What do you think?
Rate this book
Why Intelligence Fails: Lessons from the Iranian Revolution and the Iraq War
The U.S. government spends enormous resources each year on the gathering and analysis of intelligence, yet the history of American foreign policy is littered with missteps and misunderstandings that have resulted from intelligence failures. In Why Intelligence Fails, Robert Jervis examines the politics and psychology of two of the more spectacular intelligence failures in recent memory: the mistaken belief that the regime of the Shah in Iran was secure and stable in 1978, and the claim that Iraq had active WMD programs in 2002.
The Iran case is based on a recently declassified report Jervis was commissioned to undertake by CIA thirty years ago and includes memoranda written by CIA officials in response to Jervis's findings. The Iraq case, also grounded in a review of the intelligence community's performance, is based on close readings of both classified and declassified documents, though Jervis's conclusions are entirely supported by evidence that has been declassified. In both cases, Jervis finds not only that intelligence was badly flawed but also that later explanations--analysts were bowing to political pressure and telling the White House what it wanted to hear or were willfully blind--were also incorrect. Proponents of these explanations claimed that initial errors were compounded by groupthink, lack of coordination within the government, and failure to share information. Policy prescriptions, including the recent establishment of a Director of National Intelligence, were supposed to remedy the situation.
In Jervis's estimation, neither the explanations nor the prescriptions are adequate. The inferences that intelligence drew were actually quite plausible given the information available. Errors arose, he concludes, from insufficient attention to the ways in which information should be gathered and interpreted, a lack of self-awareness about the factors that led to the judgments, and an organizational culture that failed to probe for weaknesses and explore alternatives. Evaluating the inherent tensions between the methods and aims of intelligence personnel and policymakers from a unique insider's perspective, Jervis forcefully criticizes recent proposals for improving the performance of the intelligence community and discusses ways in which future analysis can be improved.
The Iran case is based on a recently declassified report Jervis was commissioned to undertake by CIA thirty years ago and includes memoranda written by CIA officials in response to Jervis's findings. The Iraq case, also grounded in a review of the intelligence community's performance, is based on close readings of both classified and declassified documents, though Jervis's conclusions are entirely supported by evidence that has been declassified. In both cases, Jervis finds not only that intelligence was badly flawed but also that later explanations--analysts were bowing to political pressure and telling the White House what it wanted to hear or were willfully blind--were also incorrect. Proponents of these explanations claimed that initial errors were compounded by groupthink, lack of coordination within the government, and failure to share information. Policy prescriptions, including the recent establishment of a Director of National Intelligence, were supposed to remedy the situation.
In Jervis's estimation, neither the explanations nor the prescriptions are adequate. The inferences that intelligence drew were actually quite plausible given the information available. Errors arose, he concludes, from insufficient attention to the ways in which information should be gathered and interpreted, a lack of self-awareness about the factors that led to the judgments, and an organizational culture that failed to probe for weaknesses and explore alternatives. Evaluating the inherent tensions between the methods and aims of intelligence personnel and policymakers from a unique insider's perspective, Jervis forcefully criticizes recent proposals for improving the performance of the intelligence community and discusses ways in which future analysis can be improved.
248 pages, Hardcover
First published February 4, 2010
50 people are currently reading
864 people want to read
About the author
Robert Jervis
47 books60 followersRobert Jervis is the Adlai E. Stevenson Professor of International Affairs at Columbia University, and has been a member of the faculty since 1980. Jervis was the recipient of the 1990 University of Louisville Grawemeyer Award for Ideas Improving World Order. Jervis is co-editor of the Cornell Studies in Security Affairs, a series published by Cornell University Press, and the member of numerous editorial review boards for scholarly journals.
While Jervis is perhaps best known for two books in his early career, he also wrote System Effects: Complexity in Political and Social Life (Princeton, 1997). With System Effects, Jervis established himself as a social scientist as well as an expert in international politics. Many of his latest writings are about the Bush doctrine, of which he is very critical. Jervis is a member of the American Association for the Advancement of Science and the American Academy of Arts and Sciences. In 2006 he was awarded the NAS Award for Behavior Research Relevant to the Prevention of Nuclear War from the National Academy of Sciences. He participated in the 2010 Hertog Global Strategy Initiative, a high-level research program on nuclear proliferation. He was also president of the American Political Science Association in 2001.
While Jervis is perhaps best known for two books in his early career, he also wrote System Effects: Complexity in Political and Social Life (Princeton, 1997). With System Effects, Jervis established himself as a social scientist as well as an expert in international politics. Many of his latest writings are about the Bush doctrine, of which he is very critical. Jervis is a member of the American Association for the Advancement of Science and the American Academy of Arts and Sciences. In 2006 he was awarded the NAS Award for Behavior Research Relevant to the Prevention of Nuclear War from the National Academy of Sciences. He participated in the 2010 Hertog Global Strategy Initiative, a high-level research program on nuclear proliferation. He was also president of the American Political Science Association in 2001.
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 19 of 19 reviews
December 26, 2018
A useful but somewhat odd book about U.S. intelligence failures. I say that it is odd because almost half of the book is just an intelligence report that Jervis wrote as a consultant to the CIA in 1980. It offers some interesting ideas as to why US intelligence missed/failed to anticipate that the Shah was in big trouble in 1978. This memo, did not exactly make enjoyable or efficient reading. It would have been better if Jervis had written up a shorter chapter for Iran, just like he did for Iraq. That criticism aside, there are still tons of helpful ways of thinking about intelligence work in this book. 2 things struck me in particular:
1. HIs chapter on Iraq WMD failure did a nice job challenging some of the common wisdom about this stuff. For instance, he argues that true politicization of intelligence did not occur. Admin hawks did exert pressure on the agencies to look into certain things over and over again, they frowned upon and criticized intel that didn't fit their narrative, and they set up alternative channels of intelligence. However, the agencies apparently did not change their estimates of IQ's WMD programs to fit what the administration was looking for. Had the agencies done a better job, they would have produced less overall certainty than they did in reality because the intelligence was murkier than the agencies assessed. I'd say this is a good read for anyone looking for a balanced but still incisive account of the WMD fiasco.
2. Jervis raises a great way of thinking about any problem, especially prediction-type problems. You should always imagine what kind of evidence you would be seeing if the alternative/opposite(s) to your prediction/assessment was actually happening. This is the "dogs that didn't bark" thing from Sherlock Holmes. If this perspective had been approached more systematically via IQ, it might have sowed more doubt about their WMD program. After all, you weren't seeing a large-scale effort to acquire key materials, build new facilities, organize personnel, etc. We were arguably hunting for everything that fit the WMD narrative (which, as Jervis pointed out, wasn't a crazy thing to believe) and missing the larger pile of actions and non-actions that suggested Saddam wasn't making much of a WMD effort.
This book is pretty technical, so I'd recommend skipping the Iran chapter and focusing on his general discussion of intelligence failures and his Iraq chapter. If you are a beginner to political science and psychology and intel work, might want to start elsewhere.
1. HIs chapter on Iraq WMD failure did a nice job challenging some of the common wisdom about this stuff. For instance, he argues that true politicization of intelligence did not occur. Admin hawks did exert pressure on the agencies to look into certain things over and over again, they frowned upon and criticized intel that didn't fit their narrative, and they set up alternative channels of intelligence. However, the agencies apparently did not change their estimates of IQ's WMD programs to fit what the administration was looking for. Had the agencies done a better job, they would have produced less overall certainty than they did in reality because the intelligence was murkier than the agencies assessed. I'd say this is a good read for anyone looking for a balanced but still incisive account of the WMD fiasco.
2. Jervis raises a great way of thinking about any problem, especially prediction-type problems. You should always imagine what kind of evidence you would be seeing if the alternative/opposite(s) to your prediction/assessment was actually happening. This is the "dogs that didn't bark" thing from Sherlock Holmes. If this perspective had been approached more systematically via IQ, it might have sowed more doubt about their WMD program. After all, you weren't seeing a large-scale effort to acquire key materials, build new facilities, organize personnel, etc. We were arguably hunting for everything that fit the WMD narrative (which, as Jervis pointed out, wasn't a crazy thing to believe) and missing the larger pile of actions and non-actions that suggested Saddam wasn't making much of a WMD effort.
This book is pretty technical, so I'd recommend skipping the Iran chapter and focusing on his general discussion of intelligence failures and his Iraq chapter. If you are a beginner to political science and psychology and intel work, might want to start elsewhere.
January 6, 2022
In the first decade of the twenty-first century, the US launched two costly wars, ostensibly because of intelligence failures. The first was failing to prevent 9/11. The second was waging a horrible war in Iraq, based on faulty intelligence about WMDs (weapons of mass destruction). Were they because of substandard intelligence? Or were they because of the Bush administration pressuring the CIA to give it favorable intelligence to justify going to war? Or was it collective cognitive failure across the administration, military, and the media? Prof. Jervis is a distinguished political scientist known for his work on perceptions and misperceptions in foreign policy decision-making. In this book, he analyses two of the more spectacular intelligence failures in the past four decades. The first failure was the mistaken belief that the Shah of Iran was secure in 1978, despite massive protests in the country. Ayatollah Khomeini led a revolution to overthrow the Shah months after this intelligence assessment. The second was the claim that Iraq had WMDs in its possession in 2002. Three years later, the Iraq Survey Group (ISG) concluded that Iraq had abandoned its quest to develop chemical, biological and nuclear weapons. It added Iraq had already destroyed all its existing stockpiles of chemical and biological weapons. The ISG comprised hundreds of intelligence analysts and military personnel from the United States, the United Kingdom, and Australia.
Analyzing the intelligence failure on the Shah of Iran, Prof. Jervis sees four major errors. First, the CIA figured the Shah was a strong, self-confident, and arrogant leader and would do what he needed to remain in power. Second, they considered the situation not dangerous in 1978 because the Shah did not crack down on the opposition. Third, no one in and out of the US government understood the role of religion in Iran and Ayatollah Khomeini’s appeal. In 1978, most analysts believed that a retrograde force such as religion was unlikely to win over the Iranians by challenging the Shah’s modernization. Fourth, the US misunderstood the power of Iranian nationalism. They failed to realize the Iranian people saw the Shah as an American puppet. This helped disparate opposition groups to come together to overthrow him.
Apart from this, the author says the CIA had few resources allocated to Iran. It had only two political and two economic analysts covering Iran. Iran’s crisis in 1978 was, at its roots, an economic one, but the economic analysts had scant communication with the political analysts. The CIA also had few contacts with academics and scholars who specialized in Iran. One surprise was the CIA being unaware the Shah had cancer in 1978. This was important in assessing why the Shah acted irresolutely against the opposition. Last, Khomeini’s opposition resulted in a theocratic revolution. History shows revolutions have always been difficult to forecast.
On the WMD fiasco in Iraq, Prof. Jervis begins by giving credit to conventional wisdom for being correct on three points regarding intelligence failure. First, the Intelligence Community’s (IC) judgments stated the existence of WMDs in Iraq with excessive certainty when the evidence did not prove their existence. Second, the IC did not consider alternative explanations for the evidence at hand. For example, the CIA was sure the aluminum tubes Iraq was importing were for enriching uranium. Other government departments believed correctly these tubes were not well-suited for this purpose. It turned out that Saddam did it to bluff Iran into believing he had an active WMD program. Third, the IC showed a lack of imagination in evaluating available intelligence. For example, Saddam’s behavior eighteen months before the war showed he was certain the US would overthrow him. National leaders always seek to avoid this fate, but Saddam’s conduct was suicidal. But the IC showed little curiosity in analyzing it and continued to work according to the field reports. However, Prof. Jervis discounts other conventional explanations for the intelligence failure. He does not agree that the IC bowed to political pressure or told the administration what it wanted to hear. Nor does he agree that the decision-makers suffered from groupthink or a failure to share key information amongst themselves.
The last chapter in the book looks at the role psychology affects intelligence analysts and how they evaluate information. Premature cognitive closure influences analysts and affects their decision making. This is a type of cognitive error in which the analyst neglects to consider reasonable alternatives after he makes an initial conclusion. The decision on Saddam’s aluminum tube import is a good example. This is humbling because the IC gets trained much better than the public in guarding against its own intellectual fragilities. The Rashomon effect was another major element at play in Iraq in 2002. It refers to an instance when different people view the same event in rather different and contradictory ways. The US found Saddam’s fears and beliefs hard to grasp and understand. Saddam also failed to understand the US and couldn’t predict what the Bush administration would do. The chapter closes with ideas for improving the IC’s performance and how analytical outcomes can get better in the future.
It is important to remember the author was only looking at the causes of intelligence failures in Iran in 1979 and Iraq in 2002. He is not arguing that the US may not have gone to war in Iraq had intelligence on WMDs pointed to their absence. Neither is he arguing that more accurate intelligence could have thwarted Khomeini’s revolution in 1979. Twenty years after the Iraq war, we now have enough evidence that both George Bush and Tony Blair were hell-bent on removing Saddam Hussein from power. This was so even if WMDs were not found in Iraq. So, the 106000 Iraqi civilian deaths and close to 5000 coalition soldiers’ deaths might have happened regardless of the intelligence. Nor could we have avoided the $800 billion spent on Iraq between 2003 and 2011. In a similar vein, on the Iranian front, Iranian nationalism was rising high against the Shah of Iran because of the perception that the Shah was an American stooge. He might have fallen from power even if the quality of intelligence and analysis had been high.
Prof. Jervis chose the examples of Iran and Iraq for his analysis because this is what the CIA commissioned him to do. These examples raise moral issues which cloud the conclusion as intelligence failure. With Iran, the US was on the side of a dictator who was trying to stop a leader with immense popular support. It was in line with American values the dictator got defeated. With Iraq, the US faked intelligence to wage an illegal war. This too is against American values. To call them both ‘intelligence failures’ is tantamount to admitting we expect the CIA to work against American values. It would have been more useful to take value-neutral events to do the analysis of intelligence failure. Failing to predict 9/11, the inability to foresee the Soviet collapse in 1991 or the fall of Eastern Europe in 1989 may have been better candidates for the investigation.
We are now in the era of pervasive electronic surveillance. Governments amass exabytes of information from around the world. Such surveillance generates thousands of leads each day to follow. It is impossible to determine which leads to spend resources on without the help of artificial intelligence (AI). But AI has not advanced enough to generate manageable and productive leads out of the thousands. So, the more information we gather, the more likely we would miss the valuable leads because they get buried among thousands of dubious ones. Even a casual look at the problem shows AI needs to develop artificial general intelligence (AGI) to achieve the required accuracy. But AGI is decades away from happening or perhaps ‘never’, according to researchers in the field. So, we may likely see more intelligence failures despite all the supercomputers and petabyte of storage.
It is an analytical, slow-paced book.
Analyzing the intelligence failure on the Shah of Iran, Prof. Jervis sees four major errors. First, the CIA figured the Shah was a strong, self-confident, and arrogant leader and would do what he needed to remain in power. Second, they considered the situation not dangerous in 1978 because the Shah did not crack down on the opposition. Third, no one in and out of the US government understood the role of religion in Iran and Ayatollah Khomeini’s appeal. In 1978, most analysts believed that a retrograde force such as religion was unlikely to win over the Iranians by challenging the Shah’s modernization. Fourth, the US misunderstood the power of Iranian nationalism. They failed to realize the Iranian people saw the Shah as an American puppet. This helped disparate opposition groups to come together to overthrow him.
Apart from this, the author says the CIA had few resources allocated to Iran. It had only two political and two economic analysts covering Iran. Iran’s crisis in 1978 was, at its roots, an economic one, but the economic analysts had scant communication with the political analysts. The CIA also had few contacts with academics and scholars who specialized in Iran. One surprise was the CIA being unaware the Shah had cancer in 1978. This was important in assessing why the Shah acted irresolutely against the opposition. Last, Khomeini’s opposition resulted in a theocratic revolution. History shows revolutions have always been difficult to forecast.
On the WMD fiasco in Iraq, Prof. Jervis begins by giving credit to conventional wisdom for being correct on three points regarding intelligence failure. First, the Intelligence Community’s (IC) judgments stated the existence of WMDs in Iraq with excessive certainty when the evidence did not prove their existence. Second, the IC did not consider alternative explanations for the evidence at hand. For example, the CIA was sure the aluminum tubes Iraq was importing were for enriching uranium. Other government departments believed correctly these tubes were not well-suited for this purpose. It turned out that Saddam did it to bluff Iran into believing he had an active WMD program. Third, the IC showed a lack of imagination in evaluating available intelligence. For example, Saddam’s behavior eighteen months before the war showed he was certain the US would overthrow him. National leaders always seek to avoid this fate, but Saddam’s conduct was suicidal. But the IC showed little curiosity in analyzing it and continued to work according to the field reports. However, Prof. Jervis discounts other conventional explanations for the intelligence failure. He does not agree that the IC bowed to political pressure or told the administration what it wanted to hear. Nor does he agree that the decision-makers suffered from groupthink or a failure to share key information amongst themselves.
The last chapter in the book looks at the role psychology affects intelligence analysts and how they evaluate information. Premature cognitive closure influences analysts and affects their decision making. This is a type of cognitive error in which the analyst neglects to consider reasonable alternatives after he makes an initial conclusion. The decision on Saddam’s aluminum tube import is a good example. This is humbling because the IC gets trained much better than the public in guarding against its own intellectual fragilities. The Rashomon effect was another major element at play in Iraq in 2002. It refers to an instance when different people view the same event in rather different and contradictory ways. The US found Saddam’s fears and beliefs hard to grasp and understand. Saddam also failed to understand the US and couldn’t predict what the Bush administration would do. The chapter closes with ideas for improving the IC’s performance and how analytical outcomes can get better in the future.
It is important to remember the author was only looking at the causes of intelligence failures in Iran in 1979 and Iraq in 2002. He is not arguing that the US may not have gone to war in Iraq had intelligence on WMDs pointed to their absence. Neither is he arguing that more accurate intelligence could have thwarted Khomeini’s revolution in 1979. Twenty years after the Iraq war, we now have enough evidence that both George Bush and Tony Blair were hell-bent on removing Saddam Hussein from power. This was so even if WMDs were not found in Iraq. So, the 106000 Iraqi civilian deaths and close to 5000 coalition soldiers’ deaths might have happened regardless of the intelligence. Nor could we have avoided the $800 billion spent on Iraq between 2003 and 2011. In a similar vein, on the Iranian front, Iranian nationalism was rising high against the Shah of Iran because of the perception that the Shah was an American stooge. He might have fallen from power even if the quality of intelligence and analysis had been high.
Prof. Jervis chose the examples of Iran and Iraq for his analysis because this is what the CIA commissioned him to do. These examples raise moral issues which cloud the conclusion as intelligence failure. With Iran, the US was on the side of a dictator who was trying to stop a leader with immense popular support. It was in line with American values the dictator got defeated. With Iraq, the US faked intelligence to wage an illegal war. This too is against American values. To call them both ‘intelligence failures’ is tantamount to admitting we expect the CIA to work against American values. It would have been more useful to take value-neutral events to do the analysis of intelligence failure. Failing to predict 9/11, the inability to foresee the Soviet collapse in 1991 or the fall of Eastern Europe in 1989 may have been better candidates for the investigation.
We are now in the era of pervasive electronic surveillance. Governments amass exabytes of information from around the world. Such surveillance generates thousands of leads each day to follow. It is impossible to determine which leads to spend resources on without the help of artificial intelligence (AI). But AI has not advanced enough to generate manageable and productive leads out of the thousands. So, the more information we gather, the more likely we would miss the valuable leads because they get buried among thousands of dubious ones. Even a casual look at the problem shows AI needs to develop artificial general intelligence (AGI) to achieve the required accuracy. But AGI is decades away from happening or perhaps ‘never’, according to researchers in the field. So, we may likely see more intelligence failures despite all the supercomputers and petabyte of storage.
It is an analytical, slow-paced book.
Read
August 3, 2025The conclusion is quite alarming:
“In retrospect, the most reasonable assessment would have been that Iraq probably (but not certainly) had active and broadly based WMD programs…”
It’s chilling to consider that the analysis that led to something as catastrophic as the Iraq War was well reasoned, and in large part correct.
“In retrospect, the most reasonable assessment would have been that Iraq probably (but not certainly) had active and broadly based WMD programs…”
It’s chilling to consider that the analysis that led to something as catastrophic as the Iraq War was well reasoned, and in large part correct.
March 9, 2019
This book would be beneficial to anyone with an interest in intelligence, policy/politics (especially foreign policy), and even critical rational thinking as such. In terms of style, Jervis is at his best when speaking freely in a broad, sweeping style. His writing becomes a bit more tedious during his two case studies (the Islamic Revolution in Iran and WMD in Iraq), but he is understandably thorough. The Iran case study comes with some CIA reactions to his study attached at the end, which are also interesting. Some portions of both case studies have been redacted, but that doesn't impact readability too badly.
Overall, Jervis gives a fairly pessimistic view, rightly noting that intelligence failure postmortems are a repeat exercise and it is very unlikely that intelligence will ever be failure-proof. Part of that is the nature of the beast; while media and laymen like to discuss "evidence" and "proof" and speak of intelligence assessments and conclusions as if they were convictions in criminal cases, the fact is that intelligence by definition deals with the murky and partially known. But even under the best of circumstances, Jervis points to weaknesses in intelligence analysis (he does not focus on collection) that likely will not be corrected anytime soon. Indeed, he is fond of pointing out that politicians would rather have faulty intelligence to blame rather than the best of intelligence and have to accept more of the burden for failure themselves. Ultimately, a lot of policy decisions come down to subjective value systems, assumptions, paradigms about the world, and so on, that cannot be proven and exist independently of specific empirical evidence in individual cases. Jervis identifies a very narrow window when intelligence might affect politicians' decisions, "after the leaders have become seized with the problem but before they have made up their minds." (location 3995)
One of Jervis' interesting observations is that intelligence "false alarms" and intelligence successes ought also to be studied. In that way, even if a given factor is present in all studied failures, should that same factor be present in successes and false alarms, then it may not be as causal as first indicated.
Jervis also notes that intelligence failures tend to be bilateral, in that one side badly misjudges the other's probable reactions to a planned course of action, and the other badly assesses the first side's plans and intentions as being unbelievable in light of the second side's certain response. He points out the obvious-but-still-true maxim that continuity tends to get focus over change, particularly radical or otherwise unexpected change. In fact, Chapter 4 is chock full of a good discussion on critical rational thought generally, not just tied to intelligence matters.
So in all his pessimism, Jervis reminds us that "perhaps the most important function of intelligence is to raise questions." (location 4321). For an academic like Jervis, this is certainly germane, but for the sorts of action-first think-later people who "we the people" often favor and elect, it really does probably make very little difference to outcomes.
In terms of its substance, ability to provoke thought, its deep handling of the topic, and its prose in Chapters 1 and 4, it deserves four stars. The case studies probably rate closer to three, as they are (understandably, perhaps) a bit drier, more difficult reading. Still, this work is not a long tome and it is not highly technical. Jervis defines his acronyms, but his repeated use of some may still mystify those without a deep background with intelligence. Overall a good read, I recommend it.
Overall, Jervis gives a fairly pessimistic view, rightly noting that intelligence failure postmortems are a repeat exercise and it is very unlikely that intelligence will ever be failure-proof. Part of that is the nature of the beast; while media and laymen like to discuss "evidence" and "proof" and speak of intelligence assessments and conclusions as if they were convictions in criminal cases, the fact is that intelligence by definition deals with the murky and partially known. But even under the best of circumstances, Jervis points to weaknesses in intelligence analysis (he does not focus on collection) that likely will not be corrected anytime soon. Indeed, he is fond of pointing out that politicians would rather have faulty intelligence to blame rather than the best of intelligence and have to accept more of the burden for failure themselves. Ultimately, a lot of policy decisions come down to subjective value systems, assumptions, paradigms about the world, and so on, that cannot be proven and exist independently of specific empirical evidence in individual cases. Jervis identifies a very narrow window when intelligence might affect politicians' decisions, "after the leaders have become seized with the problem but before they have made up their minds." (location 3995)
One of Jervis' interesting observations is that intelligence "false alarms" and intelligence successes ought also to be studied. In that way, even if a given factor is present in all studied failures, should that same factor be present in successes and false alarms, then it may not be as causal as first indicated.
Jervis also notes that intelligence failures tend to be bilateral, in that one side badly misjudges the other's probable reactions to a planned course of action, and the other badly assesses the first side's plans and intentions as being unbelievable in light of the second side's certain response. He points out the obvious-but-still-true maxim that continuity tends to get focus over change, particularly radical or otherwise unexpected change. In fact, Chapter 4 is chock full of a good discussion on critical rational thought generally, not just tied to intelligence matters.
So in all his pessimism, Jervis reminds us that "perhaps the most important function of intelligence is to raise questions." (location 4321). For an academic like Jervis, this is certainly germane, but for the sorts of action-first think-later people who "we the people" often favor and elect, it really does probably make very little difference to outcomes.
In terms of its substance, ability to provoke thought, its deep handling of the topic, and its prose in Chapters 1 and 4, it deserves four stars. The case studies probably rate closer to three, as they are (understandably, perhaps) a bit drier, more difficult reading. Still, this work is not a long tome and it is not highly technical. Jervis defines his acronyms, but his repeated use of some may still mystify those without a deep background with intelligence. Overall a good read, I recommend it.
December 27, 2023
A really interesting book about intelligence that uses a few case studies to discuss issues of intelligence making and policy. Those questions and problems are endemic to the enterprise of intelligence and some of them are nearly impossible to solve. Jervis is an advocate for more systematic methods of training analysts and having them perform their jobs - using social science methodologies - and managing the writing and dissemination process of intelligence better at the middle management level.
This is a book that I'll refer to in both undergraduate and graduate courses on intelligence and national security. Very good.
This is a book that I'll refer to in both undergraduate and graduate courses on intelligence and national security. Very good.
April 12, 2024
This book is dry, but it contains some nuggets of wisdom pertaining to analytic tradecraft. The case studies are very dense. It's by no means a thrilling read. I can't say I'd read it again – it's a giant roast and so astoundingly detailed that it could bore one to sleep. I've already taken notes and highlighted lines I found particularly insightful. There are very good points made and as a result, this should be required reading (maybe excerpts, not the whole thing) for entry level analysts. To understand the pitfalls that can happen easily and unintentionally is valuable. I found this an excellent resource, though perhaps not the most compelling read.
April 11, 2020
This is an excellent book for what it is, which is an incredibly detailed report of intelligence failures related to the fall of the Shah in Iran, and the run up to the Iraq war in 2003. There are also excellent descriptions of cognitive biases that are useful in other disciplines. It seems this book would be mandatory reading for anyone interested in working for the Agency, the history of the Agency, the history of the 2003 Iraq war, or the history of the 1979 Iranian revolution. Other than that, you may find it dry and a little too detailed to sustain continued interest.
December 3, 2017
Good read on important issues. A bit dated. Reading chapter 3 was a bit tedious - I recommend skipping that. Chapter 2 is an overview and summary of the report in Chapter 3 and is sufficient. I'd love to see another similar study/book on the current state of the IC.
July 1, 2016
Using two case studies, one about the fall of the Shah in Iran and the other about the WMD failures the author shows how and why intelligence organizations fail. The premise that he sells in this relatively short book (<300 pages) is that intelligence analysts both don't know how to ask questions of others and are given a chance to really become subject matter experts in areas. Some of this the author blames on the apparent lack of analysts to foster connections outside of their bubble inside an organization. Such as the CIA analysts always talking to CIA analysts and not to folks over at State or even going over to Georgetown and asking some of the think tanks there. The other major blame he has for intelligence failures is that there appears to be no or limited "interest" in setting folks up to become subject matter experts. Thru the use of anecdotal information, the author who was a contractor at the time of the Iranian Revolution to the CIA, shows that there were folks who not only didn't speak Farsi; some of them had never been to the region or had studied it in depth beyond the fact book published by the CIA and was depending on folks at the embassy to provide information. Most of whom were out of their element and were Russian linguists or Russian experts to work the intelligence gathering apparatus directed towards the USSR at the time. After throwing out this summary of information for the Iranian revolution as well as how he became involved in a postmortem report on what went wrong. The author then republishes all the un-redacted parts of his report that was declassified about the failures of intelligence during the Iranian Revolution up to the summer of 1978 when the Shah abdicated. The biggest problem here though is that this is typical bureaucratic writing and it is very dry. Even for what should be an informational book from a professor. This report easily makes up over 70% of the pages in the book.
The rest of the book is made up of a postmortem that the author did on his own based on released information and experiences within the CIA to show that even the story that most folks were told by the media about the WMD hunt and cause belli for Iraq wasn't so much wrong. Nor was it politicized, he makes this point numerous times through out the last half of the book. Rather it that because of a successful deception campaign and misdirection campaign by the Hussein government. Most of the intelligence seemed to point to the fact that the Iraqis were trying to get WMDs started again. The author then explains that there was so much evidence that seemed to point this way as well as biases already set in by the analysts they saw that it was easy to draw the conclusions that they drew. Even when there was evidence to the contrary, simply because "guilty people have reason to hide stuff". Therefore, the failure here the author explains is again that folks didn't know the region well, even though we had been watching for 20+ years, because there was a serious draw down of experience from the end of the Cold War to the start of the War on Terror. As well folks rotating in and thru and around the various desks all through out the intelligence communities in both the US and the UK seemed to carry their baggage about Iraq with them. Therefore, just as before with Iran, key evidence was missed or dismissed as not useful to talk about. The experience levels weren't there and in the end the failures were more from the lack of interests and lack of time to develop a coherent policy about what was being seen versus what was actually being written up in a final product. In the Iranian case, events happened so fast that there was no time to throw out the alternative explanations. While in Iraq, there was so much time that once an idea set in it grew to be the permanent idea within the community even before the policy makers in the government pushed for the information to bend to fit the policy being drafted.
The fixes proposed by the author are interesting, but if you have studied the intelligence community in any depth by outsiders or even insiders; they are the same ones that have been tossed around since the Stansfield Turner as the head of the CIA and Church Hearings days.
This is an interesting book, but not really a 4 or even 3 full stars book; mainly because the middle of this gets muddled down with a redacted paper that is over 30 years old written in dry government language that seems to take forever to wade thru with findings or suggestions for improvements. I would suggest giving it a whirl if only because the author offers up a different take on the debacle that was the intelligence on the Iraqi WMDs that might be different than the official line in the government and the official line that was pushed by the media.
The rest of the book is made up of a postmortem that the author did on his own based on released information and experiences within the CIA to show that even the story that most folks were told by the media about the WMD hunt and cause belli for Iraq wasn't so much wrong. Nor was it politicized, he makes this point numerous times through out the last half of the book. Rather it that because of a successful deception campaign and misdirection campaign by the Hussein government. Most of the intelligence seemed to point to the fact that the Iraqis were trying to get WMDs started again. The author then explains that there was so much evidence that seemed to point this way as well as biases already set in by the analysts they saw that it was easy to draw the conclusions that they drew. Even when there was evidence to the contrary, simply because "guilty people have reason to hide stuff". Therefore, the failure here the author explains is again that folks didn't know the region well, even though we had been watching for 20+ years, because there was a serious draw down of experience from the end of the Cold War to the start of the War on Terror. As well folks rotating in and thru and around the various desks all through out the intelligence communities in both the US and the UK seemed to carry their baggage about Iraq with them. Therefore, just as before with Iran, key evidence was missed or dismissed as not useful to talk about. The experience levels weren't there and in the end the failures were more from the lack of interests and lack of time to develop a coherent policy about what was being seen versus what was actually being written up in a final product. In the Iranian case, events happened so fast that there was no time to throw out the alternative explanations. While in Iraq, there was so much time that once an idea set in it grew to be the permanent idea within the community even before the policy makers in the government pushed for the information to bend to fit the policy being drafted.
The fixes proposed by the author are interesting, but if you have studied the intelligence community in any depth by outsiders or even insiders; they are the same ones that have been tossed around since the Stansfield Turner as the head of the CIA and Church Hearings days.
This is an interesting book, but not really a 4 or even 3 full stars book; mainly because the middle of this gets muddled down with a redacted paper that is over 30 years old written in dry government language that seems to take forever to wade thru with findings or suggestions for improvements. I would suggest giving it a whirl if only because the author offers up a different take on the debacle that was the intelligence on the Iraqi WMDs that might be different than the official line in the government and the official line that was pushed by the media.
April 17, 2010
Why does intelligence fail? Is there anything we can do to address failure? Those are key questions addressed by political scientist Robert Jervis. His methodology is appropriate: he uses two case studies to examine intelligence failure and how one might diagnose failure and improve matters to reduce the odds of major failure in the future.
The two case studies are the Iranian Revolution and the fall of the Shah and the inaccurate intelligence on weapons of mass destruction (WMD) in the run up to the Iraq War and the deposing of Saddam Hussein. The first of the case studies is based on Jervis' own report to the CIA after the failure of intelligence in the Iran event. Here, the Shah of Iran was overthrown and the United States was caught unaware. Intelligence breakdowns of one sort or another were a part of this (from descriptive rather than analytic intelligence, pre-existing beliefs about what was going on, and the like). While Jervis notes some means of addressing problems, he notes that it would not be certain that--even then--the rapid fall of the Shah would have been clearly predicted.
The second case study is the failure of intelligence regarding WMD in Iraq. The book examines the nature of the failure and then tries to explain that failure. Among factors leading to failure--confirmation bias (seeking information to demonstrate that there were WMD), individual analysts'/operatives' failures (e.g., trusting Curveball, a source of information on Iraq who was, as it turns out, not a reliable source).
How to address intelligence failure? Jervis is critical of the "reforms" initiated within the intelligence community. Among these: postmortems (discussion of intelligence efforts after an event, to learn what went right and went wrong and try to draw lessons for future intelligence operations), peer review of intelligence reports, and so on.
Overall, an interesting--if somewhat dry--account of how one might improve intelligence. Also, there are valuable suggestions about simple solutions to a complex problem. The book does provoke some reflection, which is all to the good.
The two case studies are the Iranian Revolution and the fall of the Shah and the inaccurate intelligence on weapons of mass destruction (WMD) in the run up to the Iraq War and the deposing of Saddam Hussein. The first of the case studies is based on Jervis' own report to the CIA after the failure of intelligence in the Iran event. Here, the Shah of Iran was overthrown and the United States was caught unaware. Intelligence breakdowns of one sort or another were a part of this (from descriptive rather than analytic intelligence, pre-existing beliefs about what was going on, and the like). While Jervis notes some means of addressing problems, he notes that it would not be certain that--even then--the rapid fall of the Shah would have been clearly predicted.
The second case study is the failure of intelligence regarding WMD in Iraq. The book examines the nature of the failure and then tries to explain that failure. Among factors leading to failure--confirmation bias (seeking information to demonstrate that there were WMD), individual analysts'/operatives' failures (e.g., trusting Curveball, a source of information on Iraq who was, as it turns out, not a reliable source).
How to address intelligence failure? Jervis is critical of the "reforms" initiated within the intelligence community. Among these: postmortems (discussion of intelligence efforts after an event, to learn what went right and went wrong and try to draw lessons for future intelligence operations), peer review of intelligence reports, and so on.
Overall, an interesting--if somewhat dry--account of how one might improve intelligence. Also, there are valuable suggestions about simple solutions to a complex problem. The book does provoke some reflection, which is all to the good.
March 12, 2019
With „Why Intelligence Fails“ Robert Jervis provides a detailed and very insightful critique of two intelligence estimates but discerns broader problems as well as possible solutions to make intelligence analysis better from there.
The author was contracted by the CIA to conduct a post mortem analysis of CIA estimates of the stability of the Persian government before the revolution of 1978. He uses this as well as the documents on the NIE concerning Iraqi WMD to analyze why the estimates were wrong. Jervis did not write a book for the average reader but for intelligence professionals and therefore does not make an attempt to simplify his language or catch the reader’s attention. He neither guides the reader nor dramatizes his account but very objectively states the facts and makes deductions.
After a short introduction Jervis provides some background to the CIA estimates on the Shah’s regime and his own report. He then provides the bulk of his original unclassified report and discusses his findings as well as the CIA reaction to them. His discussion of the WMD NIE is more to the point as less details are publicly available. Again some context is given and objective details are critiqued. Especially interesting here is his assessment the political level of the administration indeed made it clear, what results it demanded and exerted pressure to get them but neither interfered nor rewritten any part of the intelligence estimate. In the final chapter Jervis provides possible solutions to the problems he discussed and points toward a different organizational as well as analytical culture within the Intelligence Community.
In summary this is a well done critique of the intelligence analyses in question which could have been written better. While mainly interesting to the professional reader concerned with intelligence analysis it does hold many interesting insights for readers interested in the historical events concerned and what part the intelligence estimates played.
The author was contracted by the CIA to conduct a post mortem analysis of CIA estimates of the stability of the Persian government before the revolution of 1978. He uses this as well as the documents on the NIE concerning Iraqi WMD to analyze why the estimates were wrong. Jervis did not write a book for the average reader but for intelligence professionals and therefore does not make an attempt to simplify his language or catch the reader’s attention. He neither guides the reader nor dramatizes his account but very objectively states the facts and makes deductions.
After a short introduction Jervis provides some background to the CIA estimates on the Shah’s regime and his own report. He then provides the bulk of his original unclassified report and discusses his findings as well as the CIA reaction to them. His discussion of the WMD NIE is more to the point as less details are publicly available. Again some context is given and objective details are critiqued. Especially interesting here is his assessment the political level of the administration indeed made it clear, what results it demanded and exerted pressure to get them but neither interfered nor rewritten any part of the intelligence estimate. In the final chapter Jervis provides possible solutions to the problems he discussed and points toward a different organizational as well as analytical culture within the Intelligence Community.
In summary this is a well done critique of the intelligence analyses in question which could have been written better. While mainly interesting to the professional reader concerned with intelligence analysis it does hold many interesting insights for readers interested in the historical events concerned and what part the intelligence estimates played.
August 13, 2022
Why Intelligence Fails: Lessons from the Iranian Revolution and the Iraq War by Robert Jervis reads a bit like apologia, but if it is, it is for a good cause. There is a lot of defense being played for intelligence and the art of intelligence in this book, as while he notes that there were mistakes and unchallenged assumptions, the judgments reached were not necessarily bad. Indeed, Jervis remarks that policy makers, when they complain about intelligence failures, do not actually desire better intelligence. Whether the intelligence stays too close to the facts, or whether the intelligence is supposed to make an argument and intuit from the facts, issues can arise. The vast majority of the text is devoted to Iran, with Iraq getting second billing at best, but this appears largely the result of Jervis himself having written a massive report on the supposed intelligence failures in the fall of the Shah. For Iran, he blames the idea that the Shah would crack down if he was in danger as a major impediment to good analysis, as it allowed the intelligence community to assume that things were going well until he did crack down, which rendered intelligence as a timely, actionable tool impossible as only during the worst moments when the Shah actually did crack down could the danger be seen. That the Shah also believed everything in his country that was going wrong was ultimately because of the US, and that he had his own plans for his liberalization program, suggests that the Shah may never have understood his danger until it was too late, and would necessarily have to give too much power to the military in order to survive. For Iraq, it was the curse of conventional wisdom, as truth was stranger than fiction.
The book is powerful, though occasionally repeats itself. The key lessons are to be taken and understood, even if it is perhaps not as readable as its flashier counterparts.
88/100
The book is powerful, though occasionally repeats itself. The key lessons are to be taken and understood, even if it is perhaps not as readable as its flashier counterparts.
88/100
February 2, 2017
Very interesting and somewhat disturbing book about the Intelligence community! Detailed study of the failures to detect the fall of the Shah of Iran and to detect that there were no WMD's in Iraq by an expert in identifying perceptions and misperceptions in the decision making process. Lengthy study of intelligence analysts and their process for creating the briefings about what is happening and will happen in the world. Somewhat dated about the fall of the Shah but apropos given similar failings in the process for shaping the intelligence briefings and analysis of the existence of Weapons of Mass Destruction in Iraq for caparison purposes. One observation I liked was Jervis's analogy to the Silver Blaze short story where the dog did nothing in the night time - the absence of the dog barking is just as important as is the dog barking. Similarly in intelligence,the absence of evidence may be just as important as evidence itself! A more tedious read but informative.
May 25, 2015
The author's central thesis 'Don't be surprised that the Intel world can't predict the future'. Policy maker's love to blame Intel for their faulty policies instead of taking any kind of responsibility themselves. They expect the intelligence community to be their crystal ball and then get upset when it doesn't work. The author then goes on to explain not only all the failures but the sheer difficulty of the task expected of them. The author doesn't really add much more to his thesis than that. Basically what is
September 1, 2011
This book covers some very important topics in the realm of dometic and foreign intelligence programs. But the author's writing style is very dry and dull. He does a poor job of capturing the reader's interest. Consequently, I had a very difficult time following this book and understanding some of the concepts due to the dullness of the writing. Perhaps I will give it another chance and try again at some point.....but I doubt it.
May 11, 2013
Robert Jervis examines two important U.S. intelligence lapses and tries to account for what went awry. After both, the CIA hired Mr. Jervis—a longtime student of international affairs—to help the agency sort out its mistakes. He thus brings an invaluable perspective as a smart outsider with sufficient inside access to appraise the agency's blind spots. An absorbing and important book that belongs in the library of anyone interested in the problems of intelligence.
June 14, 2012
Were these intelligence failures, or policy failures? The answer is, and perhaps always will be elusive. This book does lay out some possible solutions, and ex-posits on why neither of these failures had to be pre-ordained. Negative evidence and alternative analysis are two disciplines that have proven difficult to master, and not only in the social science of intelligence forecasting.
November 10, 2013
An excellent study of two pivotal moments in intelligence. Don't skip his intro or last chapter in which he writes in more general term about what ails intelligence.
March 12, 2024
another one of my 300 books goodreads lost my rating to
Displaying 1 - 19 of 19 reviews