Nine Lessons from a Career of Analysis

I think of the craft of intelligence as telling stories and adjusting them. I also have come to think that people remember analytic points better if they are attached to a story, or an arresting fact. So here are anecdotes that produced critical lessons in my life as an analyst.

by Gregory F. Treverton

Note: The views expressed here are those of the author and do not necessarily represent or reflect the views of SMA, Inc.

My first lesson was perhaps the most important. It was my first job in government as a young whippersnapper on the staff of the first ever Senate committee to investigate intelligence. When Salvador Allende was elected president of Chile in 1970, the Nixon Administration pressed President Eduardo Frei, in power, to support a coup or other means to deny Allende the presidency. Desperate, the administration soon set in motion a parallel operation, creatively labelled “Track II,” to be done urgently and without the involvement of the embassy. It differed primarily in that it was prepared to countenance a coup without Frei’s acquiescence.

And so, in the early summer of 1975 I was deputed to meet with Edward Korry, who had been the American ambassador to Chile in 1970. A journalist and editor who had worked in post-war Europe, Korry had something of a reputation in the State Department for his “Korrygrams,” sometimes undiplomatic musings rendered in colorful prose. I quickly finished my main mission and determined that he had not known of Track II. But I was treated to five hours of verbal Korrygrams aimed at both Chileans and others who had let him down. I managed only a handful of questions but did get his agreement to testify in open session; to be sure he would be a dramatic and voluble witness!

With that, I passed on to other work. Trouble was, the open session kept getting postponed as the committee argued with the Ford Administration. Before it happened, the Committee issued its interim report on alleged assassination plots involving the CIA.[1] One of those was connected with Track II and ended in the killing of the Chilean chief of staff. The interim report made clear that Korry had not known of Track II, but it did quote several Korrygrams: “Not a nut or a bolt will be allowed to reach Chile under Allende…We shall do all within our power to condemn Chile and all Chileans to utmost deprivation and poverty.”[2]

A day or two later I was sitting in the office when I got a call from Korry. He was angry and blunt: he had contemplated suicide once before, and if he did so again, it would be on the committee’s and my head.[3] I instantly understood my mistake: I had been focused on substance, made the confirmation I needed and identified Korry as ideal for the public session. I should have understood that he was on the edge: what I needed to do was invite him to Washington, get a stenographer and let him spill it all out for a day on the record.

The lesson has stayed with me for a lifetime. It is all too easy for us analysts to become preoccupied with the issues—my ex-wife reminds me that analyst begins with anal—after all, they are what drew us to public service. Yet behind the issues are human lives, full of hopes and fears, dreams and nightmares, marriages, and mortgages. We need always to be mindful of those.

A second lesson derives from the beginning of the Carter administration, in which I handled Europe for the National Security Council. While on an early trip to Europe, Vice President Mondale was impressed by King Juan Carlos of Spain, and asked the President to call him. I was proud of the talking points I did, but only when I was crossing the street to the White House did I realize I had done nothing to set up the call. I sheepishly went to the appointments secretary, Tim Kraft, and suggested we postpone. But he picked up the phone and said to the redoubtable White House telephone operators—remember this was the 1970s— “I’m placing a call for the President to King Juan Carlos, Madrid Spain.” I was the notetaker, so I sat on a couch in the Oval Office with the extension phone to my ear. Minutes that seemed like hours passed. I began to sweat. Shirley MacLaine arrived to see the President. Finally, a voice came: “This is Juan Carlos.” I stammered, “Your majesty, I’ll put the President on.”

The lesson is not so different, perhaps, from my Korry story: substance matters but good staff work is imperative.

The third lesson I learned while I was a senior fellow at the Council on Foreign Relations in New York. It was then a small operation, a handful of fellows who ran their own fiefdoms and a director of studies. But the director, more devious than micromanaging, would buttonhole my secretary looking for information. In the end, he undercut me with a major foundation supporter of mine.

The lesson was one I had applied in another form to intelligence but now realized could apply to organizations as well. For intelligence, it was “the less you understand about them over there, the more you need to understand about you, your organization and its proclivities.” A friend and colleague, Richard Fisher, who later become president of the Dallas Federal Reserve, put it colorfully. He had won the Democratic primary for a Senate seat from Texas. However, the party was liberal while he was conservative, and so it sat on its hands while he got creamed by Kay Bailey Hutchison. Richard’s lesson: you can always tell your friends from your enemies because your friends are the ones that stab you in the chest!

Lesson four came to me slowly through intelligence but was sharpened in my first stint at the National Intelligence Council (NIC), the government’s lead interagency group for intelligence analysis. I know it is true of intelligence. I think it applies to life as well. In my second stint at the NIC I had it engraved on paperweights for my closest colleagues (though I did it small enough so they could actually have the paperweights on their desks): assumption is the parent of fuck-up.

Again, another colleague had expressed an even more colorful version years earlier. Bob Baraz, a State Department intelligence officer who knew more about the military balance on the central front in Europe than anyone alive, would respond when anyone came to him with a conditional woulda, coulda, shoulda by saying, “Yeah, yeah, yeah, and if I had a square asshole, I could shit bricks.”

I was at RAND for nearly twenty years off and on, and that left me with four lessons. The first was again about the importance of staff work. I didn’t do many projects for the military, but I realized that, if the project were to succeed, I’d need the equivalent of a three-star to be interested, but I’d also need a lieutenant colonel to be available and make the trains run on time.

A second lesson was procedural and perhaps rather particular to RAND. RAND was an internal market, with managers looking for analysts and analysts looking for projects. As an analyst you couldn’t always count, in any given year, on sponsors asking for what you wanted to do. So you needed to be the go-to guy or gal for some set of issues but also have other arrows in your analytic quiver. In one dry period, I ended up doing a project for Reserve Affairs. It was a great idea, using the reserves to get access to tech talent by relaxing some of the physical requirements. The sponsor hated it. The reserves want to be holier than the pope, just as rigorous as the active forces, if not more so.

The third and fourth RAND lessons are directly about analysis and are straightforward if often ignored – including by me. I came across a study I wish RAND had done, which looked across the intelligence agencies asking what methods they used in thinking about the future. The answer was doubly bad: most agencies used no method, and to the extent they did, it was one we know is flawed, what we used to call BOGSAT – bunch of guys and gals sitting at a table. We know that is a recipe for groupthink, so better to consult experts individually, then combine their answers. In any case, use a method when you can.

The last RAND lesson is even more elementary: pay attention to the numbers. In the mid-1990s Mexico had launched a new financial instrument, called the tesobono. Both Wall Street and the U.S. Treasury were enthusiastic about these bonds and told a complicated story about how they were no serious risk. However, at the NIC, the National Intelligence Officer (NIO) for Warning was worried. She was not an economist by training but simply watched Mexico’s foreign exchange reserves going down. (In fact, the decline was still more than she chronicled because Mexico was cooking the books.) Mexico will have to devalue the peso, she warned, which got her as rebuttal the complicated story from Wall Street and Treasury. In the end, she was right, and Mexico had to devalue the peso in 1994.

In the run-up to the Russian invasion of Ukraine, I made two rookie mistakes, which my former colleagues in intelligence avoided. The first was not paying attention to the numbers. I assumed Putin was making a show of force to get serious conversations with the United States and NATO. But Russia had deployed many more troops than it needed for that purpose. It had to be an invasion. I compounded my error by mirror-imaging: I asked myself why Putin would invade Ukraine and could think of no reason. But I am not Putin.

My final lesson comes from my second stint at the NIC, as Chair, and it is perhaps more about management than analysis. Another wise colleague, Hal Sonnenfeldt, observed that you go into government itching to accomplish several weighty aims but quickly realize that you will be a success if you just prevent a couple of idiotic things from happening. I had realized over my career how easy it is to be driven by your inbox, which can be overwhelming. So, I hired a full-time strategic planner. We spent time working through what I wanted to accomplish while at the NIC, then he developed benchmarks for how we were doing. I made sure we met every other week to discuss progress. I wished I had done something similar many times earlier in my career. As the old saw has it: if you don’t know where you want to go, any road will get you there.

Gregory F. Treverton stepped down as Chair of the US National Intelligence Council in 2017. He is now Professor of the Practice at Dornsife College, University of Southern California, Chair of the Global TechnoPolitics Forum, and an SMA Executive Advisor.

[1] See Alleged Assassination Plots Involving Foreign Leaders, An Interim Report of the Select Committee to Study Governmental Operations with Respect to Intelligence Activities, United States Senate, 94 Cong.; 1 sess., November 18, 1975, available at https://www.aarclibrary.org/publib/church/reports/ir/pdf/ChurchIR_0_Title.pdf.

[2] Anthony Lewis, “Not a Nut or a Bolt,” New York Times, November 25, 1973, C35.

[3] Ambassador Korry has been dead for a number of years, and I feel confident he would not mind my telling this part of the story to draw lessons from it, especially for people just embarking on public service.