, and you can confirm them with multiple sources. At times you will not have this level of assurance, but you’re still pretty sure that your analysis is correct. It’s okay to go with that if it’s all you have; but in every case, tell me why you are sure and your level of assurance.
During Desert Storm, our intelligence community was absolutely certain that the Iraqi army had chemical weapons. Not only had the Iraqi army used them in the past against their own citizens and against Iran, but there was good evidence of their continued existence. Based on this assessment, we equipped our troops with detection equipment and protective gear, and we trained them to fight in such an environment.
What you don’t know is just as important. There is nothing worse than a leader believing he has accurate information when folks who know he doesn’t don’t tell him that he doesn’t. I found myself in trouble on more than one occasion because people kept silent when they should have spoken up. My infamous speech at the UN in 2003 about Iraqi weapons of mass destruction (WMD) programs was not based on facts, though I thought it was.
The Iraqis were reported to have biological agent production facilities mounted in mobile vans. I highlighted the vans in my speech, having been assured that the information about their existence was multiple-sourced and solid. After the speech, the mobile van story fell apart—they didn’t exist. A pair of facts then emerged that I should have known before I gave it. One, our intelligence people had never actually talked to the single source—nicknamed Curveball—for the information about the vans; he was a source whom some of our intelligence people considered flaky and unreliable. (They should have had several sources for their information.) Two, based on this and other information no one passed along to me, a number of senior analysts were unsure whether or not the vans existed, and they believed Curveball was unreliable. They had big don’t-knows that they never passed on. Some of these same analysts later wrote books claiming they were shocked that I had relied on such deeply flawed evidence.
Yes, the evidence was deeply flawed. So why did no one stand up and speak out during the intense hours we worked on the speech? “We really don’t know that! We can’t trust that! You can’t say that!” It takes courage to do that, especially if you are standing up to a view strongly held by superiors or to the generally prevailing view, or if you really don’t want to acknowledge ignorance when your boss is demanding answers.
The leader can’t be let off without blame in these situations. He too bears a burden. He has to relentlessly cross-examine the analysts until he is satisfied he’s got what they know and has sanded them down until they’ve told him what they don’t know. At the same time, the leader must realize that it takes courage for someone to stand up and say to him, “That’s wrong.” “You’re wrong.” Or: “We really don’t know that.” The leader should never shoot the messenger. Everybody is working together to find the right answer. If they’re not, then you’ve got even more serious problems.
We need that kind of courage. We have to encourage it in our subordinates. I hate having to say, “Jeez, why didn’t someone tell me?”
If I act on what you tell me you know and don’t know, I am adding my experience and broader knowledge to yours. If my decision turns out badly, I am responsible, but so are you, and you should expect to be held accountable. Welcome to the real world!
In 1991, as we prepared for Operation Desert Storm, our intelligence people were sure the Iraqis had chemical weapons, but there were unresolved questions about whether or not they would use them. Some analysts and experts thought they would; others thought they wouldn’t. It was a classic “don’t know” situation. We thought they had them, and they had certainly used them. But they had to fear retaliation and worldwide condemnation, and it wasn’t clear that their troops were still trained to use such weapons. I accepted this “don’t know.” We could have no certainty about whether or not they would use them until they used them . . . or tried to. And they had plenty of incentive not to.
Tell me what you think. Though verified facts are the golden nuggets of decision-making, unverified information, hunches, and even wild beliefs may sometimes prove to be just as important. Without wild beliefs there would be no stock market or hedge funds.
Your thoughts and opinions are vital, even if you can’t prove or disprove them, and even if they are nothing more than hunches. You may be right. I have frequently found that someone’s hunch is a more accurate view of reality than his knowledge. But if I act on your thinking or hunch, then I bear all the responsibility for the outcome, not you.
Many intelligence analysts and experts believed the Iraqis would use chemical weapons. That was their opinion. The facts could be taken either way. My own judgment was that they wouldn’t use them. There was too much to lose. We had communicated to them that we would respond in an asymmetric way if they did, and we left them to imagine what that might be. They were aware of our capabilities.
I further believed that we could fight through any Iraqi chemical attacks. The possible effects back home worried me—public outrage and near-hysterical reactions. But I felt we could manage these. In making these judgments, I was relying on my experience and instincts. If I was wrong, the responsibility and accountability would be upon me and not the intelligence community.
It turned out that the Iraqis did not use chemical weapons.
Always distinguish which from which. I want as many inputs as time, staff, and circumstances allow. I weigh them all—corroborated facts, analysis, opinions, hunches, informed instinct—and come up with a course of action. There’s no way I can do that unless you have carefully placed each of them—facts, opinions, analysis, hunches, instinct—in their proper boxes.
Years ago, one of my best friends, then Major General Butch Saint, got thrown out of the Army Chief of Staff’s office for delivering bad news about one of the Chief’s favorite programs. Butch knew before he walked in that he was entering the lion’s den, and he wasn’t surprised when he got thrown out. Word quickly spread around the Pentagon, as it always does when things like that happen. Not long after I heard about it I ran into Butch in a hallway. As we walked along, I offered him comforting words. “Hey,” he said quietly, “he don’t pay me to give him happy talk.” I have never forgotten that. Butch retired as a four-star general.
CHAPTER EIGHTEEN
Tell Me Early
There’s an old Army story about a brand-new second lieutenant just out of airborne jumpmaster school who is supervising his first drop-zone exercise. He is standing there by the drop zone—a big, open field—watching the approaching planes. Standing next to him is a grizzled old sergeant who has been through this hundreds of times. The lead planes will be dropping artillery, trucks, and ammunition.
Everything is looking good and the lieutenant gives the okay to drop. The first chute comes out and deploys fully. The second one is a streamer and doesn’t deploy. It hits the first one, which collapses. Subsequent chutes get caught up in the mess and they all start hitting the ground at full speed. Pieces of wreckage are flying everywhere, gasoline fires break out, touching off the ammunition and starting a brushfire that rapidly spreads into the surrounding woods.