Reviewing Intelligence Analysis

Diego Bolchini

Source Link

Sixteen years after publishing the first edition in 2003, Johns Hopkins lecturer Robert Clark has delivered a sixth edition of his influential textbook Intelligence Analysis: A Target-Centric Approach. The book fits in the context of other general works about intelligence analysis. But, as a main theme of the work, the author sheds light on two critical nexuses in intelligence machinery: the collector-analyst and the analyst-customer relationships. Clark argues these different professional segments should be highly collaborative, avoiding blind spots, mistrust, or implicit bias.

A strategic bridge should be built between both groups of people dealing with national security while keeping in mind the differences between them. Former Central Intelligence Agency director, retired General Michael Hayden, once said that “while intelligence officers and the analytic community work to create the general from a sea of particulars (fact-based, inductive), the policy maker (vision-based) trends deductive, trying to apply principles to specific circumstances.”[1] From this perspective, intelligence sets the right and left handrails for any rational policy discussion and high-level decision.[2]

A BRIEF SUMMARY OF CLARK’S WORK

In Part I, Clark describes the process, the participants, and the product of intelligence. In an age of persistent geopolitical and strategic disorder, the author highlights the vital function of intelligence, examining on one hand the key actors involved and, on the other, the constraints and the paradoxes of intelligence: sometimes it discovers secrets in military affairs (e.g., Egypt’s plan to attack Israel in 1973, via precious human intelligence sources), but it does not predict mysteries of macro-politics (e.g., the fall of Soviet Union in 1990 or the 2011 Arab revolutions in Tunisia or Libya).[3]

Part II focuses on the intelligence analysis process. Here, it is evident that posing the right questions regarding an informational gap is critical, focusing the appropriate questions through a proper decomposition of the argument. This is a crucial part of the intelligence cycle.[4] Questions are not self-evident, and, if they are wrong, they can produce false-positive results. A specific chapter in the second part is also dedicated to the issue of denial and deception, which can generate misleading pictures and judgments.

Part III is dedicated to anticipatory analysis or estimates about what will happen in the future with due consideration given to driving forces and scenarios, as well as the so-called unintended consequences that result from actions of individuals. In the study of the future, extrapolation is the easiest approach, as it assumes that the existing forces in the present will not change. On the other hand, projections assume a probability that the forces will change and consider the effects of such changes. Forecasting, according to Clark, is the most difficult predictive technique because it should also identify possible new forces and assess their impact.

BETWEEN EPISTEMOLOGY AND INTELLIGENCE

Clark also adopts a definition of intelligence close to that of former U.S. Deputy Director of National Intelligence for Analysis, Thomas Fingar. Fingar notes that “intelligence is about reducing uncertainty.”[5] Clark also states that a typical goal of intelligence is to establish facts and then to develop precise, reliable, and valid inferences (i.e., hypotheses, estimations, conclusions, or predictions) for use in strategic decision making.[6]

There is always the risk of mismatch between the estimates and what later information reveals. Notable failures in U.S. intelligence machinery include the unexpected fall of the Shah in Iran in 1979 and the 2002 Weapons of Mass Destruction case in Iraq. The latter case was striking because—as Professor Robert Jervis points out—it dealt with capabilities rather than intentions, which are normally considered less difficult to discern with good technical intelligence (e.g., imagery intelligence).[7]

Clark also cites the Iraqi case of Weapons of Mass Destruction in the final chapter of the book, as an example of flawed prediction methodology, contrasting it with the successful 1990 Yugoslavia U.S. national intelligence estimate. The author argues predictive failure in Iraq was due to several causes: a short timeline to produce the estimate; a poor issue definition; and a failure to consider alternative target models and hypotheses, attempting to fit evidence into a single and pre-defined model.

View fullsize

Secretary of State Colin Powell addresses the United Nations Security Council. (Thomas Monaster/NY Daily News)

More generally, Clark identifies at least three levels or dimensions of possible intelligence failures: failure to share information, failure to analyze collected material objectively due to several biases and preconceptions, and failure of the customer to act on intelligence.[8] These three perspectives explain almost all types of intelligence misfires. Clark recalls the case of Operation Barbarossa as a representative example, pointing to Stalin ignoring a mounting body of intelligence indicating the Germans were preparing a surprise attack.[9] Clark also cites British failure, despite available intelligence, to anticipate Argentina’s invasion of the Falklands in 1982, and attribute it to excessive optimism and willful avoidance of an unpleasant choice (as well as the Soviet Union’s decision to support the existing Afghan government) based on flawed intelligence read through the ethno-centric bias of their political doctrine.[10]

THE UNIFYING TARGET APPROACH

Clark proposes a new target-centric approach, which he suggests would ensure more accurate analysis of intelligence. In this model, all stakeholders, including customers, are part of the intelligence cycle. The goal is to construct a shared picture of an informational target. Under a target-centric approach, all participants can contribute in a networked, recursive process rather than a linear, sequential process. The author then focuses on the concept of targets. Specifically, he handles them as complex systems—targets each have a structure; their components and the relationships among them have a function; and the effects produced often represent a process (i.e., a sequence of events or activities).[11]

Clark’s target-centric approach captures his comprehensive view of value in any physical target. An intelligence target may be a selected geographic area, an object, a capability, a person, an organization—including human agents—or an understanding of the context and the behaviour of the population.[12]

Clark also argues competitive target frameworks may exist between policymakers and intelligence analysts. This occurred, for example, in the 1982 case of the Lebanon situation.[13] The U.S. Marines were seen by the policymakers as peacekeepers, while intelligence analysts considered them as potential targets. There were conflicting views of the Lebanon situation, its factions, and political drivers.

CONCLUSIONS

Intelligence Analysis: A Target-Centric Approach has many strengths. The style is concise, and the main concepts are understandable, focusing on important state and non-state actor related issues. Each chapter ends with a summary, critical thinking questions for students and readers, as well as a detailed bibliography and notes section. Possible weaknesses derive, on the other hand, from a low level of granularity offered in the text when it comes to the illustration of some specific techniques such as structured analysis technique (SATs) that could help intelligence professionals to refine their level of analysis.[14]

The book highlights many of the complex issues involved in producing intelligence products. But the intelligence process will never be totally rational, and it requires for its heuristic purposes art in addition to science. When only fragments of information are available, plausible or rational inferences may not be sufficient to solve a puzzle. A certain degree of art is needed for the consideration of un-plausible yet possible, alternatives, the “dogs that do not bark.”[15]

The image of the silent dogs properly connects to our contemporary world, to some extent (or in some measure) made up also of several strategic shocks and black swans, such as the global COVID-19 pandemic. A world where the virus micro-level can impact the macro-level, such as—among other things—the operational status of a nuclear aircraft carrier affecting the three interconnected spheres of military affairs, national security, and (grand) strategy.[16]