House begins with a primer on the intelligence community and the process of turning raw information into usable intelligence to set the stage for the discussion of the interface between consumers and producers of intelligence. Although this may be redundant for readers with experience in the intelligence community, it is a very clear explanation that would be useful to students and those readers whose ideas about intelligence were formed primarily by popular media. Chapter 2 concisely explains the various sources of intelligence—SIGINT, HUMINT, etc—as well as the terminology used in assessments to indicate the level of confidence analysts have in a particular piece of information. This gets to a critical issue that House explores further in the following chapters: the inherent mismatch between what policy makers often desire and what intelligence analysts can provide. As House notes, “analysis is not simply ‘establishing the facts’ or ‘connecting the dots,’ since in many instances some of those facts or dots are unknown or unknowable.” Despite this, policy makers often desire a level of certainty that cannot be provided and become frustrated with the language used by analysts to maintain objectivity. This can result in situations where, “some decision makers regard qualifying adverbs-including ‘apparently,’ ‘probably,’ and ‘presumably’- as being an effort not to report accurately but rather to avoid responsibility for being wrong.”
Intelligence assessments are usually phrased in terms of relative probabilities, which gives policy makers an ability to choose the assessment that comports most closely with their preferred outcome.
Intelligence analysts may have detailed knowledge, but they must also confront the biases inherent in policy makers whom they serve. Intelligence products are rarely definitive, and there are likely different pieces of information that can support conclusions that confirm policy makers’ desired course of action, even if the analysts’ assessment points in a different, even opposite, direction. Intelligence assessments are usually phrased in terms of relative probabilities, which gives policy makers an ability to choose the assessment that comports most closely with their preferred outcome. This is a long-standing problem in the relationship between analysts and policy makers, but it takes on additional importance as increasingly large amounts of raw information are available to anyone through internet search engines. As House notes, “search algorithms frequently identify the most often cited sources, but ‘most often cited’ is not the same as ‘most accurate.’ The subconscious belief that a Google search can correctly answer any question constitutes yet another barrier to communication between decision makers and analysts.”
Chapter 3 gets to the heart of the matter and here the author’s personal experience and expertise shows most clearly. A fairly detailed assessment of the relationship between the intelligence analysis and policy makers in the lead up to the 2003 Iraq War is particularly illustrative of the issues raised by the author in the first two chapters. Rather than accept the assessment of intelligence analysts, policy makers continually pushed intelligence analysts to go back to find the evidence needed to support the preferred policy. The lead up to the 2003 Iraq War also demonstrates another significant point to which House gives particular attention: the role of intelligence community managers. The policy makers’ expectation of constant briefings by high-level intelligence community managers means that managers spend much more time on the briefing process than running their departments compared to similarly placed officials in other bureaucracies.
A closely related point raised by House is highly salient: the importance of the right personnel in the key management positions in the intelligence community. High level IC officials must understand the intelligence process and have the confidence of the policy making officials. Yet, those same officials must also not be so close to an administration that they become wrapped up in their policy preferences and offer a level of confidence in intelligence assessments that is potentially misleading. As House states, “Intelligence exists to serve policy decisions, but that does not mean the analytical products should bend to the preferences of policy.” Still, in the inherent dynamic between policy maker superiors and intelligence analyst subordinates, there is always a danger that this can happen with potentially disastrous results.
House devotes a considerable amount of attention to the development of intelligence services in Europe and in the U.S. in chapters 4 and 5. It is an interesting history that makes an important point: most European intelligence services developed before the U.S. did so, and mostly for domestic security reasons. The foreign intelligence component was built in because of international connections to domestic threats to the regimes. But while interesting, there is little connection made to the main themes of the book. The same largely goes for the historical section on the US. It is a useful historical overview, but much of it is not directly relevant to the core ideas of the volume. As a result, the volume often seems disjointed, covering a range of topics without linking them together in a completely coherent manner.
The Paradox of Warning: If the intelligence community gets it right and effective action is taken by policy makers, then the predicted event never occurs, and “…the intelligence officer/agency becomes the boy who cried wolf.
House also focuses on the unwelcome surprises that have confronted the intelligence community. From Pearl Harbor through the September 11th terrorist attacks, there is an unfortunately lengthy list of major strikes against the United States that slipped through the web of the intelligence community. House dives into a number of American cases such as Pearl Harbor, the Chinese intervention in the Korean war, and the Cuban missile crisis, as well as notable cases of warning failure in states as divergent as Nazi Germany and modern Israel. We should be careful to avoid characterizing all surprise as forms of intelligence failure, because as the author stresses, our understanding of surprise is based entirely on the things that slip through the net. Chapter 7 is titled, “The Paradox of Warning” for this exact reason. If the intelligence community gets it right and effective action is taken by policy makers, then the predicted event never occurs, and “…the intelligence officer/agency becomes the boy who cried wolf. The very fact that the intelligence system succeeds causes it to look like a failure.”
Despite the author’s extensive treatment of the key issues at hand, there are also some oddly short-changed sections. Politicization of intelligence is tacked on to a chapter on the more modern history of the US intelligence community, but it seems to be an afterthought. Given the alarming intrusion of partisan politics in the intelligence process documented by both the Robert Mueller and John Durham investigations into Russian interference in American elections, it seems to be a topic that would merit more than a page in a work of this nature. Likewise, there is a section on cognitive biases that is helpful but somewhat incomplete given how important these are to the problems addressed in the book. A bit more detail here and more reference to the voluminous literature on the subject would be welcome. Cognitive biases are well known, and the human mind is highly likely to attach more significance to information that confirms a desired course of action and discount information than disconfirms it. House mentions that analysts are trained to minimize the cognitive biases that often distort decision making, but does not go further on this. It would be an interesting point to bring out because there is a possibility that some of that training could be generalized to a broader policy community.
On the whole, however, the flaws in House’s book are minor compared to the value it offers. Intelligence and the State would be a welcome addition to any course on intelligence at an undergraduate or graduate level, as it provides both historical depth and contemporary relevance in a compact format. It will also hopefully be read by policy makers and intelligence professionals seeking to avoid some of the errors of their predecessors.