H-Diplo | ISSF Roundtable 14-5: Amy Zegart, Spies, Lies and Algorithms

christopher ball's picture

H-Diplo | ISSF Roundtable 14-5


Amy B. Zegart. Spies, Lies and Algorithms: The History and Future of American Intelligence, Princeton: Princeton University Press, 2022. ISBN 9780691147130

28 November 2022 | https://issforum.org/to/ir14-5  
Editor: Diane N. Labrosse | Commissioning Editor: Carleigh Cartmell | Production Editor: Christopher Ball


Introduction by Carleigh Cartmell, University of Waterloo. 2

Review by Sarah-Jane Corke, University of New Brunswick. 4

Review by Erik Dahl, Naval Postgraduate School, Monterey, California. 8

Review by Rhodri Jeffreys-Jones, University of Edinburgh. 12

Response by Amy Zegart, Stanford University. 17



Amy Zegart’s book Spies, Lies and Algorithms: The History of Future of American Intelligence, provides a well-written and easy to read overview of the multiplicities of American intelligence; everything from what intelligence is, to intelligence in classrooms and the effects of “spytainment,” and of course, intelligence of the digital age. Zegart describes the initial concept for the book as an “Intelligence 101”; she was forced to adapt this concept given the ever-changing environment of intelligence. Each chapter acts as a standalone, focusing on a different aspect of intelligence. Zegart provides a broad and accessible overview of American intelligence that will most likely come to rest on the bookshelves of many students.

One aspect of Zegart’s argument that all three reviewers remark on is that change is constant, and intelligence agencies must adapt. As such, this work is timely in fitting in to a broad expanse of literature concerned with the largely obfuscated world of intelligence.[1] In her response Zegart writes that “people often think spy agencies are in the secrets business. They’re not. They’re in the insights business.” In that, I believe, they are alike with scholars of intelligence. Intelligence scholars attempt to write coherently and insightfully about intelligence, with many of the pieces missing, obscured, and with additional irrelevant pieces. This roundtable proves that even within the same work intelligence scholars will glean wildly different insights. 

All three reviewers praise the style and accessibility of Zegart’s writing, which is key for an introductory intelligence book, especially in a field that can often trend towards the technical. Rhodri Jeffreys-Jones provided the most critical review of the three reviewers, and calls Zegart’s book “fluently written” but ultimately wishes that the book made a narrower, more focused argument. Eric Dahl argues that the book ultimately provides a “well researched account of the state of American intelligence today” and, despite a few “quibbles,” that it succeeds as a useful book for both students of intelligence and general interest readers. Sarah-Jane Corke provided the most positive of the reviews, arguing that “Zegart’s book clearly does more than simply fill a gap. It provides an excellent summary of the key topics in the field” of Intelligence Studies. Both Dahl and Corke comment that they will definitely—or will most likely—be using the book for future classes. Corke admitting that she has ordered Spies, Lies and Algorithms for one of her graduate seminar classes, and Dahl writes, “I can’t wait to use the book in my classes” but adds the small caveat that he “also can’t help wishing it were a little more.” 

While all three reviewers have varying degrees of praise for the book, noting, for example that it provides some “useful nuggets,” while Corke states, “I am just going to come out and say it. I really liked this book,” they focus on different details for their reviews. This shows just how much Zegart’s book covers for intelligence scholars. Jeffreys-Jones review spends a lot of space on Zegart’s first and tenth chapter. For the former he takes issue with the sometimes “indiscriminate” way that Zegart makes arguments. For the later, he praises Zegart’s chosen examples of media disinformation and her thought-provoking discussion point of data as the “new oil.” Dahl’s review discusses Zegart’s portrayal of the US intelligence community. He critiques Zegart her for perhaps being too sympathetic to the community and wishing that she had covered more of the modern challenges to the intelligence community. Corke focuses the majority of her review on the chapter about the “education crisis” of intelligence studies. She clearly agrees with Zegart’s arguments and relates her own challenges trying to do research within a broken classification system and the challenges of intelligence studies engaging in the wider disciplines of political science and history. Even if the reviewers disagree on the merits and support of the arguments being made within the book, it clearly provides much food for thought. 

In her response Zegart offers a succinct rebuttal to several of the criticisms mentioned in the reviews and a useful discussion of the reviews taken together. With her response she summarizes the roundtable and the reviews as a “Rorschach test that reveals the reviewers’ approaches to the study of intelligence, with rich lessons for scholars and the field more broadly.” It would be difficult to provide a better introduction for the discussion that follows.


Amy Zegart is the Morris Arnold and Nona Jean Cox Senior Fellow at the Hoover Institution, Senior Fellow at the Freeman Spogli Institute for International Studies, and Professor of Political Science by courtesy at Stanford University. Her previous intelligence books have won awards from the American Political Science Association and the National Academy of Public Administration. She writes about American intelligence and national security issues in Foreign Affairs, the Atlantic, and elsewhere.

Carleigh Cartmell is a Ph.D. Candidate at the University of Waterloo and the current Administrator for the Women in Intelligence Network (WIN). Her past work includes the comparing the intelligence sharing of the EU and Five Eyes. Her current work is creating a feminist critical lens for intelligence studies and testing it using several historical case studies.

Sarah-Jane Corke is the co-founder and past president of the North American Society for Intelligence History (NASIH). She has written extensively on early American covert operations. Her first book, titled US Covert Operations and Cold War Strategy: Truman, the CIA and Secret Warfare, was published by Routledge in 2008. Her next monograph, a dual biography of John Paton and Patricia Grady Davies, recently received funding from the Social Sciences and Humanities Research Council of Canada. She is an associate professor of post-1945 American History at the University of New Brunswick where she teaches courses on US Foreign Relations and American Intelligence.

Erik Dahl is an associate professor of national security affairs at the Naval Postgraduate School in Monterey, California, and the author of Intelligence and Surprise Attack: Failure and Success from Pearl Harbor to 9/11 and Beyond (Georgetown University Press, 2013). His latest book is The COVID-19 Intelligence Failure: Why Warning Was Not Enough (forthcoming with Georgetown University Press). 

Rhodri Jeffreys-Jones, is professor emeritus of American history at the University of Edinburgh and honorary chair of the Scottish Association for the Study of America. His latest book is A Question of Standing: The History of the CIA (Oxford: Oxford University Press, 2022).


I am just going to come out and say it. I really liked this book. So much so that I ordered it for my graduate seminar on Intelligence in War and Diplomacy. It is, to my mind, an excellent introduction to the field of Intelligence Studies. It fills a much-needed gap in the literature as many of the books, that I would traditionally have relied on, are either woefully out of date or too expensive to be used in a course where I expect students to purchase several monographs. Here I am thinking of texts like Abram Shulsky’s Silent Warfare: Understanding the World of Intelligence—the last edition of which was published in 2003—or Mark Lowenthal’s Intelligence: From Secrets to Policy—the 2022 paperback version, was recently on sale at Amazon for $106.00.[2]

But Zegart’s book clearly does more than simply fill a gap. It provides an excellent summary of the key topics in the field and includes chapters on the crisis in intelligence education and the rise of what Zegart terms “spytainment,” a basic overview of the history of US espionage from the eighteenth to the twenty-first century—a feat she accomplishes in less than thirty-five pages—an astute introduction to what intelligence actually is and more, importantly what it is not, as well as additional chapters on intelligence analysis, counter-intelligence, covert operations, congressional oversight, the importance of open source intelligence; and of course, all things cyber. It is this last chapter that really captures the essence of the book. Her argument that the United States is woefully unprepared to face the challenges of the twenty-first century is hard to miss. Indeed, her introduction comes off as nothing less than a call to arms, as she argues “American intelligence agencies must either adapt or fail” (12) as changes in technology have dramatically changed the way intelligence is conducted.

Zegart suggests that the combination of technological shifts has transformed the intelligence landscape in three crucial ways. First, it generated new uncertainties and empowered new adversaries (4). Second, it has led to a dramatic increase in information, or data, that has left American intelligence agencies “struggling to keep up.” (5) Third, she believes this has created new problems around the issue of secrecy. She argues the digital age has “blurred nearly all the old boundaries of geopolitics” forcing intelligence agencies to “engage the outside world, not stand apart from it” (8). She opens her book with the perfect example of this: the CIA’s first tweet, which scholars of intelligence history are all too familiar with, the dreaded Glomar response. A Glomar denial is a type of rejection one gets after submitting a request to a government agency. It goes something like this, “The CIA can neither confirm nor deny that there are records matching your request.” In other words, we have them, but you can’t see them, because we can’t talk about them.

Clearly there is a lot I like about the book. Its organization and the topics covered, the stories told, and Zegart’s ability to make big arguments stand at the forefront.  However, it was her third chapter that I found the most compelling. Aptly titled “The Education Crisis” Zegart argues that given the lack of education about intelligence in the United States, and I would argue Canada, fictional accounts are playing a critical role in how Americans, and I would argue Canadians, think about their history. While we are all too familiar with President Ronald Reagan’s penchant for bringing movies into the oval office, Zegart goes beyond this example to show how fiction influences public opinion on issues as important as, torture, assignations, and the general approval or disapproval, of government intelligence agencies.

This no doubt helps explain why the CIA has met with movie makers several times in the run-up to the release of several films. However, as Zegart notes, “depiction is shoddy education” (28). I could not agree more and, ideally, I would prefer it if intelligence officers spent more time meeting with archivists and declassifying documents than meeting, and spending time, with movie producers. Of course, I can certainly understand why they would prefer the latter; there is probably a lot less complaining and I would imagine that the food is better.  But as we all know, the Freedom of Information Act (FOIA) system is broken. According to Zegart, as of 2019 “simple requests to the CIA took an average processing time of 225 days and complex requestions took an average of 530 day” (35). This of course was prior to the pandemic and as a senior archivist warned me last month, the situation has only gotten worse. FOIA requests are now two and a half years behind what they were prior to COVID. No wonder it takes intelligence historians so long to get our work into the public realm.

As is evident from my point above, I recently spent a few weeks at the National Archives going through CIA files to see what had recently been declassified since I was last there. The answer, in short, is very little. Pair this with the US government’s decision to reclassify more than 25,000 documents since 1995 and you begin to get a snapshot of where we are (31). However, when you combine the lack of documents with the problems scholars have had accessing the CIA’s Freedom of Information Electronic Reading Room (formally known as CREST)¾its name changed by the way¾there is not a lot that is positive one can say about the CIA’s ongoing commitment to transparency.[3] One longs for halcyon days when Director of Central Intelligence Robert Gates committed the agency to a policy of Openness. Yes, it may have taken twenty years to see the documents that Gates promised us he was going to declassify in 1990, but at least things were moving in the right direction.

Unfortunately, the perils of writing intelligence scholarship do not stop with issues around declassification. As Zegart notes, publishing intelligence studies in top tier journals has been equally difficult. Although there are three journals dedicated to the subject, the International Journal of Intelligence and Counterintelligence, Intelligence and National Security and the Journal of Intelligence History, the top three journals in Political Science¾The American Political Science Review, American Journal of Political Science, and the Journal of Politics¾only published five articles on intelligence between 2001 and 2016. As Zegart adds, literally, this represented only 0.02 percent of the articles that appeared in the three journals (36).

Given that I am an historian, and a curious one at that, and this review was to appear in an historical forum, I performed a similar, albeit shorter, overview of the top tier historical journals¾The Journal of American History and The American Historical Review—to get a sense of how many intelligence articles appeared over the last five years. I have to say I was somewhat impressed, as I expected the number to be 0.[4] However, I found out that two articles appeared in the last five years that dealt with intelligence history. The first was an intellectual history that touched, tangentially, on the Dreyfus Affair. The second was a more fulsome discussion of Interwar Political Policing in the United States and the United Kingdom. I will leave it to those more qualified than I am to figure out the actual statistical average. I avoid math at all costs.

Sadly, even the field of American Foreign Relations has seen a marked decrease in interest in Intelligence History. When I looked at last month’s Society of Historians of American Foreign Relations, (SHAFR) conference, which used to regularly feature panels on intelligence, I found only one paper presentation. Now the lack of participation of intelligence scholars in SHAFR may well be because of the recent formation of the North American Society of Intelligence History (NASIH), which was developed in 2016 and has since sponsored two scholarly conferences on intelligence history.[5] However, while there is no doubt that NASIH has filled a much-needed void, it remains critical that intelligence scholars continue to have their work appear outside intelligence-only venues.

No doubt part of the problem lies in the decline of scholars working in the area of intelligence at American and Canadian Universities. Both the United Kingdom and Australia seem to have a better track record, but in North America there are very few professors teaching courses on intelligence. In part, this is due to the dramatic cuts in our university programs that have occurred over the last decade. However, this has been combined with a general disinterest, and sometimes outright hostility, to the teaching of subjects like foreign policy, military history, and intelligence; a problem that extends back farther in time, at many smaller universities and colleges.

Yet, our need to understand the world today still hinges on our ability to critically think about these historical topics. And as Zegart concludes, when we fail to provide a scholarly examination of these topics, “fiction too often substitutes for fact, creating fertile ground for conspiracy theories to grow and influence the formulation of real intelligence policy” (43). Nowhere is this made more evident than in the book’s final chapter, on cyber. Herein she explores how and why cyber threats evolved, what they look like today, and what they mean for intelligence practice in the future.  She concludes that cyber is “a different kind of battleground” one that is not only “manmade” but “inherently insecure.” (255).

More importantly, perhaps, it is both a foreign and domestic threat with both foreign and domestic implications. Unfortunately, the fact that it lies at the nexus of the internal and the external complicates who can target these new threats. The US elections of 2016, and 2020, in different ways, illustrate the problems that intelligence agencies face when trying to stop these types of operations. In 2016 the United States faced a Soviet disinformation campaign. In 2020 American voters found themselves caught up in a domestic disinformation campaign spearheaded by the “twice impeached, disgraced former American President.”[6] This of course was problematic, as American intelligence agencies were set up to respect the division between the foreign and the domestic, with the CIA preoccupied with the former and the FBI the latter. Although 9/11 forced more cooperation between the two, formally hostile, bureaucracies, as Zegart makes clear, there is still a lot of work to be done.

I have been an intelligence historian for over twenty years and read hundreds of books on the subject. Rarely, however, have I come across a book that I liked right out of the gate and that held my interest throughout. Zegart’s accomplishment should not be surprising, however. As she notes in her acknowledgements, the project is the result “of nearly thirty years of research on “the evaluation, operation, and challenges” faced by American intelligence agencies (277). Her work, combined with her experience as a former policy advisor who has interviewed numerous intelligence officers and has taught several courses on intelligence at both UCLA and Stanford, places her at the forefront of the field.

Now I could, of course, stop the review here, and leave celebrating only Zegart’s accomplishments, foremost of which I would suggest is her ability to make big arguments, in a big way. But big arguments can sometimes have little problems.[7] And I would be remiss, if I did not at least point out a couple of places where I would have liked the book to have engaged a bit more energetically with the new historical work that has appeared over the last decade. Of course, the first has to do with my own area of research. I was somewhat disappointed to see that Zegart embraces the traditional, albeit outdated view of George Kennan’s role in the development of American post World War Two covert operations (185). As several scholars have shown, myself included, Kennan played the pivotal role in the development of the CIA’s Office of Policy Coordination.[8] Although he later tried to distance himself from the development of these operations, his fingerprints are all over them. This is a small matter certainly, and of interest to only few, but to my mind understanding the origins of these operations help us understand how they unfolded over time.

I would also add that I found the book’s discussion of the Cuban Missile Crisis to be overly traditional in nature. For example, in chapter five on intelligence analysis, Zegart discounts any defensive rational Soviet leader Nikita Khrushchev may have had when he installed nuclear weapons in Cuba. Instead, she argues that none of the intelligence assessments considered the benefits of the decisions, which she believes were leverage over the situation in Berlin, embarrassing President John F. Kennedy, or leveling the nuclear playing field (126). Yet, as several revisionist and post-revisionist historians of the crisis have illustrated, Khrushchev’s decision was largely defensive in nature.[9]

Despite these two examples, where I would have liked a more fulsome acknowledgement of the shifts in the historiography, Zegart certainly done her part to highlight the importance of intelligence history to a wide audience. The book will no doubt be used over the next decade in countless university classes, including mine, ensuring the new generation of scholars are introduced to the field in a thoughtful and scholarly way.


This book offers a very readable and well researched account of the state of American intelligence today.  Amy Zegart’s engaging writing style makes the book often read more like a novel than a typical textbook or academic work about intelligence, and I expect it will be a great source of information for intelligence students and for general readers.

The theme of the book is captured in chapter 1, where Zegart writes that “Never before has the United States faced a more dynamic and dangerous world” (2). The part of that dangerous world that Zegart is most interested in is technological change, and she describes how American intelligence agencies are being challenged today by new technologies such as artificial intelligence, quantum computing, and synthetic biology. In response, she writes, “America’s intelligence agencies must adapt or they will fail” (12).  

Having read Zegart’s previous work, I find a lot that is familiar here. For instance, Zegart made the same argument about the need for intelligence agencies to adapt in her best-selling book, Spying Blind.[10] The title of the new book echoes that of an article she co-wrote with former CIA deputy director Michael Morell a few years ago in Foreign Affairs.[11] And some sections and chapters are clearly drawn on earlier work; chapter 2, for example, which describes how little most Americans know about intelligence, appears to be an updated version of a 2010 article on what she called “Spytainment.”[12]     

The book provides a broad introduction to the American intelligence community, and, in that sense, it resembles textbooks from authors including Mark Lowenthal,[13] Roger Z. George,[14] and Loch K. Johnson.[15] And much like those other authors, Zegart’s account of American intelligence is a sympathetic one. For example, in a set of vignettes about “Who works in intelligence and why?” she paints a picture of honest, patriotic Americans who decided to work in intelligence. And later she writes that “working in an intelligence agency is not a job. It’s a mission” (157). Even granting that the book attempts to explain the culture found within the American intelligence community, the statement is less than objective.[16] 

Elsewhere the book goes easy on the intelligence community, whether through treating the NSA gently for its collecting of Americans’ telephone calls after 9/11, or in Zegart’s description of the controversy that arose after a number of former senior intelligence officers went public with their criticisms of then president Trump. Some experts have argued that this public criticism represents a dangerous politicization of intelligence,[17] but Zegart writes simply that “nobody knows what the long-term effects of this hyper-political moment will be for U.S. intelligence agencies” (90).      

I have three sets of comments about the book. First, a few quibbles. Zegart argues that part of the reason why Americans do not understand intelligence is that “few professors study or teach intelligence” (35). She notes that relatively few articles are published in major political science journals about intelligence, and that few courses on intelligence are taught in major universities. But although this argument was convincing when she made it 15 years ago in Spying Blind—she then called intelligence an “academic no-man’s land”—I think the situation is changing today more than this new book acknowledges. There are several respected intelligence studies journals such as Intelligence and National Security and the International Journal of Intelligence and Counterintelligence (both of which have published her work); intelligence studies scholars are active in organizations such as the International Studies Association and the International Association for Intelligence Education;[18] and the number of college and university intelligence programs is growing, in many cases sponsored by the US intelligence community.[19]  

Zegart provides a good discussion of the controversies surrounding the use of covert action, but at one point she writes that when the CIA was created it was not supposed to “engage in sponsoring coups, influencing foreign elections, or conducting other covert operations” (63-64). This is a point of debate among historians of intelligence, and a number of authorities argue that covert actions were in fact always intended to be part of CIA’s mission, but without being publicly mentioned.[20] This may be in part a problem of editing; the book later includes a more nuanced discussion about covert action, noting that at its founding, the CIA was not specifically authorized to conduct covert actions (172)—a statement that is correct, but also unenlightening. 

Another quibble might be due to a lack of understanding on my part. In Chapter 5 Zegart discusses the many cognitive biases that make intelligence analysis challenging, and she illustrates these problems with what she calls the “Beauty Pill exercise” that she used to conduct with her students (127). This is the familiar kind of problem where subjects are given two choices, which are expressed differently but which actually amount to the same thing—and presto, we find that humans do a bad job of crunching numbers and assessing risk. But unless I’m having trouble assessing risk myself—which is certainly possible—I think the exercise is flawed. 

Zegart writes that she tells the students to imagine a beauty pill that “was rigorously tested and found to be 99.9 percent safe, with no side effects,” and almost all of them say they would take it. But then she restates the situation, telling them that this time “there’s a 1 in 1,000 chance that if you take this beauty pill, you will drop dead, right here, right now.” Not surprisingly, few now say they would take it. For Zegart, this demonstrates how presumably intelligent individuals are unable to understand risk. The point, she explains, is that “statistically speaking, 99.9 percent safe is exactly the same as a 1-in-1,000 risk of death.” But I am not sure about that, because to my mind being unsafe is not necessarily the same as being dead. In the first case, if the pill turned out to be unsafe, couldn’t that just mean it would make you sick, or result in other side effects?

My second comment is that the book appears to try to do too much, covering the waterfront of intelligence issues, and as a result it reads like a cross between a textbook and an edited collection without quite succeeding as either. There are gaps that make it less useful as a textbook, such as the chapter on intelligence oversight that focuses solely on Congressional oversight and neglects other important sources such as those from within the executive branch itself, or from the judicial branch.[21] Another example is the book’s cursory description of the intelligence disciplines, or “ints,” such as signals intelligence and human intelligence. This is enough for the casual reader, but a student looking to learn deeply about the intelligence community would be better off with one of the standard textbooks,[22] which not only provide significantly more detail, but also cover less well-understood aspects of intelligence such as what is known as measurement and signatures intelligence (MASINT), which Zegart’s book does not discuss. 

A firmer editorial hand might also have helped to reduce the overlap between chapters. For example, anecdotes often appear more than once, for example the twice-told one about a disinformation video that was produced in which House Speaker Nancy Pelosi was made to sound drunk.

Third, I would have preferred a more ambitious, more cohesive book that extends Zegart’s argument about the challenges facing the intelligence community throughout all the chapters. More coverage of how intelligence can be used to address nontraditional threats, such as COVID-19 and climate change, would also have been useful. Most important, it would have been useful if Zegart had included advice for the future. Even though the subtitle of the book is “The History and Future of American Intelligence,” there is little about the future to be found here. 

This book is clearly of a piece with Zegart’s previous work, which is pessimistic about the prospect for significant change and improvement in American intelligence. In Spying Blind Zegart argued that adaptation in the intelligence community “is not impossible, but it is close.”[23] Today it would be nice to know whether we are getting any closer—or at least, to hear recommendations for how the intelligence community could get a better handle on the many challenges it faces. Zegart has elsewhere provided thoughtful advice for American intelligence that could have been developed in this book. Her co-authored article argued for the creation of a new open-source intelligence agency,[24] and included other recommendations such as changes in intelligence community hiring policies. An updated examination of these ideas would be welcome. But in this book Zegart writes simply that “Many believe that publicly available information, or open-source intelligence (OSINT), has become so important in the Internet age that it needs its own agency” (82).    

Finally, the book could use a concluding chapter—as it is, the final chapter is about the increasingly important area of cyber threats, which is a good topic to end on, but adds to the feeling that this book is a collection of very well written, standalone chapters, rather than a cohesive monograph.

These comments are not meant to detract from the value of the book. Even though much of this material is familiar, Zegart has written the most up to date account available of American intelligence today. And her writing makes the book a pleasure to read, with frequent quips and quotable comments such as “Congress has many more powdered milk experts than intelligence experts” (216). I can’t wait to use the book in my classes, but I also can’t help wishing it were a little more.


Spies, Lies, and Algorithms makes only a partial recovery from an opening chapter titled “Intelligence Challenges in the Digital Age: Cloaks, Daggers, and Tweets.” Here Amy Zegart argues that America is facing a security crisis and that the public is aware neither of its gravity nor of the essential role that intelligence must play on order to keep the nation safe. The book’s mission is better to inform Americans about these crucial issues. A case might indeed be made for improving understanding of and support for the intelligence analysts who rarely make it into the headlines and need to be bolstered for two reasons: to command the respect of policymakers and to resist those policymakers’ demands for intelligence to please. Zegart’s argument is, however, indiscriminate. Her book echoes without making distinctions the laments of a variety of intelligence actors and their supporters over the decades that they are the dimension that is missing from the history books and that they are under-appreciated.

There is a nostalgic populism in some of Zegart’s remarks. For example, in demonstrating the threats posed in our digital age, Zegart cites a study that predicts information technology (IT) will displace 25 percent of the workforce over the next 15-25 years. The argument would be credible if it were accompanied by an explanation of why the Luddites got it wrong, whereas this source is correct. She further maintains that IT advances mean we are living in an age of fastest-ever change but offers no measurement of this. One generation after another, after all, has complained about the speed of change. The argument that the Cold War era was a “simpler” time (4) is not supported.  That period did not seem simple to the more thoughtful of the Cold Warriors.

Zegart argues that there are “bad actors” who leverage technology in such a way as to create problems for America “with a click of a mouse” (4). She warns that “data is democratizing”—advances in digital technology have levelled the playing field, challenging US hegemony. Later in the book, though, she asserts that in Stuxnet, a system operated jointly with Israel, the United States has the “most sophisticated cyber weapon in the world (264).” Stuxnet she explains, was used to disrupt Iran’s production of centrifuges. This was alright because, according to her tacit assumption, the United States is self-evidently not a “bad actor”.

Zegart holds that the privatization of national security is an inevitable and acceptable trend. She states that the world is now governed by Amazon and its mega-siblings, and that it would be uncompetitive to try to combat foreign security threats by old-fashioned public means. It is, however, open to question whether it is in the interest of American taxpayers or efficient in any way whatsoever to have the CIA’s cafeteria in Langley at the disposal of corporation head-hunters who hire the Agency’s best, double their salaries, and subcontract their new recruits to the nation at triple the original cost.[25] With China in mind, Zegart notes that private intelligence ventures in countries “with different values” pose a problem and that there is a need “to establish standards” (250), thus expressing the improbable expectation that the United States, having allowed initiative to pass beyond government, may be able to persuade other nations to exercise tighter and beneficial control. She notes that in the past the direction of IT developmental travel was from government to the private sector, and that this has now reversed in the United States. The book lacks a discussion of the desirability of the reversal, or of the fact that the change happened, in large measure, for reasons of ideology, not effectiveness or the needs of national security.

Zegart effectively makes a plea in the name of US patriotism for the principle that private firms will best defend America. In fact, many of the larger corporations are multinational and, like monarchs who service their own dynasties, not necessarily patriotic at all. Kathryn Olmsted recently wrote a book called The Newspaper Axis: Six Press Barons Who Enabled Hitler. It is a study of American, British and Canadian media moguls whose support for Hitler and antisemitic sentiments reached a readership 66 million and weakened resistance to the German dictator. That resistance, had it been timely, could have saved mankind from one of its greatest tragedies. This is not to say that there should be governmental control of the media, but Olmsted has demonstrated that “fake news” is not new, and that private media ownership does not reliably serve the best interests and ideals of Americans.[26]

What weakens Zegart’s first chapter particularly is an assumption that affords too high a priority to the CIA and its sibling agencies. Zegart argues that the CIA “beats the best alternatives” (4). The CIA does indeed have its strong points and serves an essential need. But Zegart does not highlight that US foreign policy and security making owes much to non-secret branches of policymaking, such as the Departments of State and of Commerce, as well as to advice from the medical, scientific, and other professions which are international by nature.

Moreover, intelligence has had a recruitment problem. The State Department for decades tended to recruit candidates with foreign experience and summa cum laude degrees, the CIA less so, and there is little reason to suppose that things have changed.[27] The book venerates secrecy in a way that calls into question its objectivity. Zegart writes, “While open-source information has always been important, secrets have reigned supreme in America’s intelligence agencies (10).” This elevation of secrecy over quality has helped to give the intelligence community and its more strident champions a bad name.

Chapter 1 is not, then, persuasive. Subsequent chapters are, by and large, not much better. Chapter 3 is titled “American Intelligence History at a Glance” and, as that wording suggests, the constriction of space imposes a limit on what can be said. In an erroneous expression of reverence for UK intelligence, Zegart here writes, “When British codebreakers at Bletchley Park cracked [Adolf] Hitler’s ciphers in World War II, it was the third time in 250 years that British codebreaking prevented a foreign invasion (46).”[28] Zegart later praises General George Washington’s “talent for deception (50).” In the same chapter, Zegart’s discussion of the origins of the FBI ignores the political problems that beset the Theodore Roosevelt administration—in spite of the recent appearance of an additional scholarly study of the subject.[29]

Chapter 3 outlines the confusion imposed on the US intelligence community by the Intelligence Reform and Terrorism Prevention Act of 2004. The act created a new Director of National Intelligence who was independent of the CIA. As Zegart observes, this new official had big responsibilities that were not backed up by a commensurate degree of authority. A pre- and post-reform comparison would have strengthened the chapter. Did the 2007 Iran estimate that stopped in its tracks a dangerous tendency in the administration of George W. Bush indicate that things had improved since the disastrous failures over 9/11 and the detection of non-existent weapons of mass destruction in Iraq? The book offers no guidelines for thinking this problem through.

Chapter 7 is titled “Covert Action” and takes as its subtitle a quotation attributed to Leon Panetta, director of the CIA, 2009-2011 and later secretary of defence, 2011-2013: “A Hard Business of Agonizing Choices.” Zegart offers some points for consideration. She suggests that while covert action is risky, it can also succeed. What is not discussed is that a nation that resorts to covert action can experience failure as an intrinsic part of apparent success. What happens to soft power when it is universally accepted, in foreign countries, that a particular power is prone to overthrowing democratically elected governments? Some intelligence officials and their acolytes inside the newly constructed Beltway hailed the Iran and Guatemala debacles of the 1950s as brilliant successes, but knowledge of such operations could not be kept from the wider world and the United States lost its majority in the United Nations General Assembly. This is surely the kind of framework needed for the discussion of the merits and demerits of covert action.

Zegart points out in this chapter that liberal, Democratic presidents who were prone to criticising intelligence practices when in opposition have nevertheless resorted to covert action when in office. There can be no quarrel with the examples she gives, President Jimmy Carter’s abortive attempt to rescue the American hostages held in Iran, and President Barack Obama’s dispatch of al-Qaeda leader Osama bin Laden. Republicans tend to offer these arguments as rejoinders to Democratic assertions that Republican presidents wage more wars than their Democratic equivalents. The argument can be taken further: are liberal presidents so fuelled by morality, or so moralistic to put it less kindly, that they will resort to war in preference to being thwarted? Data on the subject suggests, rather, that Republicans are more warlike—but not by a significant margin. Zegart has given credible new impetus to a thought-provoking issue.[30]

Spies, Lies, and Algorithms contains some useful nuggets in chapters that roam across diverse issues. There is a pungent review of the utility (or not) of torture in the securing of intelligence, such as that which led to the discovery of the whereabouts of Osama bin Laden, even if Zegart errs in saying that the killing of the terrorist leader—as distinct from his discovery—was an intelligence triumph. She is surely right to warn that the nuclear threat is in danger of being overlooked by the public but remains “important to intelligence officials and policymakers” (227). Recent sabre-rattling by Russia’s Vladimir Putin may challenge the public’s complacency.

Although the book is meant to educate the American public, it does not provide a reliable guide to the literature. In some ways, the arguments are careless. One would like to know on what basis James Fenimore Cooper’s The Spy (1821) can be considered “America’s first ever bestselling novel” (24). Zegart cites a detail supplied by the historian Frank Smist in support of her contention that Congress failed in its intelligence oversight duties in the 1950s (201, 351 n.40), but she does not include the context, the fact that Smist sees intelligence oversight in that decade as having been secure, appropriate and adequate.[31] Zegart references the political scientist and veteran intelligence-committee staffer Loch Johnson, but does not discuss the arguments of critics who differ from Johnson’s view that the intelligence reforms of the 1970s were a significant turning point in oversight history.[32] Even more, a book that deplores the American public’s ignorance of intelligence matters does not include a classic hypothesis about the ebb and flow of oversight, the Ransom thesis that postulates an inverse relationship between international tension and the aggressiveness of congressional intelligence inquiry.[33]

Chapter 10, “Decoding Cyber Threats,” explores the Kremlin’s cyberattacks and media disinformation but does not put this process into context. There is no discussion of the fact that RT, formerly Russia Today, had ample precedents in US foreign policy, for example CIA-sponsored radio, TV and magazine outlets. Fake News is not solely an import but is also a home-bred product. Nor does one find an answer to the question posed in one of the subheadings, “What does Intelligence have to do with Cyber (269)?”

Chapter 10 does have its merits. Zegart has usefully collated examples of media disinformation, whether or not Russian in origin. She graphically details the clash in Houston on 21 May 2016 between a Confederate-conscious group called the Heart of Texas and supporters of a more peaceful organisation called United Muslims of America. Both of the contending factions derived from Facebook groups engineered by the Internet Research Agency (IRA): “Inside nondescript offices in St Petersburg, Russia, hundreds of trolls masqueraded as Americans in around-the-clock shifts—tweeting, liking, friending, and sharing in English to attract American followers” (251). Another example she gives is the voice-distortion applied to remarks made by House of Representatives Speaker Nancy Pelosi making her sound drunk—a scam that hit the headlines and was viewed more than two million times on social media.

In Chapter 10, Zegart introduces a visionary discussion point when she asks whether data are “‘the new oil—a precious resource that fuels geopolitical competition (255).” She thoughtfully suggests that “Oil is valuable because it is scarce; data is valuable because it isn’t (255),” meaning that it is widely available and is challenging the distribution of global power. The flaws in the book under review should not obscure that fact that it is fluently written and contains some good ideas.


I am deeply grateful to Sarah-Jane Corke, Erik Dahl, and Rhodri Jeffreys-Jones for taking the time to review my book. I’d also like to thank Carleigh Cartmell for organizing this forum and H-Diplo for hosting it.

For understandable reasons, the study of American intelligence has always been highly charged. Agencies are powerful and secret. At times they have broken the law, violated American civil liberties, and betrayed the public trust. At other times, they have left the nation vulnerable to horrific surprises like Pearl Harbor and 9/11. The stakes of intelligence are high. When the Internal Revenue Service (IRS) fails, people over-pay their taxes. When the CIA fails, people can get killed. And as intelligence scholars know well, the full story is almost never really known—even by participants. Records are classified, memories are selective, and individuals and organizations reveal and conceal to serve their own interests. Of course, these kinds of challenges affect the study of American foreign policy writ large. But intelligence is especially tricky because it elicits powerful feelings of attraction and revulsion. Espionage is seen as glamorous and awful, heroic and hideous. It’s cool and yet often heated. Everyone seems to feel strongly about American spy agencies even if they don’t know much about them.

I originally set out to write an “Intelligence 101” book after my national polls found that Americans who heavily consumed spy-themed entertainment were statistically more likely to approve of waterboarding and other harsh interrogation techniques. But I took so long to write the book that the world changed, sweeping US intelligence agencies with it. When I moved to Stanford University, I saw up close and personal just how much emerging technologies—from AI to commercial satellites—were transforming industries, including intelligence. People often think spy agencies are in the secrets business. They’re not. They’re in the insights business. And that business is in the throes of radical disruption. With commercial satellites tracking Russian troop movements in Ukraine,[34] online crowdsourcing tips helping the FBI identify January 6th insurrectionists,[35] and nongovernmental nuclear sleuths discovering vast Chinese missile fields,[36] it is fair to say that intelligence is not just for governments anymore. So my intelligence 101 project evolved into intelligence 2.0—a book that seeks to examine the past as well as the future of American espionage in the digital age.

The three reviewers offer wildly different assessments. Corke begins, “I am just going to come out and say it. I really liked this book” and notes that of the hundreds of intelligence books she’s read, rarely has she felt that way. Dahl writes that the book builds on much of my earlier research to provide a broad introduction to intelligence that is  “very readable and well-researched,” though he has quibbles with some of the arguments and thinks I tried to do both too much and not enough (more on that below). Jeffreys-Jones really dislikes the first chapter and believes the book goes downhill from there—accusing me, among other things, of writing a book about US intelligence agencies that pays too little attention to all the agencies that aren’t involved in US intelligence.

Taken together, the reviews provide more than a book review. They offer a fascinating Rorschach test that reveals the reviewers’ approaches to the study of intelligence, with rich lessons for scholars and the field more broadly.

I admit I thoroughly enjoyed Sarah-Jane Corke’s review, and not just because she thoroughly enjoyed my book. Corke is a leading intelligence historian who teaches at the University of New Brunswick and co-founded the North American Society for Intelligence History. She zeroes in the education crisis chapter, which I believe is one of the most important contributions because it studies the study of intelligence—and why researching and teaching intelligence are so essential for public policy. A fellow traveler in the field and its frustrations, Corke notes how challenging it is to conduct research on the CIA even when the CIA thinks it is being helpful, and how little attention her history colleagues pay to the study of intelligence. I have written about how political science journals have largely ignored intelligence. Corke did some clever digging and found the same pattern in history—noting that only two articles related to intelligence have appeared in the past five years in The Journal of American History and The American Historical Review. Her curiosity piqued my curiosity. She didn’t tally the total number of articles published—wishing “to avoid math at all costs” as she notes—so I picked up the trail. The total turns out to be 178 articles, which means that intelligence topics constituted just 2 percent of the material published in these two top history journals. But wait, there is more. As historians know better than I do, these journals publish a great deal of other content—notably book reviews, as well as other features like AHR Forum, AHR Conversation, AHR Reappraisal, and AHR Reflections. Intelligence got even less attention there. To give just one example, I examined the last nine issues of book reviews in The Journal of American History. They numbered more than 1,000. I discovered a surprising number of books about animals and cigarettes. But only eleven reviews, or about 1 percent of the total, examined intelligence books.

To be sure, there is new and important historical research about intelligence. Corke helpfully suggests that I could have paid greater attention to more recent work about George Kennan’s role in covert operations and revisionist perspectives about intelligence in the Cuban Missile Crisis. But the overall pattern is disconcerting and unmistakable: American intelligence agencies play a central role in US foreign policy and international relations, yet they remain a sideshow in the most prestigious history and political science outlets.

Erik Dahl brings two decades of experience as a naval intelligence officer and a subsequent scholarly career to his review. His close reading raises several important points and suggestions for which I’m grateful. First, he suggests that sometimes I seem sympathetic to intelligence officials and agencies. I hope so. As he knows, nearly all of my earlier research has been explicitly and deeply critical. (So critical, in fact, that I had to hire a First Amendment lawyer when I was writing my book dissecting the CIA and FBI failures leading to 9/11).[37] In Spies, Lies, and Algorithms, my goal was to provide both an outsider’s critical analytic perspective as well as an insider’s personal view of working in the intelligence world. I want readers to understand the strongest criticisms as well as the most forceful defenses of intelligence agencies, to see both the organizational challenges and the human dimensions. That’s why I include interviews with several leading intelligence officers who reflect on the ethical dilemmas they faced, their best and worst days, and other personal experiences. My primary aim is not to tell readers what I think. It is to help them figure out what they think. Especially in today’s political climate, nuance, empathy, and genuine debate are underrated.

Second, Dahl suggests my beauty pill exercise may stack the deck, since 99.9 percent safe may not really be seen as a 1-in-1,000 risk of dying. Good point; the next time I conduct the exercise, I’ll try different wording and see if the outcome changes. The main takeaway, however, is that risk is largely a matter of perception. The same underlying risk can be interpreted differently depending on how it’s presented.

Third, perhaps Dahl’s biggest criticism is that the book takes on too much and yet does not cover enough. This the classic breath/depth conundrum. He notes that I offer a cursory overview of intelligence disciplines (the “INTs”) and provide no policy prescriptions for the future. Dahl considers these omissions a weakness, while I see them as a strength. Yes, readers looking for draft legislation of an intelligence reform bill that will never be enacted will be disappointed. Those who want to know the intricacies of MASINT (Measurement and Signature Intelligence, which identifies objects by their detectable characteristics like temperature or vibration) or the fine points of intelligence budgeting should also look elsewhere. My aim was to cover the most important intelligence issues where public misunderstanding abounds, not cover wonky intelligence details in all their glory.

Finally, Dahl notes that the book reads more like individual chapters than a coherent whole. I agree. I designed the chapters to be standalone units so that professors could assign parts of the book for an array of courses in US foreign policy or American politics without requiring students to buy the whole thing. The tradeoff may not be worth it, but I hope it was.

Rhodri Jeffreys-Jones is an historian of American intelligence in the UK who brings a strong normative bent and focuses on the role of individual people and events. He’s interested, as he puts it, in the “ebb and flow” of history. That traditional perspective can be valuable, though it’s often not particularly scientific, and in this case it’s unartfully conveyed. Let me put aside the unfortunate and distracting tone of the review so that readers of this roundtable can glean the important conceptual differences in our approaches.

Jeffreys-Jones’s review values trees. My book values forests. He focuses on the backgrounds of CIA and State Department employees based on data from 1984 and the number of Pashto speakers back in 2002. I focus on large questions and examine broad patterns that illuminate the past and affect the future—such as why all presidents, even liberal democrats, use covert action; why congressional oversight has always been weak; and why intelligence agencies have such a hard time adapting to new threats. What he criticizes as an “indiscriminate” narrative is what Corke calls “making big arguments in a big way.” It is clear why I appreciate Corke’s review so much more.

One can see our differing perspectives in the discussion of congressional intelligence oversight. Jeffreys-Jones advances the tired conventional wisdom—he calls it the “classic hypothesis”—which asserts that congressional intelligence oversight should be characterized by its ups and downs over time. That approach is myopically true but misleading in the grand scheme of things. My original datasets of congressional hearings, legislative activity, and the tenure and backgrounds of intelligence committee overseers over the past five decades suggest that the more important pattern is not variation over time, it is stasis: Congressional oversight has been remarkably and consistently weak no matter which party is in the White House or who leads the committees. Indeed, the 9/11 Commission called congressional intelligence oversight “dysfunctional.”[38] By fixating on oversight variation, the classic hypothesis overlooks the more interesting and important question: what explains consistent oversight weakness? The answer, I find, has to do with information asymmetries that give intelligence officials the upper hand, electoral incentives that drive legislators to give intelligence oversight short shrift, and institutional features of Congress that hinder the development of expertise and exacerbate partisanship in the House. Whether Senator X or Y chaired the committee in 1992 certainly matters, but these broader forces turn out to matter more. When it comes to explaining how and why Congress acts the way it does in intelligence, the classic “oversight-varies” hypothesis may be an oldie, but it is not a goodie. Jeffreys-Jones concludes that Spies, Lies, and Algorithms “contains some useful nuggets in chapters that roam across diverse issues.” Even his praise is all about the nuggets, not the bigger picture.

Together, the three reviewers showcase the ongoing challenges of conducting intelligence research, the need for more scholars to contribute to and recognize this field, the value of both insider and outsider perspectives, and the importance of considering variation as well as broad patterns of intelligence over time.


[1] See, for example, Christopher M. Andrew, Richard J. Aldrich, and Wesley K. Wark, eds., Secret Intelligence: A Reader, Second edition (Abingdon, Oxon; New York, NY: Routledge, 2019); Robert Dover, Routledge Companion to Intelligence Studies, Routledge Companions (Milton Park, Abingdon, Oxon: Routledge, 2014); R. Gerald Hughes, Peter Jackson, and L. V. Scott, eds., Exploring Intelligence Archives: Enquiries into the Secret State, Studies in Intelligence (Abingdon, Oxon, England; New York: Routledge, 2008); Loch Johnson and James Wirtz, Intelligence: The Secret World of Spies, Third Edition (New York: Oxford University Press, 2011); Daniel W. B. Lomas and Christopher J. Murphy, Intelligence and Espionage: Secrets and Spies (London; New York, NY: Routledge, 2019); and Adam D. M. Svendsen, Understanding the Globalization of Intelligence (Basingstoke, Hampshire; New York: Palgrave Macmillan, 2012).

[2] Abram Shulsky and Gary j. Schmitt, Silent Warfare: Understanding the World of Intelligence, Potomac Books, Nebraska: 2002; Mark Lowenthal’s Intelligence: From Secrets to Policy, CQ Press, Washington: 2019.

[3] I cannot tell you how often I cursed trying to find articles from the agency’s inhouse journal Studies in Intelligence, that I had spent countless hours providing hyperlinks for in my course syllabi, which no longer work. 

[4] My surprise, and initial skepticism, were no doubt the result of a recent experience I had after I was asked to fill in a form by the American Historical Review on those areas in American history that I felt competent to review books in. I was shocked to see there was no category listed for intelligence history.

[5] Full confession, I was one of the co-founders and past president of NASIH. I too was tired of getting rejections from SHAFR. See https://www.intelligencehistory.org for more information.

[6] This is perhaps the nicest thing I can say about “the former guy” but in doing so I must give credit to Nicolle Wallace for encouraging the use of both these phrases. Nicolle Wallace, Deadline White House, MSNBC. Every time she refers to the “twice impeached, disgraced, former president.”

[7] As an example, see page 248 where she argues that “the future is unlikely to look like the past.”

[8] Sarah-Jane Corke, US Covert Operations and Cold War Strategy: Truman, the CIA and Secret Warfare, Routledge, 2007, p.p. 53-54, 58-63; or Corke, George Kennan and the ‘Inauguration of Political Warfare,’” The Journal of Conflict Studies, Summer 2006, vol. 26, no. 1 pp., 98-117. See also Stephen Long, The CIA and the Soviet Block: Political Warfare, the Origins of the CIA and Countering Communism in Europe, I.B. Tauris, London:2014, p.p. 4,11, 38-40.

[9] For example, see: Sheldon Stern, The Cuban Missile Crisis in American Memory, Stanford University Press, 2012. For primary source evidence see Timothy Naftali, “Introduction: The Malin Notes: Glimpses Inside Moscow During the Cuban Missile Crisis,” Cold War International History Project, 2.3. 17/18 (Fall 2012): p.p. 219. https://www.wilsoncenter.org/sites/default/files/media/documents/publication/CWHIP_Bulleti....

[10] Amy B. Zegart, Spying Blind: The CIA, the FBI, and the Origins of 9/11 (Princeton, NJ: Princeton University Press, 2007).

[11] Amy Zegart and Michael Morell, “Spies, Lies, and Algorithms: Why U.S. Intelligence Must Adapt or Fail,” Foreign Affairs 98, no. 3 (June 2019): 85–96.

[12] Amy B. Zegart, “‘Spytainment’: The Real Influence of Fake Spies,” International Journal of Intelligence and CounterIntelligence 23, no. 4 (August 31, 2010): 599–622, https://doi.org/10.1080/08850607.2010.501635.

[13] Mark M. Lowenthal, Intelligence: From Secrets to Policy, 8th ed. (Thousand Oaks, CA: CQ Press, 2020).

[14] Roger Z. George, Intelligence in the National Security Enterprise: An Introduction (Washington D.C.: Georgetown University Press, 2020).

[15] Loch K. Johnson, National Security Intelligence, 2d ed. (Cambridge, UK: Polity Press, 2017).

[16] Full disclosure: as a retired naval intelligence officer and current Navy civilian professor, I can’t claim to be completely objective either. But I should note that the views expressed in this review are mine alone, and not necessarily those of the U.S. government or the Naval Postgraduate School. 

[17] John A. Gentry, “Trump-Era Politicization: A Code of Civil–Intelligence Behavior Is Needed,” International Journal of Intelligence and CounterIntelligence, September 1, 2021, 1–30, https://doi.org/10.1080/08850607.2021.1960252.

[18] Again in full disclosure, I am a former chair of the intelligence studies section of the International Studies Association.

[19] Michael Landon-Murray and Stephen Coulthart, “Intelligence Studies Programs as US Public Policy: A Survey of IC CAE Grant Recipients,” Intelligence and National Security 35, no. 2 (February 23, 2020): 269–82, https://doi.org/10.1080/02684527.2019.1703487.

[20] See for example David F. Rudgers, “The Origins of Covert Action,” Journal of Contemporary History 35, no. 2 (2000): 249–62.

[21] By comparison, Lowenthal’s Intelligence from Secrets to Policy and George’s Intelligence in the National Security Enterprise each have extensive discussions of these other forms of intelligence oversight.  

[22] In addition to the textbooks already mentioned by Lowenthal, George, and Johnson, see Carl J. Jensen III, David H. McElreath, and Melissa Graves, Introduction to Intelligence Studies, 2d ed. (New York: Routledge, 2018).

[23] Zegart, Spying Blind: The CIA, the FBI, and the Origins of 9/11, 170.

[24] Zegart and Morell, “Spies, Lies, and Algorithms.” 

[25] Patrick R. Keefe, “Privatized Spying: The Emerging Intelligence Industry”, in Loch K. Johnson, ed., The Oxford Handbook of National Security Intelligence (Oxford: Oxford University Press, 2010), 303; James Bamford, “This Spy for Rent,” New York Times, 14 June 2004; Dana Priest and William M. Arkin, Top Secret America: The Rise of the American Security State (New York: Little, Brown, 2011), 192.


[26] Kathryn S. Olmsted, The Newspaper Axis: Six Press Barons Who Enabled Hitler (New Haven, CT: Yale University Press, 2022), 2-3. Olmsted’s book was about the following newspaper proprietors: Lord Harold Rothermere, Lord Max Beaverbrook, William Randolph Hearst, Robert McCormick, Joseph Patterson and Cissy Patterson.

[27] James M. Doolittle and others, “Report on the Covert Activities of the Central Intelligence Agency,” 30 September 1954, Recommendation A.1: “Elimination of personnel who can never achieve a sufficiently high degree of competence to meet the CIA standard. This will entail a substantial reduction in present personnel”: https://www.cia.gov/readingroom/docs/doolittle_report.pdf. A statistical comparison of the relative educational attainment levels of CIA employees and a random sample of other government officials serving in the years 1982-84 suggested that the CIA scored below average, though not by a wide margin: Rhodri Jeffreys-Jones, “The Socio-Educational Composition of the CIA Elite: A Statistical Note,” Journal of American Studies, 19 (December 1985), 422. At the time of 9/11, the CIA had no personnel able to speak Pashto, the language mainly spoken by Taliban members who were giving shelter to Osama bin Laden: Thomas Powers, “The Trouble with the CIA,” New York Review of Books, 17 January 2002.

[28] Zegart documents this quotation with an unpaginated reference to Christopher Andrew, The Secret World: A History of Intelligence (New Haven: Yale University Press, 2018). On page 1 of that book, Andrew does refer to the 1588 Armada and to Napoleonic and Nazi invasion threats. He also states that in all three conflicts British cryptographers broke some enemy codes. He does not, however, assert that the codebreakers saved the UK from invasion. In the case of the Spanish invasion attempt Andrew, though an avowed admirer of aspects of the Elizabethan intelligence system, notes that the English fleet was “taken by surprise by the timing of the Armada’s arrival” off the Devon coast (The Secret World, 184). In the frustration of Adolf Hitler’s ambitions, codebreaking played little part. Operation Sealion, the provisional plan for a German invasion of England, did not get off the ground. Erich Pfeiffer, a maritime logistics specialist who had earlier run a successful spy operation against the United States, concluded in 1940 that Germany did not have enough ships and troop barges to launch an invasion. Shortly afterwards, German naval chief Admiral Erich Raeder told Hitler that he did not have enough boats. Hitler nonetheless authorized an attempt to gain the necessary air control. His air force lost the ensuing Battle of Britain (July-August 1940) and, from then on, the odds were stacked against the prospects of a successful invasion. See Derek Robinson, Invasion, 1940: The Truth about the Battle of Britain and What Stopped Hitler (London: Constable, 2005), 204, Leo McKinstry, Operation Sealion: How Britain Crushed the German War Machine’s Dreams of Invasion in 1940 (London: John Murray, 2014), 182 and, on Pfeiffer’s role, Rhodri Jeffreys-Jones, The Nazi Spy Ring in America: Hitler’s Agents, the FBI, and the Case that Stirred a Nation (Washington, D.C.: Georgetown University Press, 2020), 222.

[29] Willard M. Oliver, The Birth of the FBI: Teddy Roosevelt, the Secret Service, and the Fight Over America’s Premier Law Enforcement Agency (Lanham, MD: Rowman & Littlefield, 2019).


[30] See the breakdown offered by an anonymous  contributor to Daily Kos, 19 July 2016, “Are Democratic presidents more warlike than Republicans?”: https://www.dailykos.com/stories/2016/7/19/1549622/-Are-Democratic-presidents-more-warlike-than-Republicans. The contributor concludes that since 1861 Republican presidents have instigated, by a margin of 81 to 64, a greater number “offensive” as distinct from “defensive” wars and that there were 93 war years under Democratic presidents compared with 107 war years under their Republican peers.

[31] The volume in question is Frank J. Smist Jr., Congress Oversees the United States Intelligence Community, 1947-1994 (Knoxville: University of Tennessee Press, 1994). Zegart does not supply a page number. While Smist does state that the intelligence oversight committees met infrequently in the 1950s, he argues that the relevant legislators were interested, responsible, and discreet: Smist, 4-7.

[32] Zegart cites Loch K. Johnson, A Season of Inquiry Revisited: The Church Committee Confronts America’s Spy Agencies (Lawrence: University of Kansas Press, 2015). An earlier edition of A Season of Inquiry appeared in 1985. For approaches that are more critical of congressional achievements, see Olmsted, Challenging the Secret Government: The Post-Watergate Investigations of the CIA and FBI (Chapel Hill: University of North Carolina Press, 1996) and Jason Ross Arnold, Secrecy in the Sunshine Era: The Promise and failures of U.S. Open Government Laws (Lawrence: University of Kansas Press, 2014).

 [33] See the graph on page 225 of Harry H. Ransom, “Secret Intelligence in the United States, 1947-1982: The CIA’s Search for Legitimacy,” in Christopher M. Andrew and David Dilks, eds., The Missing Dimension: Governments and Intelligence Communities in the Twentieth Century (London: Macmillan, 1984): 199-226.


[34] Chris Baraniuk, “How Access to Satellite Images Shifts the View of War,” BBC News, 22 March, 2022, https://www.bbc.com/news/business-60762772.
[35] Amy Zegart, “Spies Like Us: The Promise and Peril of Crowdsourced Intelligence,” Foreign Affairs, Vol. 100, No. 4 (July/August 2021): 168-73.
[36] Joby Warrick, China Is Building more than 100 New Missile Silos in Its Western Desert, Analysts Say,” Washington Post, 30 June, 2021, https://www.washingtonpost.com/national-security/china-nuclear-missile-silos/2021/06/30/0fa8debc-d9c2-11eb-bb9e-70fda8c37057_story.html.

[37] Amy B. Zegart, Spying Blind: The CIA, the FBI, and the Origins of 9/11 (Princeton: Princeton University Press, 2007).

[38] National Commission on Terrorist Attacks Upon the United States, The 9/11 Commission Report, Authorized Version (New York: W.W. Norton, 2004), 420.