On January 16, 1979, Shah Mohammad Reza Pahlavi left Iran, the country he had ruled since 1941. The extravagant leader spent decades attempting to modernize Iran, promote industrial growth, curry favour with the United States (US), and assert regional prowess. Yet just five months after the Central Intelligence Agency (CIA) declared “Iran is not in a revolutionary or even “prerevolutionary” situation,”^[i]the Shah was defeated in an Islamic surge that dumbfounded intelligence communities (ICs). Particularly after the US helped defeat nationalist Prime Minister Mohammad Mosaddeq and began a program of economic and military aid to Iran, the Shah had been a crucial US partner in Cold War geopolitics.^[ii] Iran bordered the Soviet Union and thus provided Americans with unparalleled surveillance capabilities – and Persian Gulf oil.^[iii] Yet by the mid-1970s, the Shah’s economic reforms (which had resulted in stagnation and rising unemployment), anti-religious stance, and penchant for repressive tactics met opposition from Islamic clergymen, landowners, and professionals alike.^[iv] Meanwhile, the charismatic Ayatollah Ruhollah Khomeini, exiled in Paris, “called for justice [and Sharia law] and blamed the woes of humankind on capitalist greed.”^[v]By the time US President Jimmy Carter began paying attention to the increasingly bloody riots and demonstrations, it was too late to prevent what Zbigniew Brzezinski, Carter’s National Security Advisor, called “the disintegration of Iran” and the creation of a “political vacuum” likely to be filled by those loyal to the Soviet Union.^[vi] Indeed the protagonists in this Persian tale wore sunglasses tinted by a Cold War mindset that proved constraining and inhibited pragmatism.

While many States were impacted by the fall of the Shah, it is the US intelligence community (IC) whose failure has received the greatest attention, most notably in a recently declassified post-mortem by Robert Jervis. The Iran case provides an opportunity to analyze failures at each stage of the intelligence cycle, namely tasking, collection, processing, analysis, and dissemination. This paper will outline how existing scholarship conceptualizes intelligence failures, and then will assess the Iran case along each stage of the intelligence cycle, emphasizing that many deficiencies were located at, and between, multiple stages. The discussion of collection failures will focus on the value human intelligence and contextual knowledge. Analytical failures will be explored with an emphasis on the US IC’s misunderstandings of both the Shah and the growing religious opposition. The dissemination section will highlight how the US IC and policymakers shared not only assumptions, but also an interest in maintaining strong relationships with a resilient Shah-led Iran. Both worlds had little desire to upset the balance of US-Iran relations during a time of aggressive US detente with the Soviet Union and engagement in the Middle East. A constant thread flowing through this part of the paper will be the question of whether more actionable intelligence would have been likely to influence policy outcomes.

The paper will conclude with an examination of the key lessons of the Iran intelligence failure, all of which are highly relevant at a time when governments are seeking to improve their ability to predict revolutions in general, and to understand the dynamics behind Iran’s ‘Green Movement’ in particular. Iran’s Islamic revolution and the monumental hostage crisis that followed constituted “America’s and the West’s first encounter with contemporary radical Islam,” launching an ill-defined war that continues to this day.^[vii] The actions and inactions of the US IC in 1978 and 1979 raise a number of critical questions, including whether revolutionary political movements are predictable phenomena, and more concretely, what reasonable expectations might be for ICs in times of upheaval. This paper will not explore US imagination or lack thereof with respect to possible post-Shah political arrangements in Iran, but rather will focus on the IC’s failure to predict the fall of the Shah.

Intelligence Failures

Jervis views intelligence failures as having two dimensions: an inevitable mismatch between estimates and reality, and “a falling short of what we expect from good intelligence.”^[viii] Mark Lowenthal’s definition is more specific, referring to failures of one or multiple stages of the intelligence cycle “to produce timely, accurate intelligence on an issue or event of importance to national interests.”^[ix]Whereas the above definitions place the onus for intelligence failure on ICs, Shulsky and Schmitt focus on policymakers, whose misunderstandings lead governments towards “inappropriate and counterproductive” actions.^[x]Dahl attributes responsibility to both parties, noting that intelligence failures result when ICs fail to produce what policymakers need or when the latter fail to respond appropriately to intelligence.^[xi] Others, by contrast, feel there should be a more clear “distinction between intelligence failures and policy failures.”^[xii] Dahl usefully outlines three schools of thought. The traditional school emphasizes failures of analysis in an environment of information overload. Intelligence failures are seen as inevitable and even natural.^[xiii] For Betts, analysts cannot escape their “motivated biases resulting from organizational or operational goals”^[xiv] and these biases distort perception.^[xv] The reformist school focuses on organizational and bureaucratic deficiencies. ICs can improve when such problems are corrected, however failures are exacerbated when “hierarchies, specialization, and centralization” limit coordination and information sharing.^[xvi]Finally, the contrarian school emphasizes collection failures — not a failure to connect the dots, but rather a lack of (actionable) dots.

Jervis orients himself somewhere between the traditional and reformist schools, given his focus on cognitive deficiencies that lead to analytical failures but that can be overcome through the use of more rigorous social scientific methods in ICs. Whereas post-mortem analyses of the September 11, 2001 attacks have highlighted specific bureaucratic and intergovernmental deficiencies, Jervis’ report investigates the entire intelligence process and therefore yields numerous and valuable lessons for the field.^[xvii] Intelligence failures occur when analysts do not “articulate their assumptions, subject these arguments to appropriate scrutiny, consider rival hypotheses that fit the evidence, test arguments by offering predictions, consider negative and positive evidence when evaluating assertions, and seek information that might disconfirm their existing point of view.”^[xviii] However, Lebovic questions whether ICs have the capacity for such rigorous social scientific inquiry and whether they should focus instead on providing “facts” to policymakers.^[xix] Yet in the Iran case, such narrow intelligence left the IC unable to answer important strategic questions about the conditions underlying political revolution and the role of religion in mobilizing dissent.^[xx] This raises the question of reasonable expectations for ICs and reaffirms the need for need for “empathy with the situation confronting analysts.”^[xxi]

The Intelligence Cycle^[xxii]

Tasking

As will be discussed in the dissemination section, there were significant US priorities and interests that limited intelligence tasking, which led to inadequate collection. Not only had the US scaled back intelligence on Iran as a sign of confidence in the Shah, but the US was also stretched thin as it sought to negotiate SALT II and the Israel-Egypt peace treaty simultaneously.^[xxiii] Iran was viewed “through the prism of its other regional concerns” rather than as a unique area of concern.^[xxiv] In its obsession with preventing the spread of communism, radical Islam was not understood as a transnational movement with the potential to unite diverse factions. Moreover, the US had generated a false sense of Iran as an ‘island of stability’ following the CIA-backed coup in 1953.^[xxv] The overall result was insufficient tasking within the US IC before and throughout 1978.

Collection

For Gary Sick, who served as a staffer in the US National Security Council, there was an “astonishing lack of hard information” about events in Iran, particularly the opposition movement.^[xxvi] The failure to collect sufficient actionable intelligence on Iran can be examined through a number of angles, as a failure of diplomatic reporting, a lack of human intelligence (HUMINT), or an overreliance on the Shah’s intelligence service. William Sullivan, US Ambassador to Iran, delivered deliberately soft reports to Washington, raising no concerns about the Shah’s vulnerability until two months before the revolution.^[xxvii] Whereas other intelligence reports noted the Shah’s depressed condition and indecisiveness, Sullivan would reassure Secretary of State Cyrus Vance that there was nothing to worry about.^[xxviii] Sick feels that Sullivan was guided by personal motives, as he “knew from experience that once the system was jolted into a crisis mode,” Washington would take over the US response, leaving him with limited control.^[xxix] This may be a lesson that HUMINT must not be left unverified or seen as above human self-preoccupation. Sullivan met with the Shah on an almost daily basis, making him an invaluable, yet eventually misleading, source of HUMINT. While many feel there was an overall lack of HUMINT on Iran, there is a disagreement as to whether more strategic or more tactical intelligence was needed. Dahl notes the overabundance of strategic analysis – mostly focused on broader Soviet threat perceptions.^[xxx] By contrast, Donovan, Jervis, Wirtz recall that the US was bombarded by requests for intelligence and was under pressure to provide detailed reports on the protests as well as economic developments, leading to a lack of strategic intelligence.^[xxxi] The HUMINT deficiency was twofold, involving a failure to utilize non-secret, open sources to better understand the demonstrators, and a lack of accurate intelligence on the state of the Shah – his health and his rule.^[xxxii] Had intelligence analysts approached “people in the streets, bazaars, and movies,” they could have discovered the “breath and depth of the hatred of the Shah, the power of nationalism, and the role of religious leaders as focal points.”^[xxxiii] Unlike the US, Israel had excellent HUMINT in Iran. Many of them Iranian Jews, Israeli agents closely interacted – in Farsi – with elites and opposition members alike, who felt comfortable sharing personal estimates over and above official positions.^[xxxiv] Israeli agents read local newspapers, participated in demonstrations under cover, and were thus able to “sense (rather than systematically analyze) the revolutionary atmosphere” and anti-Shah sentiment.^[xxxv] By contrast, the CIA had almost no Arabic or Farsi speakers on staff.^[xxxvi]

The US IC’s collection failure was exacerbated by its reliance on the Shah’s internal intelligence unit, the SAVAK. With the exception of Ambassador Sullivan’s (flawed) reporting, the US was dependent on the Shah’s officials, leaving Washington unaware of the monarch’s debilitating illness or of deepening support for Khomeini.^[xxxvii] As Jervis emphasizes, SAVAK did not provide misleading information on the opposition, but rather gave no information at all.^[xxxviii] At a time when the US IC was overstretched, Washington’s reliance on the Shah eliminated its incentive to continue independent intelligence collection.^[xxxix] Yet the US learned a key lesson: in a society where citizens are unable to express their grievances through official channels to which leaders are receptive, ICs should look to monitor informal or non-state channels.^[xl] This case raises critical questions about the suitability of spying on ‘friends.’ The US scaled back intelligence collection as a sign of trust in the Shah, fully aware that the he was suspicious of any US contacts with opposition groups.^[xli] Yet the resulting intelligence failure hindered US strategic interests over the longer term.

Processing

A key lesson of the Iran case is that on-the-ground knowledge is futile unless it is properly integrated with intelligence being produced at headquarters. Consistent with the reformist school, the processing stage is where collaboration amongst agencies is critical. The CIA, State Department, Defense Department rarely shared information on Iran.^[xlii]

Analysis

At the core of the Iran intelligence failure is a series of four unchecked assumptions and prefabricated beliefs combined with an absence of methodological rigour. Failure to ‘connect the dots’ left the US unable to advance its interests.^[xliii] However, when examining analytical failures, one must be attentive to hindsight bias and the problem of noise – the challenge of discerning warning signals amid chaos.^[xliv]

a) The Shah would crack down, and since he did not, the situation was not dire:
The US IC adopted a “nonfalsifiable proposition” through its belief that the Shah would crack down on any serious instability and the fact that he had not done so was accepted as evidence that the Shah’s control over Iran was stable.^[xlv] As explained below, the IC had valid reasons to assume the Shah would respond with force, in light of his “well-entrenched regime, loyal military, and draconian security services.”^[xlvi] The disproval of the logical assumption that the Shah would not be overthrown reinforces the need for ICs to systematically assess alternate possibilities. This would have been difficult, however, since the assumption could only be disproved once the Shah’s fall became imminent. The the IC could therefore offer policymakers little more than an “eleventh-hour warning.”^[xlvii]

b) The Shah was powerful and decisive and would respond with appropriate vigour:
This assumption earned its reliability through CIA assessments explaining how the Shah, by implementing economic and social reforms, had significantly bolstered the monarchy and incapacitated the highly fragmented opposition.^[xlviii] Although they recognized that troop loyalty was a concern, the CIA reasoned that the Shah was capable of exerting authority while relying on the public support he had supposedly generated through modernization.^[xlix] Indeed some in Washington felt the Shah’s false and fleeting strength was the result of an economic boom that proved temporary.^[l]

The Shah’s decision not to deploy massive force to quell demonstrations remains a mystery.^[li] The CIA failed to penetrate the Shah’s mental world and failed to realize that Washington’s messaging was dizzyingly contradictory for the Shah. He was being encouraged to respond to protests with strong force while also maintaining liberalization programs.^[lii] In attempting to form a compromise between Vance and Brzezinski, Carter told the Shah he supported a coalition or military government in order to restore order, after which liberalization should continue.^[liii] Here the US assumed that a weakened leader could restore order through coercive means, even though the Shah was most concerned with seamlessly transferring power to his son, further discouraging the Shah from using military force against demonstrators.^[liv] The Shah’s terminal cancer likely also affected his decision making, yet explanations like these were not systematically considered in the IC’s analyses.^[lv] After all, Carter’s administration did not learn of the Shah’s cancer until late 1979.^[lvi] With this knowledge, Brzezinski might have given up advocating for a forceful response, the US might have worked to bolster political moderates in resisting Khomeini, and the US, like Israel, could have developed a plan to protect Americans in Iran and possibly avoid the subsequent hostage crisis.^[lvii] It is important to note that Jervis was similarly unaware of the Shah’s illness while writing his post-mortem.^[lviii]

The US conspicuously failed to act even as intelligence increasingly pointed to the Shah’s growing weakness. Diplomatic reporting was partly at fault. After the State Department characterized the Shah as “out of touch with the current scene”^[lix] and the embassy discovered plans for cutbacks in military purchases^[lx], Ambassador Sullivan would submit an “upbeat prognosis,” and the cycle of warning and reassurance would continue.^[lxi]

c) The role of religion was miniscule nor was Khomeini a significant concern:
A crucial question is why the US IC paid so little attention to the role of religion in fuelling the revolution under the spiritual guidance Khomeini. The problem was not merely a lack of tactical intelligence, but rather a system-wide failure to understand the role of religious leaders in responding to a population’s grievances.^[lxii] Not until November 1978 did the CIA – nor the Shah – realize the charismatic ayatollah’s success at amassing public support through “his instinctive understanding of publicity and propaganda.”^[lxiii] One must acknowledge that the religious nature of the Iranian uprising posed a major hurdle for ICs, which relied on increasingly obsolete models of secular revolutions and which were skeptical of religion as force for uniting disparate opposition elements.^[lxiv] CIA reports and the June 1978 National Intelligence Estimate reveal that the IC viewed the opposition as a “liberal, modernizing, middle class” reacting to “rapid social and economic changes that had outpaced Iran’s political development.”^[lxv] In other words, this was about correcting a democratic deficit rather than returning to Islamic principles.^[lxvi] Khomeini’s ideas about “the clerical management of political institutions” according to Sharia law were dismissed, just as some analysts predicted that clergy “would never participate directly in the formal governmental structure.”^[lxvii] The tension between secular and religious opposition elements led in part to a failure to recognize and predict the revolution in its initial phase.^[lxviii]

Yet there were signs and signals, and Washington should have been alarmed by reports that demonstrators increasingly included members of the professional middle class, rather than usual suspects like Islamic conservatives and students. The National Intelligence Daily went further, noting the possibility of “an alliance between the moderate left and the extreme Muslim right” grounded solely in their opposition to the Shah.”^[lxix] Crucially, the US failed to recognize the increasing role of the urban working class in pro-Khomeini demonstrations.^[lxx] There were tactical intelligence reports of ceremonies and sermons that – using religious symbolism involving Hussein, the founder of Shia Islam – were “explicit in their call for the Shah’s downfall,” yet these reports fell largely on deaf ears.^[lxxi] Furthermore, the CIA failed to consider multiple possibilities, assuming only a coalition government could regain public trust and discounting an Islamic revolution.^[lxxii] By late September, the CIA began to recognize Khomeini’s mass influence and his ingenious use of local religious leaders in mobilizing protestors, yet it remained convinced the Shah would stay in power in the earlier 1980s and possibly be replaced by the moderate opposition.^[lxxiii] By contrast, Israel was firm in its analysis of the Islamist opposition’s predominance and the growing irrelevance of the secular National Front.^[lxxiv]

d) Iranian nationalism and anti-Americanism played no role in the revolution:
The US failed to recognize that nationalism and anti-Americanism were intertwined in pre-revolutionary Iran.^[lxxv] A primary reason for Khomeini’s popularity was his appeal to nationalistic tendencies in opposition to a regime led by a tyrant corrupted by America.^[lxxvi] Iranians did not forget US support for the 1953 coup and they continued to see the Shah as an “American puppet.”^[lxxvii] There were graffiti and leaflets calling for the Shah’s death, but no US intelligence, partly because Washington underestimated the extent to which Iranians viewed, with annoyance, their country as a US ‘client state.’^[lxxviii] Iranians saw the 1953 coup “as an indicator of U.S. hostility toward Iran and of U.S. proclivity to trample on the rights and prerogatives of Iranians.”^[lxxix] Another reason why intelligence was so sparse was the US’ failure to recognize that anti-Americanism was being directed at the Shah becausethey saw him as an American puppet. Ignoring the fact that the antagonism was not merely anti-Shah, Washington thus paid little attention to its new detractors.

Dissemination

The process of disseminating intelligence to relevant parties is naturally subject to forms of politicization. Examining Gregory Treverton’s five forms of politicization, ranging from “direct pressure” to “house line,” the Iran case can be seen as an example of “shared mindset,” whereby strong assumptions are shared by the IC and policymakers.^[lxxx] There was a shared mindset with respect to the aforementioned four assumptions, as well as a shared interest in conveying the Shah’s continued strength. Yet the US’ long-held appreciation of the Shah as a vital anti-Soviet actor led to intelligence failure and “deafened policy makers to the warnings implicit in available intelligence.”^[lxxxi] The US IC’s strategic relationship with the Shah inflated confidence in the durability of his regime at all stages of the intelligence cycle.^[lxxxii] Policymaker receptivity – the willingness of decision makers to act on the basis of intelligence – requires “a combination of belief in the threat and trust in intelligence.”^[lxxxiii] The Carter administration, the IC, and the State Department all “had acquired a vested interest in the status quo” and none wanted to be the first to signal the Shah’s impending fall.^[lxxxiv] Once the State Department and the CIA began reporting on the Shah’s vulnerability, Brzezinski refused to act upon this intelligence and he continued to believe the Iranian leader would apply mass force and remain a Cold War ally.^[lxxxv] The IC also faintly hoped the Shah would remain prominent in a transition to “a government that will (if we are lucky) respond to US interests.”^[lxxxvi] For Jervis, this policymaker-intelligence interface was the site of intelligence failure consistent with Betts’ model, whereby self-deluded policymakers ignore “facts that fail to conform to their Weltanschauung.”^[lxxxvii] In addition to Soviet containment, Washington was also focused on securing an Israeli-Egypt peace treaty and was unenthusiastic about admitting that one of its key Middle East allies was facing an existential threat. Indeed, challenging US strategy “put individuals’ careers at risk.”^[lxxxviii] Dissemination failure was not restricted to the US IC. Uri Lubrani, Israel’s representative in Iran, tried in vain to share with Washington his reports, which, in contrast to Sullivan’s, predicted the Shah’s downfall andwere closely reviewed by Israeli Prime Minister Yitzchak Rabin, who, although he was preoccupied with Camp David, recognized his government had a vested interest in securing the Shah and gave the issue appropriate attention.^[lxxxix] Furthermore, the lack of intelligence cooperation with France remains baffling and reflects Washington’s underestimation of Khomeini.

Conclusion: Learning from Failure

The failure to predict the Islamic revolution in Iran offers several lessons that remain relevant today in the aftermath of the ‘Arab Spring.’ ICs have not arrived at criteria for political revolutions, nor have they established whether revolutions are predictable phenomena at all. The problem was not that Carter was disinterested, but rather that he voiced his concern too late and by then was mostly focused on avoiding future intelligence failures. Carter demanded that his IC chiefs assess the failure, though in doing so he embarrassed Stansfield Turner, Director of Central Intelligence, and merely deepened rifts in Washington.^[xc] This “set off a scramble of defensive countermoves and self-justifications” that detracted from efforts to improve the IC’s capabilities.^[xci] Carter quickly became embroiled in the Tehran hostage crisis, and his IC had little capacity to implement Jervis’ recommendations, all of which remain topical. Jervis’ report ended with a “dual appeal” for analysts to reassess their assumptions and for managers to “create an environment conducive to analyzing foreign affairs, not just reporting them.”^[xcii] He calls on ICs to use more rigorous social sciences methods, such as peer review, though he overestimates the ability of ICs to measure intelligence failure.^[xciii] Jervis’ recommendations range in feasibility, and the US evidently had too few analysts specializing in Iran to allow for peer review or rigorous analysis of strategic intelligence.^[xciv] He is correct that alternative explanations must be presented and considered. Since they must defend their views, analysts are more likely to be thorough, systematic, and conscious of biases or subjective beliefs, such as the Shah’s durability or the capacity of religion to fuel revolution.^[xcv] Sharpened analysis could enable agents to mark indicators for confirming or denying predictions. For example, they could have set a threshold level of violence after which they would have committed to revising estimates of the Shah’s strength.^[xcvi] However, Jervis is quick to note that this style of competitive analysis must be valued and rewarded in ICs in order to bring about meaningful change.

Other proposals are arguably too ‘academic’ for time-pressured intelligence practitioners, such as the call for a pre-mortem analysis.^[xcvii] As one critique of Jervis put it, “the intelligence business model now thrives on speed more than accuracy,” where breaking the news is more valued than in-depth analysis, let alone peer review.^[xcviii] Donovan contradicts Jervis, noting that there was ample debate and disagreement over how the Iran crisis should be framed, and the resulting delays in releasing reports brought about the intelligence failure. Either way, the failure to predict the fall of the Shah woke up the intelligence and policy communities to the growing threat of religious revolution in general, and charismatic Islamism in particular. The CIA was restructured, and new approaches to collection and analysis were introduced. What remains unclear is whether enhanced HUMINT and more rigorous analysis would have changed policy outcomes in light of the absence of policymaker receptivity combined with the shared mindset that left the IC, the White House, and the US government (including its embassy in Tehran) equally unwilling to make the case for the decline of a stalwart US ally. Moreover, the unprecedented post-1978 “institutional response to concerns about regional instability, the forces of nationalism, and the growth of terrorism” evidently failed to secure the US IC’s capacity to predict political revolutions.^[xcix] It is not apparent whether such foresight will ever be within reach and if not, whether intelligence communities can be reasonably expected to achieve the impossible. The hostage crisis that followed the Iran intelligence failure leaves little doubt that the stakes could not be higher.

Bibliography

Bar-Joseph, U. (2010), Forecasting a Hurricane: Israeli and American Estimations of the Khomeini Revolution, Journal of Strategic Studies, 36(5), 718–742.

Connell, M. (2010). Why Intelligence Fails: Lessons from the Iranian Revolution and the Iraq War (Review). The Middle East Journal, 64(4), 677-678.

Dahl, E.J. (2013). Intelligence and Surprise Attack, Failure and Success from Pearl Harbor to 9/11 and Beyond (pp. 7), Georgetown University Press.

DIA report (1978), ‘Assessment of the political situation in Iran, 9/30/78’, NSA, Vol. 1, 47.

Donovan, M. (1997), National intelligence and the Iranian revolution, Intelligence and National Security, 12(1), p.143-163

Etges, A. (2011). All that glitters is not gold: The 1953 coup against Mohammed Mossadegh in Iran. Intelligence and National Security, 26(4), 495-508.

Federation of American Scientists, “The Intelligence Cycle,” Retrieved from https://www.fas.org/irp/cia/product/facttell/intcycle.htm.

George, R.Z. (2013). Robert Jervis, Why Intelligence Fails: Lessons from the Iranian Revolution and the Iraq War. Intelligence and National Security. 28(5), 761-765.

Gill, P. and Phythian, M. (2006). Intelligence in an Insecure World. Polity Press.
Hedley, J.H. (2005), Learning from Intelligence Failures, International Journal of Intelligence and CounterIntelligence, 18, 435–450.

Jervis, R. (2010). Why Intelligence Fails: Lessons from the Iranian Revolution and the Iraq War.New York: Cornell University Press.

Lebovic, J. (2010) “Why Intelligence Fails: Lessons from the Iranian Revolution and the Iraq War” Review), Critical Dialogue 8(4): 1167-1169.

Leffler, M.P. (2008), For the Soul of Mankind: The United States, the Soviet Union, and the Cold War. New York: Hill and Wang.

Litwak, R. (2011). Why intelligence fails: Lessons from the Iranian Revolution and the Iraq War (Review). Journal of Cold War Studies, 13(1), 223-225.

Lowenthal, M. (1985). The Burdensome Concept of Failure. In A.C. Maurer, M.D. Turnstall, and J.M. Keagle (Eds.), Intelligence: Policy and Process (pp. 43-56). Boulder, CO: Westview.

MacEachin, D. and Nolan, J.E. (2004), ‘Iran: Intelligence Failure or Policy Stalemate?’ Institute for the Study of Diplomacy, Edmund A. Walsh School of Foreign Service, Georgetown Univ., Working Group Report No. 1.

Marrin, S. (2012). Evaluating the quality of intelligence analysis: By what (mis) measure? Intelligence and National Security, 27(6), 896-912.

McDermott, R. (2007). The Use and Abuse of Medical Intelligence. Intelligence and National Security, 22(4), 491-520.

Munton, D. and Matejova, M. (2012). Spies without borders? Western intelligence liaison, the Tehran hostage affair and Iran’s Islamic revolution, Intelligence and National Security, 27(5), 739-760.

Murray, D. (2010). The carcass of dead policies: Lessons for Obama in dealing with Iran. Contemporary Politics, 16(2), 209-223.

Pillar, P.R. (2010), The role of villain: Iran and U.S. foreign policy, Political Science Quarterly, 128(2), 211.

Prillaman, W. C., and Dempsey, M. P. (2004). Mything the point: What’s wrong with the conventional wisdom about the C.I.A., Intelligence and National Security, 19(1), 1-28.

Shulsky, A.N. and Schmitt, G.J. (2002). Silent Warfare: Understanding the World of Intelligence. Potomac Books.

Sick, G. (1985), All Fall Down: America’s Tragic Encounter With Iran. New York: Random House.

US CIA, National Intelligence Daily Article (1978), ‘Iran: Shah’s Strategy’, 13 Nov. 1978 in ‘Making of US Policy’, doc. 01728.

US House of Representatives, Subcommittee on Evaluation, Permanent Select Committee on Intelligence, Staff Report (1979), Iran: Evaluation of U.S. Intelligence Performance Prior to November 1978 (Washington DC: GPO).

Welch, D.A. (2011). Why Intelligence Fails: Lessons from the Iranian revolution and the Iraq war – by Robert Jervis. Political Psychology, 32(4), 717-720.

Wirtz, J.J. (2014), The Art of the Intelligence Autopsy, Intelligence and National Security, 29(1), 1-18.

^[i] US House of Representatives, Subcommittee on Evaluation, Permanent Select Committee on Intelligence, Staff Report, Iran: Evaluation of U.S. Intelligence Performance Prior to November 1978 (Washington DC: G P O 1979): 5.

^[ii] Pillar, P.R. (2010), The role of villain: Iran and U.S. foreign policy, Political Science Quarterly, 128(2), 211.

^[iii] Leffler, M.P. (2008), For the Soul of Mankind: The United States, the Soviet Union, and the Cold War. New York: Hill and Wang: 299.

^[iv] Ibid. 300.

^[v] Ibid.

^[vi] Ibid. 301.

^[vii] Munton, D. and Matejova, M. (2012). Spies without borders? Western intelligence liaison, the Tehran hostage affair and Iran’s Islamic revolution, Intelligence and National Security, 27(5), 739-760.

^[viii] Jervis, R. (2010). Why Intelligence Fails: Lessons from the Iranian Revolution and the Iraq War.New York: Cornell University Press: 2.

^[ix] Lowenthal, M. (1985). The Burdensome Concept of Failure. In A.C. Maurer, M.D. Turnstall, and J.M. Keagle (Eds.), Intelligence: Policy and Process (pp. 51). Boulder, CO: Westview.

^[x] Shulsky, A.N. and Schmitt, G.J. (2002). Silent Warfare: Understanding the World of Intelligence (pp. 63). Potomac Books. One should not assume, however, that accurate understanding of situations necessarily (or even usually) leads to productive and appropriate actions by governments and militaries.

^[xi] Dahl, E.J. (2013), Intelligence and Surprise Attack, Failure and Success from Pearl Harbor to 9/11 and Beyond (pp. 7), Georgetown University Press.

^[xii] Prillaman, W. C., and Dempsey, M. P. (2004). Mything the point: What’s wrong with the conventional wisdom about the C.I.A., Intelligence and National Security, 19(1), 1-28.

^[xiii] Wohlstetter, R. (1962). Pearl Harbor: Warning and Decision. Stanford University Press.

^[xiv] Marrin, S. (2012). Evaluating the quality of intelligence analysis: By what (mis) measure? Intelligence and National Security, 27(6), 901.

^[xv] Ibid.

^[xvi] Dahl 11-12.

^[xvii] Wirtz, J.J. (2014), The Art of the Intelligence Autopsy, Intelligence and National Security, 29(1): 2.

^[xviii] Lebovic, J. (2010) Why Intelligence Fails: Lessons from the Iranian Revolution and the Iraq War (Review), Critical Dialogue, 8(4): 1167.

^[xix] Ibid. 1169.

^[xx] Ibid.

^[xxi] Wirtz 3.

^[xxii] Federation of American Scientists, “The Intelligence Cycle,” Retrieved from https://www.fas.org/irp/cia/product/facttell/intcycle.htm.

^[xxiii] Donovan, M. (1997). National intelligence and the Iranian revolution, Intelligence and National Security, 12(1), 149.

^[xxiv] Murray, D. (2010). The carcass of dead policies: Lessons for Obama in dealing with Iran. Contemporary Politics, 16(2), 209-223.

^[xxv] Prillaman and Dempsey.

^[xxvi] Sick, G. (1985), All Fall Down: America’s Tragic Encounter With Iran (pp. 90). Random House.

^[xxvii] It is important to note that Jervis’ postmortem did not include a November 9 telegram from Ambassador Sullivan that “marked the day when the administration came to terms with the possibility that the end of the Shah’s rule was close.” Bar-Joseph, U. (2010), Forecasting a Hurricane: Israeli and American Estimations of the Khomeini Revolution, Journal of Strategic Studies, 36(5), 735.

^[xxviii] Sick 53.

^[xxix] Ibid. 66.

^[xxx] Dahl 22.

^[xxxi] Donovan 159, Jervis 67, and Wirtz 9.

^[xxxii] McDermott, R. (2007). The Use and Abuse of Medical Intelligence. Intelligence and National Security, 22(4), 491-520.

^[xxxiii] Jervis 26. It is important to note that Jervis was required by the CIA to focus on analytical rather than collection failures in his postmortem, yet he still managed to shed light on problems with collection.
Also see Litwak, R. (2011). Why intelligence fails: Lessons from the Iranian Revolution and the Iraq War (Review). Journal of Cold War Studies, 13(1), 223.
Also see Welch, D. A. (2011). Why intelligence fails: Lessons from the Iranian Revolution and the Iraq War – by Robert Jervis. Political Psychology, 32(4), 718.

^[xxxiv] Ben-Joseph 721.

^[xxxv] Ibid. 739.

^[xxxvi] Prillaman and Dempsey 6.

^[xxxvii] McDermott 500. By contrast, in Israel senior policymakers were being told in early 1978 of the Shah’s declining popularity (Bar-Joseph 728).

^[xxxviii] Jervis 17.

^[xxxix] Sick 38

^[xl] Ibid 161.

^[xli] Sick 32. Also see Donovan 159.

^[xlii] Connell, M. (2010). Why intelligence fails: Lessons from the Iranian Revolution and the Iraq War (Review). The Middle East Journal, 64(4), 677-678.

^[xliii] Gill, P. and Phythian, M. (2006). Intelligence in an Insecure World (pp. 152). Polity Press.
Jervis would disagree, noting such thoughts just add to myth of Iran as an American puppet.

^[xliv] Wirtz 5.

^[xlv] Jervis 190, Lebovic 1168, Litwak 224, and Welch 718.

^[xlvi] Connell 677.

^[xlvii] Connell 677, Jervis 190, and Wirtz 9.

^[xlviii] Wirtz 8.

^[xlix] Donovan 145, 148.

^[l] Sick 97.

^[li] Litwak 224.

^[lii] Jervis 32.

^[liii] Sick 68. Also see Donovan 159.

^[liv] Sick 77. The Shah’s speech on November 6, 1978 “should have provided incontrovertible evidence that the shah was incapable of adopting and sustaining the kind of hard line that many were pressing on him.”

^[lv] Jervis 31. See also McDermott 500, “the psychological effect of his diagnosis may have ultimately proved more destructive politically for the Shah than his physical disabilities.”

^[lvi] McDermott 496.

^[lvii] Ibid. 499.

^[lviii] Wirtz 8.

^[lix] Donovan 150.

^[lx] Sick 56.

^[lxi] Ibid. 53.

^[lxii] Jervis 91.

^[lxiii] Sick 58. Also see US CIA, National Intelligence Daily Article, ‘Iran: Shah’s Strategy’, 13 Nov. 1978 in
‘Making of US Policy’, doc. 01728. (quoted in Donovan 152).

^[lxiv] Sick 167.

^[lxv] Jervis 23.

^[lxvi] Wirtz 9.

^[lxvii] Ibid. 113, 164.

^[lxviii] Sick 164.

^[lxix] Donovan 145.

^[lxx] Ibid. 146.

^[lxxi] Ibid. 153

^[lxxii] Ibid. 155. Quoted from US CIA, Alert Memo., ‘Iran – Prospects for Moharram’, 29 Nov. 1978, doc. 01818.
^[lxxiii] Bar-Joseph 733. Quoted from DIA report ‘Assessment of the political situation in Iran, 9/30/78’, NSA, Vol. 1, 47.

^[lxxiv] Bar-Joseph 729.

^[lxxv] Welch 718.

^[lxxvi] Jervis 87.

^[lxxvii] Ibid. 89.

^[lxxviii] Etges, A. (2011). All that glitters is not gold: The 1953 coup against Mohammed Mossadegh in Iran. Intelligence and National Security, 26(4), 505.

^[lxxix] Pillar 211.

^[lxxx] Gill and Phytian 153.

^[lxxxi] Donovan 160.

^[lxxxii] Gill and Phythian 152.

^[lxxxiii] Dahl 23.

^[lxxxiv] Sick 42. Also see Gill and Phythian 152.

^[lxxxv] Donovan 153, 160. As Robert Bowie, the former head of the National Foreign Assessment Center, put it, “I think certainly by September ’78 we [the CIA] had a better grasp of the situation than the policy establishment, but we were providing intelligence they were not necessarily interested in using.”

^[lxxxvi] Bar-Joseph 734, Quoted from Dept. of State paper, ‘Iran Political Assessment’, Washington, 10/18/78, NSA, Vol. 1, 47. 01602.

^[lxxxvii] Gill and Phythian 149.

^[lxxxviii] Bar-Joseph 738, Quoted from MacEachin, D. and Nolan, J.E. (2004), ‘Iran: Intelligence Failure or Policy Stalemate?’ Institute for the Study of Diplomacy, Edmund A. Walsh School of Foreign Service, Georgetown Univ., Working Group Report No. 1.

^[lxxxix] Ibid. 720, 733. Ironically, “While the Israeli preparations were underway, President Carter and his aides focused their attention on the Egyptian-Israeli peace talks at Camp David. The Egyptians and the Israelis were more tuned to the events in Iran”. Also see Sick 37. The problem of noise remains relevant here in assessing US reluctance to receive the Israeli report.

^[xc] Donovan 151.

^[xci] Sick 91.

^[xcii] Jervis 108.

^[xciii] Welch 720.

^[xciv] Wirtz 11.

^[xcv] Jervis 48.

^[xcvi] Ibid.

^[xcvii] Ibid. 192.

^[xcviii] George, R.Z. (2013). Robert Jervis, Why Intelligence Fails: Lessons from the Iranian Revolution and the Iraq War. Intelligence and National Security. 28(5), 764. Also see Connell 678-679.

^[xcix] Hedley, J.H. (2005), Learning from Intelligence Failures, International Journal of Intelligence and CounterIntelligence, 18, 435.

—
Written by: Adam Moscoe
Written at: University of Ottawa
Written for: Greg Fyffe and Wesley Wark
Date written: March 2014

Further Reading on E-International Relations

• Policy Failure and Unipolarity on the Eve of Operation Desert Storm
• The Gouzenko Affair and the Development of Canadian Intelligence
• On the Role of Intelligence in the Pacific Theater of the Second World War
• The Puzzle of U.S. Foreign Policy Revision Regarding Iran’s Nuclear Program
• The UN Failure in Yugoslavia: Lessons from Canadian Peacekeeping
• Can the Use of Torture in Intelligence Gathering Be Justified?