A Policymaker's Perspective on Intelligence Analysis
A Policymaker's Perspective On Intelligence Analysis
Editor's Note: This article is based on the author's interviews during 1991-93 of Ambassador Robert D. Blackwill. The author has written several other articles on intelligence and policy.(1)
Sherman Kent, in Strategic Intelligence for American World Policy, his path-breaking effort to join intelligence doctrine and practice for post--World War II America, concluded that:
There is no phase of the intelligence business which is more important than the proper relationship between intelligence itself and the people who use its products. Oddly enough, this relationship, which one would expect to establish itself automatically, does not do this. It is established as a result of a great deal of conscious effort . . .(2)
Despite guidance from Kent and numerous subsequent authors, the terms of engagement between intelligence analyst and policymaker are still ill-defined doctrinally and thus practiced as much to suit the immediate preferences of the players on both sides of the relationship as to meet the fundamental demands of sound policymaking. The quest to join sage principle--what should work--to solid practice--what does--is more important than ever in post--Cold War America, as resources for intelligence support of policymaking are cut back more rapidly than responsibilities.
The original pillar of Ambassador Blackwill's doctrinal views on intelligence and policy was self interest--his effort to make the relationship work for him personally under trying conditions. He served as Special Assistant to the President and Senior Director for European and Soviet Affairs, National Security Council Staff, during 1989-90, a tumultuous period that witnessed the collapse of the Soviet Union and the reshaping of Europe. The more lasting pillar is his concern for the national interest--a belief that the United States can ill afford prevailing patterns of ineffective ties between experts on events overseas and policymakers in Washington.
Some Key Points
The Ambassador's framework for defining the requirements for sound intelligence--policy relations consists of four key points:
- Roughly 90 percent of what passes for national security analysis in the US Government, including structured study of events overseas, is done by intelligence analysts.
- The national interest requires that this effort be effectively joined to the policymaking process.
- The officials who carry most of the day-to-day burden of policymaking on key issues are so besieged by time-consuming responsibilities that decisions on how much to stay informed on events overseas and in what way are narrowly based on self interest in managing the pressures and getting the job done.
- Intelligence professionals have to carry nearly all the burden to convince each key policy official that they are committed to servicing his or her analytic needs via customized expert support.
Thus, to meet their responsibilities in promoting the national interest, intelligence professionals have to become expert not only on substantive issues but also on serving the self interest of policy professionals by providing specialized analytic support.
Ambassador Blackwill's career as a Foreign Service Officer began in 1967, after a stint in Africa with the Peace Corps. During 1979-80, he served on the NSC Staff as Director for West European Affairs. In the early 1980s, he worked at the State Department as Principal Deputy Assistant Secretary, first for Political-Military Affairs, and then for European Affairs. From 1985-87, he was US Ambassador and head of the US Delegation to the NATO--Warsaw Pact negotiations for reduction of conventional military forces in Europe.
Ambassador Blackwill's recent tours of duty as policy decisionmaker have been interspersed with periods as an administrator, lecturer, and program director at Harvard University's John F. Kennedy School of Government. In his current stay at Harvard, Ambassador Blackwill directs a program on public policy for the Russian General Staff. His publications--including his latest book, New Nuclear Nations(3)--address issues of arms control and European affairs rather than intelligence per se.
A Shaky Start
I first met the Ambassador in November 1987, when he was teaching in the CIA-funded Kennedy School Seminar on Intelligence and Policy. He seized the attention of the class of some 30 Directorate of Intelligence (DI) division chiefs and managers from elsewhere in the Intelligence Community by asserting that as a policy official he never read DI analytic papers. Why? "Because they were nonadhesive." As Blackwill explained, they were written by people who did not know what he was trying to do and, so, could not help him get it done:
When I was working at State on European affairs, for example, on certain issues I was the Secretary of State. DI analysts did not know that--that I was one of a handful of key decisionmakers on some very important matters. Why bother to read what they write for a general audience of people who have no real responsibility on the issue.
More charitably, he now characterizes his early periods of service at the NSC Staff and in State Department bureaus as ones of "mutual ignorance":
DI analysts did not have the foggiest notion of what I did; and I did not have a clue as to what they could or should do.
An unpromising start. Yet during his 1989-90 NSC Staff tour, Ambassador Blackwill--by the lights of DI analysts working with him on European affairs--raised analyst--policy relations to an exemplary level. Time after time, the DI's Office of European Analysis (EURA) provided much-needed intelligence support under stringent time constraints. In a tribute with resonance in the hometown of the Washington Redskins, Blackwill called the EURA crew his "analytic hogs," opening up holes in the line for him to run through. At least one EURA analyst considers this period "the most exciting and meaningful" of his career.
The balance of this article consists of the Ambassador's replies to my questions.
From Mutual Ignorance to Mutual Benefit
Q: What caused your apparent change of mind about the utility of DI analysis?
A: I had started to rethink my position even before our 1987 classroom encounter. As chief negotiator for the MBFR talks,(4) I worked closely for the first time with Agency analysts--those assigned to the US delegation. They regularly came up with information and interpretations that helped me sharpen my approach to the individual negotiating issues. When I gave them a special task, they delivered to suit my schedule, even if it meant considerable inconvenience to them.
One more matter important to negotiators, and to heavily engaged policymakers generally. Unlike other intelligence people I had worked with in the past, including those from State, my informal talks about possible US tactical initiatives with CIA analysts from the Arms Control Intelligence Staff did not end up in The Washington Post.
My understanding of the role of intelligence was also broadened by my work at Kennedy School. In addition to the CIA seminar, I collaborated with Professors Ernest May and Richard Neustadt on a course called "Assessing Other Governments." Here, the importance of country expertise, of language skills, of perspective and a sense of history were underscored by well-documented case studies.
Then there was the survival factor. I knew soon after [President] Bush's election in November 1988 that I was to be selected for the NSC Staff job on both Europe and the USSR. This meant longer hours and more pressures for me than ever before. Frankly, I was concerned about forgetting what my 10-year-old daughter looked like. So I sat down in Cambridge and planned how I was going to interact with Executive Branch colleagues, with Congress, with the press--and with intelligence. I decided that in my own self interest I had to arrange to get as much support as practical from Agency analysts.
Q: Why Agency analysts?
A: You mean besides the fact there are many more of them in my areas of responsibility than in the other intelligence outfits? My experience at State convinced me that INR [Bureau of Intelligence and Research] works for the Secretary. I suppose it is the same at Defense. I judged that Agency analysts would be much more likely to provide close and continuous support to an NSC Staff director.
Back to State. From my White House perspective, the State Department almost never met a deadline it could not miss. Then there is also the confidentiality factor. As I said earlier, your musings about possible policy initiatives are not leaked to the press by the DCI to shoot down your policy.
The most important consideration is that Agency analysts are better informed about individual countries than anyone else in the [US] Government. And I judged they had the wit--the historical perspective I spoke of--to interpret this information for my benefit and the President's benefit. I just had to determine whether they had the professional interest and enterprise to be responsive to my overtures.
Let me expand on one point. Intelligence analysts--essentially DI analysts--do 90 percent of the analysis by the USG on foreign affairs. Policy officials, even those with academic backgrounds, are too busy with more pressing matters.
In some administrations, the most heavily engaged and influential policy officials on any given issue spend 90 percent of their time assessing their policy competitors in Washington. I am talking here about getting ready to leverage competing Administration officials, not just Congress. Busy decisionmakers concentrate what little time they have for foreign policy analysis on narrowly focused aspects of key agenda issues--often how to deal effectively with their foreign counterparts. Let me tell you, any policy official who can do his own research on all aspects of an issue, cannot be very important--because he is not fully engaged in the coalition-building and power-leverage games essential for getting serious policy work done in Washington.
And there is no second team. If Agency analysts do not do the work of keeping up with developments overseas that the decisionmakers need to know about, it does not get done. It was in my self interest to see if I could get those analysts working for me, to help me keep up with a broad range of developments I could not possibly follow on my own.
What Works, and What Does Not
Q: You have mentioned self interest a couple of times.
A: Let me explain. The policymakers who count the most--those five to 10 on any issue who have the most power for getting anything done, decided, implemented--work much harder than intelligence analysts. During 1989-90, I was often at my desk from 7 in the morning till 10 at night. Others at the NSC Staff, Brent Scowcroft and Bob Gates for instance, started even earlier.(5) Unlike analysts, we had no evening tennis games. No weekends.
Even with these hours, as I indicated, I needed help to stay informed. But it had to be the right kind of help. I could not afford to read intelligence papers because this or that intelligence agency was entitled to produce them. It did not matter to me how much work the Agency had put into its products, or how polished they were in scholarly terms. In fact, I could not afford the time to read intelligence papers written by personal friends and colleagues. I could only read intelligence products tailored to help me get through my substantive schedule. There was no other rational choice.
Q: The old issue of "adhesive analysis."
A: You asked, so let me unload here. During my [1989-90] NSC tour, the Agency was still putting out gobs of analytic products that I never read. During the two years I did not read a single [National Intelligence] Estimate. Not one. And except for Gates, I do not know of anyone at the NSC who did. The reason, at least for me, is simple. There was no penalty to be paid for not reading an NIE. It did not cost you anything in terms of getting done the most important policy things you had to get done.
The same goes for your other general audience papers. I got them, but I did not read them. I am sure somebody did, or you would not bother to put them out. Let me grant without hesitation that there is a lot you put out for good reason that has nothing to do with policymakers at my level. I think, however, that you ought to consider the cost--benefit ratios of producing papers that are read mostly by specialists at the desk level at State and Defense, or by policy officials with general interest but no direct say on an issue.
Q: What about the NID [National Intelligence Daily]? I've heard a number of NSC Staff members praise its utility over the years.
A: Of course, I was interested in the PDB [President's Daily Brief] because President Bush read it. As for the NID, I would spend, literally, 60 seconds a day on it. This was a defensive move. I wanted to know in advance what would likely be leaked to the press by readers in Congress. Other than that, there was, again, no cost to me, no penalty, from not having read the NID.
Q: What did you read, aside from what you commissioned directly from DI analysts?
A: Despite what you hear about policymakers not having time to read, I read a lot. Much of it was press. You have to know how issues are coming across politically to get your job done. Also, cables from overseas for preparing agendas for meetings and sending and receiving messages from my counterparts in foreign governments. Countless versions of policy drafts from those competing for the President's blessing. And dozens of phone calls. Many are a waste of time but have to be answered, again, for policy and political reasons.
Q: Let's turn to what you commissioned from DI analysts.
A: One more minute, please, on what I did not find useful. This is important. My job description called for me to help prepare the President for making policy decisions, including at meetings with foreign counterparts and other officials. One thing the Agency regularly did was send me memos on the strategic and tactical agendas of foreign officials; in effect, what they wanted from the United States. Do you think that after I have spent long weeks shaping the agenda, I have to be told a day or two before the German foreign minister visits Washington why he is coming?
O.K. What did I want from analysts? I want their reading of what is going on in the domestic affairs of country "X" or "Y"--countries the President is planning to visit to advance foreign policy or countries from which we are going to receive important visitors to discuss problems and bilateral strategy, or countries on which, for one reason or another, we feel a need to get US policy into better shape.
What is going on domestically in these countries that could have an impact on how the President's counterparts and my counterparts will behave? What pressures are they under at home? Although I knew the national security issues cold, I could not become expert on all important issues affecting Germany or France or Italy at the national level, much less at the provincial or state levels. DI analysts knew this, and they helped me bone up on what I needed to understand to nuance and sharpen the US approach.
You also have to consider that President Bush, as a political animal, was naturally interested in the domestic politics of other leaders, even when there was no pressing bilateral business on the table.
Q: We variously call this "opportunity analysis," or "value-added analysis." Sometimes we call it "targeted tactical analysis."
A: I never put a label on it. Your terms are all good ones. Incidentally, the MacEachin metaphor you told me, about scouts and coaches, is also useful.(6) Yes, intelligence analysts should help key policymakers make the best game plan by telling them what they do not know or appreciate sufficiently. Regarding my own needs, this was mostly, as I said, on the domestic politics of the countries I was dealing with.
Whatever label you put on it, the service I got on Europe from EURA was superb and invaluable. As you know, when I traveled to Europe, EURA analysts prepared a daily cable for me on key developments. They got it to me first thing in the morning European time, which means they worked late into the night in Washington to get it done. I appreciated that immensely. Once a senior State Department colleague joined me for breakfast in Brussels as I was reading my very own newsletter. He studied it with great interest and asked me where it came from. I chose not to give him a clear answer.
EURA people met without exception whatever deadlines I set for informal memos while I was in Washington. They also were responsive and quick with some major projects I laid on with little advance notice. My only problem with their written work is sometimes the text had gone through too many levels of review and began to read like a NID article. If I wanted a NID article, I could read one. What I wanted was the analyst's unvarnished response to my questions. After I made this point, the incidence of overpolished papers diminished.
Qs: What about briefings?
A: Yes, because you get a chance to ask questions, briefings can be more helpful than memos. Here, too, I got first-rate customized service. Whenever I asked for briefings in my office, the analysts who came were both informed and responsive. Really terrific people.
Again, I was mostly interested in domestic affairs in this and that country. From time to time, though, I would ask the analysts in my office what the response of a European government would be to the policy initiatives the President was considering or that I was thinking of recommending to the President. Their unrehearsed responses here were also useful. I always hesitated to put such requests into writing for fear of leaks to the press. I learned you can trust DI analysts. They were well informed. Ready to help. And they kept their traps shut.
Q: That sounds like a good advertisement for DI analysts.
A: You bet. They were expert on their subjects. They were responsive to my needs. And they did not leak my confidences to the press.
Politicization Not an Issue
Q: Did your NSC Staff colleagues resent your close ties to DI analysts?
A: Not that I was aware of. The people who worked for me, rather than being resentful, made use of EURA support on their own.
Q: What about this kind of closeness pushing analysts across the line into policymaking?
A: Again, I saw no problem with EURA analysts. When I asked, they provided advice on tactics to support an established policy. They were good at that too. But the EURA people did not get into policy prescription. And where it did happen on occasion with others, when intelligence people started recommending policy, I pushed them back.
Q: What about telling you what you wanted to hear, or avoiding bad news?
A: Not a problem. I wanted their help in avoiding setbacks as well as for advancing policy goals. If there were negative developments I had to know about, they let me know. We had trust going both ways.
I would like to continue with this for a minute. I know during the Gates confirmation hearings [for DCI during 1991] the media were full of charges of analysts writing to please policymakers. My experience was different. I would argue that at least in my experience close professional relationships encouraged frankness--not politicization. But I know it does not always turn out that way.
Just as top policy aides have got to deliver bad news to the President when called for, intelligence people have got to have the intellectual courage to tell key policy officials that something is not working, or is not going to work. It is tough, really tough, to stop a policy failure based on ignorance of the ground truth. Intelligence analysts have got to rise to this challenge. I am not talking about shouting it from the rooftops. NSC directors are especially resentful when Congress is told bad news before they have a chance to think about it. But limited distribution memos should work. Private briefings might be even better, since that gives the policy official a chance to ask questions.
Often it is important to decisionmakers to know how to get to the least bad outcome, to limit the damage. I think options papers work very well here, especially if they are delivered after bad news forces key policymakers to focus on an issue. Somalia is a good example. The analysts could table a paper or lay on a briefing outlining three possible outcomes six months down the road, and what opportunities, leverage, and so forth the United States has to influence the outcome.
Intelligence and Policy Tribes
Q: Why do not more overworked policy officials lean on Agency analysis the way you did?
A: I guess some do, though I do not personally know of any case quite like mine with EURA. The absence of a pattern of effective relations probably reflects a combination of professional differences and mutual ignorance about what really makes the relationship work.
I am not the only policy official who decided that too many intelligence products still are nonadhesive. They are, or were when I last served, too long and complex. Analysts love words and complexities; it is one of their strengths. Good policymakers are driven by the need to take action. They need problems broken down, simplified. You and I have been through this before, and you can probably make a better list of tribal differences than I can. The key still is getting close enough to the individual policymaker to find out what he needs.
Policymakers do not as a rule know what intelligence analysts can do for them. They read Estimates, think pieces, the NID, and say, in effect, "What does this have to do with my problems"? They do not see it as their job to teach analysts how to be helpful. Besides, they would not have the time.
Q: How did your counterpart NSC Staff senior directors stay informed, and, for that matter, others in the Bush administration who were the kinds of key hands-on policy officials you think the Agency should cultivate?
A: The only honest answer is, I do not really know. I was too busy with my own affairs. But I seriously doubt that any of them [during 1989-90] received the kind of customized support from the Agency that I am talking about.
A: Yes. As a citizen and taxpayer it sure does. I am talking here about the national interest. Let's go back to my statement that the Intelligence Community does 90 percent of foreign affairs analysis in the USG. Policy choices are made and policy actions are taken whether or not the expertise of analysts is brought to bear. But how can anyone argue that we should pay for this expertise and not make use of it?
I do not mean to say it is all the analysts' fault, but I am fully prepared to argue that if an analyst's work does not have an impact on policymaking as a process, including in the long run, he or she is taking pay under false pretenses. A lot that you do is useful to someone. You have to make it more useful to those who count.
Let me say this: the Agency's understanding of the world is probably needed more today than ever. The world and the challenges the United States faces are changing so rapidly. Also, the new [Clinton] administration does not seem to have yet defined its policy approach. The costs of tribal tensions between analysts and policymakers--mutual ignorance, really--may be rising.
At a Lower Level
Q: Much of what you have had to say relates to officials at your level, the NSC senior director and departmental assistant secretaries and above. What about one level down--deputy assistant secretaries, office directors?
A: I would say, much the same. Find out who counts--the five or 10 midlevel officials who have the most influence on more senior decisionmakers--and cultivate close relations with them. Trade customized support for access to the real agenda, and so forth.
A Program for the DI
Q: How would you combine your various recommendations for Agency analysts into a program? If you were advising the DCI or DDI, what measures would you propose to enhance the effectiveness of relations between analysts and policy decisionmakers?(7)
A: Thank you for letting me know in advance this question was coming. It is a good question, and I have given it considerable thought. Let's see if the seven measures I have sketched out add up to a program.
- For new appointees, send a letter asking for an appointment and spelling out your areas of expertise and the services you are ready to extend.
- For serving officials, anticipate a major pending visit or event and offer to send over your analysts for a briefing on any one of several related aspects. For example, if the prime minister from Denmark is to visit the President, the DI manager should signal that he will bring over his Denmark analyst to fill the policymaker in on any gaps in understanding in time for the latter to prepare briefing memos for his or her principal, be it the Secretary of State or the President.
- Whenever DI managers know of travel plans by a key policy official, offer to send over country analysts who can fill in the official's knowledge on areas of his choice.
- Have the DCI set up luncheon meetings in town (CIA Headquarters is just too inconvenient), at which analysts and their managers can establish their credentials as entrepreneurial experts.
1. Identify the 30 or so senior policy officials who count--those who really carry weight with administration Cabinet officers on key foreign policy issues. These officials, usually assistant secretaries in policy departments or special assistants to the President on the NSC Staff, regularly set the thinking of NSC principals on major policy decisions. As a rule, these are the assessors of foreign governments, or the analysts of last resort. To contribute to sounder policymaking, intelligence analysts have to reach this group. Remember, the list of policy notables has to be carefully worked out and kept up to date, because office titles do not always reflect real policy weight.
2. Approach the policy officials who count as if they were motivated solely by self interest. Their self-interest has to be worked on because they are just too busy to allow either institutional considerations or personal friendships to determine their attitude toward intelligence analysts.
3. Learn as much as you can about each senior official. Study them as carefully as you do foreign leaders. For example, read everything they have written on the subjects in their policy portfolios. Check them out through mutual contacts.
4. Take the initiative to establish ties. This is an essential obligation of intelligence managers, because policy officials will rarely seek them out.
5. Customize intelligence papers and briefings to solidify the relationship. Many policy officials, overwhelmed by the volume both of their activities and of seemingly important information, will welcome specialized newsletters. They will welcome even one-page summaries of key events overseas that provide the kind of information and analysis they want at the time of day or week they prefer to set aside for keeping up with developments. For the same reason--fear of being overwhelmed--many will welcome customized briefings and memos relating to their policymaking responsibilities on matters on which the DI country analyst is much better informed than they can be or than anyone else in the government. Give them something they will really miss if they do not get it.
6. Place the best and most promising analysts on tours in the policy world. The Agency could offer, free of charge, 50 first-rate people to policy officials around town. Intelligence officers can learn something about how to use intelligence resources effectively by reading about policymaking. You can learn some more by periodic visits to a policymaker's office. But the best way to learn about a different bureaucracy is just the same as the best way to learn about any alien tribe--go live with them for a couple of years.
7. Reward those managers and analysts who are successful in gaining and maintaining access. As a rule, once a win-win relationship takes hold, momentum will keep it going. Once the policy official knows the intelligence unit can and will deliver support when it is needed, he will provide in exchange access to the real policy agenda. But policy officials come and go, and the Agency has to take care of those with talent at starting over again with newcomers who, as almost always will be the case, will not seek you out.
Q: How do you stay informed on events overseas these days, while working again at Kennedy School?
A: My main current interest is Russian politics and military affairs. I have been spending one week per month in Russia, dealing directly with the General Staff. While at Harvard, I spend a couple of hours each morning on Internet. It is amazing how much good information and worthwhile commentary is out there for those with the interest and the time. While at the NSC, I had the interest but not the time.
Internet, CNN, increasing visits by all sorts of Americans. The competition for the DI analyst is becoming much stronger. This means you are going to have to work much harder to find a comparative advantage. How do you get more expertise--living there, of course, language, and history?
I worry a bit about this. Just as you cannot rely on quality alone to get your job of informing policy done, you cannot rely only on access. In fact, marketing without a quality product to deliver is worse than passivity.
Q: Final question. At the end of a long day, which is it, working for more expertise, or for more access?
A: The answer, I suppose, is more efficiency. I imagine a textbook breakdown would have the analysts spending 40 percent of their time on collection and other activities for building expertise, 30 percent on analysis and writing--putting things together, and 30 percent on assuring impact on the policymaking process. I never managed an analytic unit, and this is just a guess. I do not think you are anywhere near the last 30 percent.
One final thought occurs to me. Managers in particular should spend enough time establishing and keeping up effective links to the policymaking world that they begin to feel guilty about not having enough time for their other duties. It is that important.
(1) See "Analysis and Policy: The Kent-Kendall Debate of 1949," Studies in Intelligence, Vol. 36, No. 5 (1992), and The Challenge of Opportunity Analysis, Center for the Study of Intelligence Monograph (July 1992).
(2)Princeton University Press (1949), p.180.
(3) Co-authored with Albert Carnasale. Council on Foreign Relations, 1993.
(4) Mutual and Balanced Force Reduction negotiations between NATO and the Warsaw Pact.
(5) Scowcroft was Special Assistant to the President for National Security Affairs. Robert M. Gates, subsequently Director of Central Intelligence, was then Deputy Special Assistant.
(6) Douglas A. MacEachin, currently CIA's Deputy Director for Intelligence, uses the scout-coach metaphor for analyst-policymaker relations to underscore that it is the scout's responsibility to help the coach prepare to win the game and not to predict the outcome of the game before it is played.
(7) This question was communicated in a letter sent in October 1991 and answered in an interview in November 1991. The DI has been moving in the recommended direction for several years. When the interviewer showed an outline of Blackwill's program to DDI Douglas MacEachin (October 1993), he said, "I guess Bob [Blackwill] and I agree."