INTELLIGENCE PRIMER : HOW TO INFORM POLICY Robert David Steele IMMMMMMMMMMMMMMMMMMMMMMMMMMMMM; : : Mr. Steele has served as both : : a Foreign Service officer and a : : Marine Corps infantry officer. : DRAFT #8A : He returned to the Marine Corps : CLEARED FOR : as a Civil Service officer to : PUBLICATION : stand up the new Marine Corps : BY : Intelligence Center, where he : OASD(PA) : was the founding Special Assis- : 16 MAR 92 : tant and Deputy Director, and : : a Study Director. Now he is a : : planner and resource manager for : : the Director of Intelligence, : : U.S. Marine Corps. HMMMMMMMMMMMMMMMMMMMMMMMMMMMMM< It is the premise of this article that any executive or legislative action to improve our national and defense intelligence capabilties must not only address authority and organization, but also perspective and objectives. Only is this way will we be able to accommodate both the changed nature of the "threat", the changed fiscal environment including an anticipated decline in our intelligence manpower of major proportions, and-- last but certainly not least in its import--the order of magnitude changes in the external (public) information environment. As we consider "intelligence" and its purposes, it is helpful to review some basic definitions, such as developed by Jack Davis, one of the grand masters of analysis. The basic information which follows regarding terminology, the differences between producers and consumers, and the barriers to analysis, owe much to Davis' course on "Intelligence Successes and Failures", and to another course he helped establish, the Harvard Executive Program's "Intelligence Policy Seminar". Elsewhere I have provided critical commentary on the two bills proposing the "National Security Act of 1992". My view in brief is that we do not need legislation--we need instead a better way of integrating analysts and consumers, and a different approach to how we do intelligence. This article attempts to capture what I have learned from others about the barriers to intelligence success in informing policy, to include the competing influences on policy- makers and the fact that they pay no short-term price for ignoring intelligence. The opinions and the recommended remedies are my own. IMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM; : : : INTELLIGENCE ANALYSIS : : : : The process of producing written and oral assessments : : designed to improve the policymaking process by helping : : policy officials better understand and deal more : : effectively with current and prospective national : : security issues, including opportunities as well as : : threats to US interests. : : : : ESTIMATING : : : : The means by which intelligence professionals address : : aspects of national security issues that cannot be : : known with full confidence and thus require conditional : : judgements, interpretation of the evidence, and : : inference. : : : : INTELLIGENCE SUCCESS : : : : Support to the policymaking process that has the : : potential to assist policy officials to avoid or : : mitigate the damage of threats to US interests and to : : enhance the gain from opportunities; that is, : : assessments that are timely, insightful, relevant, : : and attention-demanding. : : : : INTELLIGENCE FAILURE : : : : The inadequate preparation of policymakers for an : : important threat to or opportunity for US interests, : : because of the absence of timely and attention- : : demanding assessments or the presentation of flawed : : assessments. : : : HMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM< Figure 1. Basic Terminology In each of the above definitions, analysis is there to inform the customer, to aid the customer is preventing or defeating threats, and in exploiting opportunities. Emphasis in each definition is my own; my point is to put "current intelligence" in perspective, to highlight the shortfalls of "research" production planned in relative isolation from the customer's decision milestones, and to drive home the fact that if the policy-maker is not reading the product and not talking to the analyst, all the authority and money in the world are not going to alter the practical outcome of restructuring. This first look reflects the experience of generations of analysts as articulated in courses offered at the the Central Intelligence Agency for their own analysts, but including participants from other organizations. IMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM; : : : Intelligence Producer Intelligence Consumer : : : :  Believes sound policy  Believes sound policy : : starts with starts with U.S. : : international realities political realities : : : :  "Expert" on problems,  Political generalist : : focuses on same in all wanting solutions, : : their complexity simple ideas that sell : : : :  Emphasizes foreign  Wants to focus on U.S. : : constraints, what U.S. opportunities, art of : : "cannot" do the do-able : : : :  Gravitates to most  Wants to understand : : likely outcomes good & bad alternatives : : : :  Prefers to be  Wants to know or at : : authoritative and least discuss the : : avoid speculation "unknowlable" : : : :  "Objectivity" first!  Get the job done! : : : :  "We Know What You Need"  "Whose Side Are You On?" : : : HMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM< Figure 2. Producer versus Consumer, Version I In developing our understanding of why intelligence so frequently fails to impact on policy-makers or decision-makers-- even when "tailored" intelligence products are known to reach them--it is essential to recognize the differences between the intelligence professionals (especially the analysts) and intelligence consumers (especially the policy-makers). This is such an important element of intelligence failure, and so vital to understanding the need for the recommendations that conclude this article, that two additional variations of this theme are presented below. I stress this aspect of the problem to make the point that changes in organization, the authority of the DCI, even major changes in how much money we invest in additional collection and information technology capabilities--these are all irrelevant if we cannot change the basic relationship between the analyst and the individual consumer in every Department and at every level. 00000000000000000000000000000000000000000000000000000000000000000 ...changes in organization, the authority of the DCI, even major changes in how much money we invest in additional collection and information technology capabilities--these are all irrelevant if we cannot change the basic relationship between the analyst and the individual consumer in every Department and at every level. 00000000000000000000000000000000000000000000000000000000000000000 Here is the second look, this one with an academic flavor, brought out by Gregory Treverton at a recent "Intelligence Policy Seminar" offered under the auspices of the Intelligence Producers Council by Harvard Executive Programs : IMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM; : : : Analysts Policy-Makers : : : :  Facts/Disengaged  Beliefs/Accountable : :  Objective  Intuitive : :  "Balanced" View  Agenda-Driven : :  Long-Term View  Short-Term View : :  Descriptive  Action-Oriented : :  Employer-Driven  Constituency-Driven : :  Protect Information  Use Information : :  International Focus  Domestic Focus : :  Perfection/Accuracy  "Good Enough"/Utility : :  Written Compendiums  Oral Shorthand : :  Facts/Things  People/Personalities : :  Tenure/Continuity  Short Tours : :  Generic Audience  Tight Upward Audience : :  Single Output  Multiple Inputs : : : HMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM< Figure 3. Producer versus Consumer, Version II As one might surmise from this organized and thorough look at the differences between our producers and our consumers, until we change the way we "train, equip, and organize" intelligence analysts (and other elements of the intelligence community) to "fit in" with our consumers and impact on our consumers, no amount of consolidated authority at the DCI level will be effective in curing our ills. 00000000000000000000000000000000000000000000000000000000000000000 ...until we change the way we "train, equip, and organize" intelligence analysts (and other elements of the intelligence community) to "fit in" with our consumers and impact on our consumers, no amount of consolidated authority at the DCI level will be effective in curing our ills. 00000000000000000000000000000000000000000000000000000000000000000 Here then is the third and final look at differences in terms of the two different groups of people, a look reflecting Department of Defense experience and ably articulated by Sumner Benson to a number of audiences : IMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM; : : :  The analyst is focused on all-source INTERNATIONAL : : DATA while the policy-maker is focused on DOMESTIC : : POLITICAL ISSUES as the primary criteria for : : decision-making; : : : :  The analyst is focused on (and driven by community : : managers) to produce "PERFECT" products over a : : lengthier timeframe while the policy-maker : : requires "GOOD ENOUGH" products immediately-- : : analysts continually run the risk of having : : ZERO IMPACT because their review process delays : : their product to the point that it is overtaken : : by events; : : : :  The analyst is accustomed to INTEGRATING : : all-source information at the CODEWORD level, : : while most policy-maker staffs, and especially : : those actually implementing operational decisions, : : have at best a SECRET clearance. "A secret : : paragraph is better than a codeword page." : : : :  The analyst and community management are focused : : on SUBSTANCE and ACCURACY while the policy maker : : is focused on POLITICS and PROCESS; in the latter : : arena, disagreement can be viewed as : : insubordination. Even if new information is : : received, POLITICAL EQUITIES may weigh against : : policy revision. : : : HMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM< Figure 4. Producer versus Consumer, Version III In short, as we evaluate the intent and utility of the two bills, we should be thinking about what we can do to increase the intellectual and the political "authority" of the analyst in terms of credibility and consumer respect. As Andy Shepard has noted elsewhere, such authority must rest in part on the analyst's direct access to the consumer, and a corresponding familiarity on the part of the analyst with the consumer's day- to-day as well as mid- and long-term concerns. Changing our organization, funding, and the authority of the DCI will not significantly alter this fundamental deficiency in our national intelligence community. Below is a summary of the barriers to analysis. These differences in perspective can also be looked at in generic terms, in the context of institutionalized barriers to intelligence success. IMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM; : : : Signal Barriers Barrier Impact : : : :  International--  International-- : : complexity of world ambiguity of informa- : : affairs; multiple tion; noise; paradigm : : interests & actors; bias; deception; : : national cultural domestic collection : : differences; impact confusion or gaps in : : of U.S. actions understanding : : : :  Policy--misperception  Policy--threat dis- : : of foreign actors tortion; distrust of : : (policy mirroring); analysts; hoarding : : wishful thinking; of information; : : policy momentum manipulation of : : information : : : :  Organization--resource  Organization--mixed : : limitations; emphasis management signals if : : on authoritative not active subversion; : : publications and pre- resistance to alternate : : defined missions and views; information : : roles; fragmentation choke points (both : : of missions, functions, internal and external) : : knowledge, and data : : : :  Analysis--substantive  Analysis--arrogance or : : biases and cognitive overconfidence if not : : traps; parochialism; naivete; tunnel vision; : : monasticism; lack of resistance to outside : : exposure to real world views and priorities : : : HMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM< Figure 5. Barriers to Intelligence Success Such barriers are relatively well understood by students of intelligence, yet they have not been systematically addressed by either legislative charter, or executive organization. It is vital that whatever legislative or executive initiatives are taken in this watershed year of restructuring be founded on a solid understanding of this dimension of the problem. Those responsible for crafting the National Security Act of 1992, or devloping a radically revised Executive Order 12333, must understand that increasing the authority of one person, the DNI, will not mitigate these predominantly cultural circumstances, and may well exacerbate the situation. Each of these barriers has doctrinal, architectural, and technical remedies of one sort or another. In all cases the two key ingredients for improving our chances of intelligence success lie in personal relationships--the relationships between individual analysts and consumers on the one hand, and the relationships between analysts and their immediate managers on the other. Further afield, in the collection management and the individual functional areas of support (imagery, signals, human, and open source intelligence, communications and computing, training, security), equally divisive and counter-productive disparities in perspective between those "doing" and those "receiving" will further enervate the intelligence community. Specific recommendations for improvement are offered in the conclusion to this article. Now, a final illustration needed to establish a foundation for remedial action. IMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM; : : : Politicians : : Executive Leadership : : Legislative Leadership : : Personal & Professional Staffs : : : : : : Government Officials Foreign Officials : : Department Heads P and Organizations : : Assistant Secretaries 0 Diplomats : : Program Managers L Counterparts : : Message Traffic I Correspondence : : C : : Private and Y Independent : : Public Sector 3 Researchers : : Lobbyists M Think Tanks : : Executives A Academics : : Citizen Groups K Authors : : Pollsters E Foundations : : Individuals R Laboratories : : : : : : Media Personal Intelligence : : CNN/C-SPAN Family Community : : Newspapers Intimates CIA : : Wire Services Church NSA/DIA : : Radio/TV Clubs State : : Pool Reporters Alumni Services : : : HMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM< Figure 6. Competing Influences on the Policy-Maker This marvelously simple yet powerful illustration has been explained to generations of analysts and managers without apparent impact on the way we do business. Note that the intelligence community is but one of many competing influences, while also lacking the political influence, economic incentives, or personal appeal such as can be brought to bear by other constituencies. There is no short-term or personal cost to the policy-maker when intelligence is ignored! 00000000000000000000000000000000000000000000000000000000000000000 ...the intelligence community is but one of many competing influences, while also lacking the political influence, economic incentives, or personal appeal such as can be brought to bear by other constituencies. There is no short-term or personal cost to the policy-maker when intelligence is ignored! 00000000000000000000000000000000000000000000000000000000000000000 It also merits comment, for those less familiar with the vagaries of public administration, that no organization is monolithic--each has its own fragmented culture to worry about, and it will not be uncommon for members of one Directorate or Bureau or Division or Service to carry entirely contradictory messages to individual policy makers, all ostensibly from the same organization. In brief then, national and defense intelligence managers are in charge of a vast conglomeration of fragmented resources, created in a piecemeal fashion over time to serve an even vaster array of consumers, most of whom do not really care one way or the other if intelligence is on their table. Only when we fail do we hear the refrain, "where was intelligence?" Such is our foundation. On that basis, below are listed a few modest areas where legislative or executive arrangements may help break down some of the traditional barriers to intelligence success, and improve the ability of our dwindling numbers of analysts to render insightful, relevant, attention-demanding judgements which prepare and encourage policy-makers for their full range of planning, programming, and execution responsibilities. Each of the four groups of ideas, as labeled, corresponds to one of the barriers of analysis outlined earlier in Figure 5. IMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM; : : :  International Signal Barriers : : -- Mandated inter-agency sharing of information : : at multiple-levels of security : : -- Required overseas assignments for most analysts : : -- Radically expanded clandestine human intelligence : : and overt information collection efforts : : : :  Policy Signal Barriers : : -- Annual Congressional review of "threat" in : : relation to each departmental activity, i.e. : : required "state of the world" report as : : precursor to Congressional review of : : President's budget : : -- Full integration of analysts into each : : Department and Country Team policy process : : : :  Organization Signal Barriers : : -- Congressional and executive intelligence : : "Ombudsman" : : -- Increased emphasis on cross-program oversight : : by functional area : : -- Establish "return on investment" program : : evaluation process (not just for weapons : : systems, but for intelligence capabilities) : : : :  Analysis Signal Barriers : : -- Mandated inter-agency training and foreign : : travel for most analysts : : -- Increased exploitation of foreign and : : domestic subject experts in development of : : competing "open source" analyses : : -- Mandated direct consumer contact with analysts : : -- Regular evaluation of analyst and product : : relevance and impact on decision-making to : : include critiques of format, medium, and : : timing of delivery : : : HMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM< Figure 7. Remedial Provisions Who is the customer? What do they need? How do we ensure they get what they need? These are issues which we have not considered as fully as we should in our executive restructuring efforts, and which are also not adequately addressed in the proposed legislation. When you get right down to it, most individuals contemplating improvements to our national and defense intelligence capabilities appear to be thinking about block and wire diagrams and funding authority, when they should be thinking about truly changing the way we do business by substantially expanding the customer base for intelligence, redefining our national security concerns, integrating the individual analyst into the daily lives of their customers, and recapitalizing our infrastructure to take full advantage of the rapidly expanding sources of unclassified information, while also empowering our dwindling work force. IMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM; : : :  Expand the customer base : : : :  Redefine national security : : : :  Integrate analysts and customers : : : :  Recapitalize the infrastructure : : : :  Fully integrate open sources & products : : : HMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM< Figure 8 : Prescription for Intelligence Success If we don't come to grips with these basics, then neither the executive initiatives nor the proposed legislation will make any real difference in our national security or our national competitiveness.