MANAGING/TEACHING NEW ANALYSTS

• Collection: FOIA Collection
• Collection: Declassified Articles from Studies in Intelligence: The IC’s Journal for the Intelligence Professional
• Document Number (FOIA) /ESDN (CREST): 0000620589
• Document Page Count: 11
• Document Release Date: July 30, 2014
• Case Number: F-2013-02322
• Publication Date: September 1, 1986

TITLE:AUTHOR:VOLUME:Approved for Release: 2014/07/29 000620589Managing/Teaching New Analysts(b)(3)(c)30 ISSUE: Fall YEAR: 1986Approved for Release: 2014/07/29 000620589t.;proved for Release: 2014/07/29 000620589STUDIES IINTELLIGENCEA collection of articles on the historical, operational, doctrinal, and theoretical aspects of intelligence.All statements of fact, opinion or analysis expressed in Studies in Intelligence are those ofthe authors. They do not necessarily reflect official positions or views of the CentralIntelligence Agency or any other US Government entity, past or present. Nothing in thecontents should be construed as asserting or implying US Government endorsement of anarticle's factual statements and interpretations.Approved for Release: 2014/07/29 0006205890Approved for Release: 2014/07/29 000620589Three missions, four essencesMANAGING/TEACHING NEW ANALYSTS(b)(3)(c)We have all been there, right? The new hire who has only one talent:driving you insane. He misses deadlines. He erases your editing and restores hisstuff in ink, thinking -you can't change it now." He orders one copy of every-thing. The print button on his terminal is taped down and the 2700 is approach-ing meltdown. His in-box looks like the registry. He refuses to take directions,and because all his professors told him he writes well, it must be your fault ifhe is struggling.You're right. Punching this guy is not enough, but what about the new hirewho is obviously trying, seems to have the talent, and is still struggling? Onetheme surfaces repeatedly in the interviews of new analysts at the end of theirfirst year: a degire for a manual describing how to do analysis. I suspect Direc-(b)(3)(c)ument.managers are facing a daunting training burden in addition to reviewingan increased volume of production, providing vital hand-holding services duringthe adjustment period, and passing on tribal lore. Growth has also meant a higherpercentage of new managers struggling with these issues for the first time.After attempting to cope for a while, many managers are ready to conclude -that analysts are born, not made?a position I am not ready to abandon entirely,but one that may be a bit extreme. Like the old jokes, however, the news is goodand bad. The bad is that there is no substitute for experience, no mechanicalformulas that an eager new hire can follow to guarantee an acceptable piece offinished intelligence. Each analyst must learn the job as each before him has,essentially by trying, falling short, and trying again. The odds are improved ifthe manager is blessed at the moment of making a hiring decision with whatMark Twain held to be the most important ingredient in success. -I'd rather belucky than smart,- he said.The good news is that the manager can facilitate the learning process if heor she can:? Communicate a sense of our mission and the difference between intel-ligence writing and academic writing.? Describe the process of intelligence analysis in a clear, cogent fashion.? Prepare the fledgling analysts for early failures and provide lots of positivereinforcement and reassurance.The First StepThe manager's first task is akin to deprogramming?undoing habits formedin four to ten years of college-level work. This comes down to impressing on thewould-be analyst a sense of what the job is and a thorough discussion of theApproved for Release: 2014/07/29 0006205891Approved for Release: 2014/07/29 000620589SECRETTeachingnature of intelligence writing. There are seven key concepts that the new analystmust absorb, three relating to his mission and four to intelligence writing. I amconvinced that the branch chief must hammer (if need be) the "three missionsand the four essences- 1 into the new hire's head, or face the prospect of yetanother journeyman who contributes little but eats up enormous amounts ofmanagerial time.The job is to make judgments about the future. The new analyst oftenhas difficulty accepting the idea we are less concerned about what actuallyhappened than the significance of the event for US interests. Moreover, con-ditioned by college to search for "truth--artistic and scientific?he/she issometimes slow to believe that what people think is true is often more importantthan what is actually the fact. And then there is that dogged determination toget all the facts, a compulsion reinforced by the mistaken notion that our job isto know everything.The manager must get the new analyst to understand that:? Judgments will invariably be on the basis of incomplete and conflictinginformation.2? There are no "incompletes- given here. The analyst never has the luxuryof asking the consumer for a little more time for the situation to clarifyor to wait until additional information becomes available.? Strange as it always sounds, our job is not so much to be right as it isto provide the best answer possible given the time and informationavailable.We are the interpreters of foreign cultures and alien problems. As such,our job is to expose the logic behind the actions of a Middle East madman andto render intelligible to the general reader the physics underlying a Sovietbarrage attack on a US missile system.Our job is to support decision makers. This is a concept that all newanalysts readily accept. Indeed, for many, the prospect of being part of the policyprocess is one of the strongest selling points of the job. But (there is always a but),it is a concept that many new analysts have difficulty putting into practice,because they are confused about what constitutes support.Many believe that if they add to the policymaker's knowledge, they havedone the job. The manager must stress that the point of analysis is the inter-pretation of information, not its presentation. Analysts must be taught to graspthe distinction between providing answers to real problems and expanding theThe Chinese are big on this sort of stuff: the Gang of Four, the Five Ups and the Five Downs and theFour Not Forgets (a spinoff of the Three Non-Divorces), the Three Emptying and Four No Retainings, andmy favorite, the Five Stresses and Four Beauties.2 One new analyst once asked me where "the plans" were filed. My puzzled look prompted him to explainthat he was struggling to reconcile some conflicting information and that he thought he could resolve the issueby looking at the Chinese plans. I told him that only in the movies does iomeone "steal the plans." In amoment of rare good judgment, I resisted adding that if we had the plans we would not need analysts.Reflect for a moment on how many times we actually know if we were right or wrong. Thewho-will-win-the-election or will-country-X-invade-country-Y questions are less common than those dealingwith intentions, options, or outlooks that require continual reassessment.2 SECRETApproved for Release: 2014/07/29 000620589Approved for Release: 2014/07/29 000620589Teaching -SteRLE-T--body of knowledge on some subject. Supporting the policymaker comes downto three related functions:? Providing answers to specific questions, only some of which may be askedby the policymaker.? Providing a framework?an illumination of the forces at work and thefactors that bear on them?that allows the policymaker to understand anissue and to process new information.? Where appropriate, to warn.The four -essences- of intelligence writing flow directly from the threedealing with the mission. The table summarizes the differences between aca-demic and intelligence writing.The Difference Between Academic and Intelligence WritingAcademicFocus on the PASTIntelligenceFocus on the FUTUREWritten for EXPERTS with no responsibility Written for GENERALISTS struggling withto act real problemsDetailed, proof-laden Essentials only, meaningful characterizationsShort on conclusions with a tendency to sum- Begins with conclusions and explores theirmarize implicationsIntelligence writing focuses on the future and it is not enough to marshalthe facts. A new analyst who was struggling to make the transition from academecaptured this problem best. In college, she just gathered all the facts and -theconclusions just fell out.- Many are under the misimpression that the primarygoal of intelligence writing is to discover truth or set the record straight, and asa result, the first instinct is to lay out in loving detail how .the present situationevolved.Managers need to impress on new analysts that because what people believeto be true is often more important than what is, discovery of the facts alone isinsufficient for and occasionally immaterial to the real job of analysis?thinkingabout the future. Students become analysts when they stop thinking in terms ofwhat happened and start thinking in terms of what the facts mean. Many justnever seem to make this transition.Intelligence is written for generalists who are grappling with real prob-lems. One of the hardest things for new analysts to grasp is the nature of theiraudience. Unlike the situation in which they now find themselves, they are usedto writing for professors and academics who demand detail and who are underno obligation to do anything with the information. New people are also slow torealize and often doubt another truth?that after a few months on the job theyare among the most knowledgeable people in the government on a -particularissue, and for the first time in their lives, they are writing for an audience thatSECRCT 3Approved for Release: 2014/07/29 000620589Approved for Release: 2014/07/29 000620589SECRET Teachingknows less than they do. They must be taught that their new audience, unliketheir professors, does not judge the value of a product by its length, devotion todetail, or complexity. Nor is a well-told tale enough. New hires need to learn thatthe value of a paper is proportional to its clarity, brevity, and focus on issues.Intelligence writing is the art of the meaningful characterization. Newanalysts resort to "data dumps" for two basic reasons: they do not know whatis important so they include everything; or they believe piling up detail is thebest way to demonstrate their expertise, a lesson learned in college. The managerhas to impress on the new analyst that the "art of intelligence- is identifying theimportant in the mountain of detail. While reporters -describe the situation,analysts characterize it by making meaningful generalizations that help thereader put events in perspective and think about them. Analysts reconcileconflicting information, isolate the principle in a sea of data, and recognize theexception that demands a reevaluation.Intelligence writing begins with the conclusions and then explores theirimplications. The idea of going beyond the evidence is new for many newanalysts. Academic writing rarely reaches this point; what passes forconclusions." is more often than not a summary of the preceding pages. Incollege good students learn by design or default to focus on how situationsdevelop and fit the evidence into intellectual constructs that are more descriptivethan predictive. Managers must retrain them to think in terms of "this is thesituation; these forces are at work; this is what it means."This is a very difficult transition for many people. The other elements ofintelligence writing can be learned. I am less sure about this one. It seems to goto the core of the thought process. People seem either to have the ability to doit, or they do not. Some are clearly uncomfortable with ambiguity and alwaysseek a little more information before writing. Others draft but cannot movebeyond the evidence or reach intellectual closure on an issue, perhaps becausethey are afraid of being wrong. In any case, the ability to think beyond theevidence and to explore the implications of a situation is the sine qua? non ofintelligence analysis.A Framework for AnalysisIn addition to hammering home what the new analyst is supposed to bedoing and how it differs from what he has done in the past, the manager needsto provide a concise and simple scheme of how to produce analysis. A -how to"diagram accomplishes a number of things. It helps reduce anxiety by giving thenew hire a crutch to lean on. It starts the individual off in the right direction.It reinforces the message of the three missions and the four essences. And, it givesthe manager and the analyst a common vocabulary and a framework forcritiquing fledgling efforts.The production of finished intelligence can be presented as a four-stepprocess:4? Identifying the intelligence issue within the topic.? Identifying the questions that need to be addressed.Approved for Release: 2014/07/29 000620589Approved for Release: 2014/07/29 000620589Teaching -ErEERE'r? After completing the research, identifying the two or three key points thepolicymaker is to take away from the paper.? Drafting, using questions to organize the paper.Step one: identify the intelligence issue within an intelligence topic. Or,deal with the -Great Title Trap.- Ask a new analyst what he is writing aboutand odds are you will get a reply along the lines of (b)( 1 )-defense strategy in the 1990s" or -Soviet activity in the Third World.- Andwhat you get is everything about Soviet activity in the Third World starting withAfghanistan and proceeding through Zimbabwe.Too often the manager must bear the responsibility for the data dump orthe rambling draft that lands on his desk. The sad fact is that most new analystsreally do not know what they are writing about. They are researching a titlegiven to them?a problem compounded by the fact that most titles are con-structed more with an eye to snagging the reader's interest than conveying thesubstance of the paper. I am convinced that more papers go wrong for this reasonthan any other.The solution is to teach the analyst that the first step is to identify specificallywhat it is he is writing about. Introducing the concept of a difference betweenan intelligence topic and an intelligence issue is helpful in this regard. Anintelligence topic is a broad question of interest, such as Soviet activity in theThird World. An intelligence issue is a development or something new anddifferent that narrows the topic and gives a focal point to the paper. There isa simple test: an issue phrase will convey a sense of change or movement oractivity; a topic will not.Examples may help clarify this subtle, but important, distinction.(b)( 1 )(b)(3)(n)The purpose of making this fine distinction is to get the analyst to stop andthink about what he is attempting to do before he attempts to do it. The newhire is not going to be able to make this distinction; the ability to identifyintelligence issues is one of the things that separates the apprentice analyst fromthe adult of the species. It is the manager's responsibility to insure that the analystknows exactly what he is working on.Step two: identify the questions. An intelligence issue is still too broad toprovide the new analyst much help. He needs something to guide him as he readsfiles and gathers information.One answer is to break the intelligence issue into a series of general ques-tions. To do this, the new analyst should be encouraged to step into thepolicymaker's shoes and ask himself: what do I want/need to know about thisissue? The questions should flow from the intelligence issue; if they do not, thepurpose of the paper is probably not clear.(b)( 1 )Approved for Release: 2014/07/29 0006205895Approved for Release: 2014/07/29 000620589SECRET Teaching(b)(1) The first cut at this shouldbe a spontaneous, stream of consciousness exercise. The analyst can then weedand consolidate the list.The list of Questions serves to sharpen the focus of the paper. The newanalyst now knows "what's in" and "what's out," for instance, whether he needsto be concerned about Soviet?North Korean economic relations. The list also tellshim what he should be looking for as he reads files, and once that is done, it helpshim identify intelligence gaps and write requirements. Because the analyst nowknows what information is relevant, it should also speed up the research andprevent the indiscriminate collection of data. 4 The questions may change as theanalyst does the actual research, but this only serves to define the paper moreprecisely.As wiCh identifying the intelligence issue, the manager will have a majorinput in identifying the questions, and the new analyst should be encouragedto touch base with his counterparts at State, Defense, and elsewhere. But, it isimportant that the analyst take the first cut and actually put the questions downon paper. The exercise furthers three goals: it gets the analyst thinking in termsof an audience; it heightens sensitivity to policy relevance; and it gets the analystthinking in terms of something besides what happened.With the questions at hand, the analyst can do the research.Step three: after completing the research, identify the two or three keypoints the policymaker is to take away. This is the most important step in theprocess. At this point, the task for the new analyst is to: (1) digest his research,(2) decide what he knows, and (3) put down in a short paragraph or as bulletsthe two or three key ideas to impress on the reader. Point three is the analyticalbottom line, the essence of the paper and probably the heart of the prospectsor outlook section. If an analyst cannot summarize concisely his bottom line, hehas not done the analysis. If he starts to write before determining his bottomline?the hope that the conclusions will fall out of the facts?he almost certainlywill never have one.It is, of course, a very big step from what you know (2) to what it means(3). How do we get the analyst there? The analyst cannot get there unless he firstdecides what it is he knows. The manager's function, then, is to get the noviceto answer explicitly?if only in his mind?the questions outlined in step two.That done, the analyst is in a position to go beyond the evidence?to thinkabout what the answers mean. The preferred methodology is to use questionsto think the issue through, questions designed to bring out the implications of' I have few illusions about this point. At a minimum, it reduces the excuse for prolonged andindiscriminate research.6 SECRETApproved for Release: 2014/07/29 000620589Approved for Release: 2014/07/29 000620589Teachingthe facts. I just happen to have a set of generic questions with me. Havingdigested the research, the analyst reflects on:? What is new or what is being done differently?? Why is it occurring?? What are the goals and/or broader concerns of the principal actors?? What factors influence success or failure? Are the actors aware of thesefactors? Do they have a strategy/program to deal with the factors?? What are the prospects for success, and more important, what are theimplications for the actors, their broader concerns, the United States, andother countries?? Where do the principal actors go from here?By preaching these questions the manager gets the analyst to focus on the"big picture." The questions cannot?must not?be answered by restating thefacts. The questions get at the processes and call out for generalizations, theessence of good finished intelligence. The key points the analyst wants to impress? I ?? ? ? 'a.. a II ?(b)(1)(b)(3)(n)Step four: draft using questions to organize the paper. The final step shouldbe the easiest. Once the analyst knows the two or three key ideas he wants tomake, the task is to organize the material in a way that makes the points mosteffectively. The best papers are those that are organized into sections that addresswhat policymakers want to know and need to know. The questions used in stepthree often can be used to organize the draft.(b)(3)(c) RET 7Approved for Release: 2014/07/29 000620589Approved for Release: 2014/07/29 000620589Does It Work?TeachingYes and no.6 The four-step process will not make bad analysts adequate.But it does help the learning process:? It provides a common framework and language for managers and newhires.? By helping new people think about the process of writing finished intel-ligence, it improves their ability to master what is an art rather than ascience.? It can be used to explain to analysts why a particular draft is deficient andoffers guidance on how to fix a sick draft.? It gets the new analyst focused on the consumer and US policy questions.? It stresses that intelligence is interpretation of fact, not the recitation offact.The merits of the system outlined here aside, the manager, and especiallythe new manager, needs to develop his/her own plan for training analysts.7 Theart of analysis is, or should be, second nature by the time individuals are tappedfor managerial positions. It may be particularly difficult to communicate tosomeone just how you do it, unless you take the time to reflect on what worksfor you and how best to get those ideas across. A system makes the manager'sjob easier, it standardizes training across the unit, and it allows the manager totest different approaches in a systematic way.Whatever he develops, the manager must do a number of other things tomake the tool useful. It is especially important to discuss the differences betweenacademic writing and intelligence with the new analyst, and lay out for the newhire what the manager looks for in a good piece. This personal philosophy ofintelligence plants a suspicion that there is indeed a method in your madness andthat not all you do is managerial capriciousness. The discussion also serves toestablish the standards you will hold the new analyst to. Mentors are fine, butthe new hire needs to know what the manager thinks.There is no substitute for practice. The more a new hire writes, the soonerhe will master intelligence writing. This has to be coupled with a careful readingof the finished product, but for style and organization rather than substance. Thenew analyst should be given examples of particularly good papers, and themanager should discuss with the novice what makes the paper exceptional.Correctly handling the first paper is also critical. The manager should goover each of his editorial changes with the new hire, explaining clearly why eachwas made. This may be more guidance than the analyst wants at times, but itis an essential part of the teaching process.6 Come on. You know that is as good as it ever gets.The system outlined in this article works for me because it is my style. Perhaps the quickest way to gowrong is to try to apply it without adapting it to your personality.8 SECRETApproved for Release: 2014/07/29 000620589Approved for Release: 2014/07/29 000620589Teaching SECRETFinally, the manager has to dredge up from his own pysche the longsuppressed feelings of frustration and anger that accompanied his own learningprocess. Uniformly, our new hires were very successful in college, doing some-thing that looks very much like what we do. In many cases, you will be the firstperson ever to tell them their work does not measure up. It is a truly traumaticexperience for many when that first draft comes back. Some will not even hearyou at first?a good reason for also putting your comments on paper?when youstart explaining the changes; they are too busy wondering if they are a failureor are about to be fired. In this situation, there is no substitute for positivereinforcement, patience, and a little judicious handholding. Stress what is rightand find a way to say that their experience is typical and not all that differentfrom yours. It is at this point that the framework for analysis can be helpful inpointing out what works, what does not, and why.A little luck helps, too.This article is classified-SEeitET7SECRET 9Approved for Release: 2014/07/29 000620589