A Stunning Intelligence Failure

Paul Monk
2002

 

Failure to detect and abort the plot that led to destruction of the World Trade Centre (WTC) and part of the Pentagon, on 11 September, has been described as the worst US intelligence failure since Pearl Harbor. Yet the United States established its vast intelligence system, from 1947, precisely in order to prevent future 'Pearl Harbors'. What went wrong? Remarkably, it was almost exactly what went wrong in the lead-up to Pearl Harbor. But this is not a weird coincidence, nor is it an aberration.

In fact, failures of this nature are endemic in the world of intelligence. It's time their endemic nature was better understood. It can be. Both the problems and their solutions have certain features in common with intelligence failures in the economic world. We have a lot to gain from looking at these common themes in intelligence failure - then systematically addressing the causes.

First, consider the case of 11 September. Intelligence officials in the United States have said that they "never imagined" that the terrorist groups they had been monitoring "had the ability to kill thousands of people in coordinated attacks on the American homeland." One senior intelligence officer declared, "It is now clear that there are some things that should have rung bells a little bit louder. I wish we had paid more attention."

Never imagined? Some things should have rung bells a little bit louder? These are incontestably the confessions of intelligence failure. Consider, however, what the "things" were which should have rung bells a little bit louder, especially at CIA HQ, Langley. In March 1993, an attempt was made by Islamic militants to blow up the WTC using a truck bomb. The attempt was only partially successful, but those responsible vowed that next time they would finish the job. The former deputy director of the CIA's Counter-Terrorist Centre has stated that the 1993 WTC bombing began to "raise public concern about terrorism". Yet somehow it slipped out of focus at CIA itself, it would appear. Shouldn't the WTC should have been considered a primary target, when reports came in, during the months prior to 11 September, of plans by Osama bin Laden's al-Qaeda (AQ) group for a big hit on the United States?

And if the WTC had been considered a target, what sort of attack on it might at least have been imagined? Consider that, in 1994, Tom Clancy, the author of Patriot Games and Clear and Present Danger, wrote a blockbuster novel which featured the following scenario. Captain Torajiro Sato, former Japanese fighter pilot, as part of an anti-American Japanese plot called (of all things) Operation Kabul, takes a civilian airliner and flies it straight into the US Congress in Washington DC, with a full payload of jet fuel, killing nine hundred people and demolishing the entire Capitol building. Did this ring no bells this northern summmer? Did no-one in CIA or FBI imagine this might be something AQ would attempt - perhaps against the WTC?

But there was no need to depend on fiction to prompt the imagination. A matter of months after the publication of Clancy's novel, in January 1995, a raid by Philippine police in Manila led to the arrest of one Abdul Hakim Murad, an Islamic terrorist and trained pilot with links to Osama bin Laden. He was wanted for involvement in the March 1993 attack on the WTC. When he was arrested, floppy disks were found on the premises - Room 603 Josefa Apartment Building, Quirino Ave, Malate, Manila - detailing a spectacular terrorist plot called Project Bojinka.

Project Bojinka was an AQ plan to hijack eleven airliners simultaneously, exploding many of them at various places over the Pacific, but flying at least two of them into major federal government buildings in the United States. The flights to be hijacked were specified. They were all United Airlines, Northwest Airlines and Delta flights. Two of the targets, Murad confessed during his interrogation, were the CIA HQ and the FBI HQ. The plan has been masterminded by one Ramzi Youssef, who was arrested in Islamabad in the wake of Murad's interrogation. Both Murad and Youssef were extradited to the United States, tried and convicted for complicity in the 1993 attack on the WTC.

The date of Youssef's conviction was 11 September 1996. From that point, given the fascination terrorists have with anniversaries, 11 September should surely have become a watch date. There seems to be no evidence that it did. Indeed, none of these things seem to have rung any bells at Langley in the northern summer of this year.

What went wrong is not, in fact, particularly mysterious. We have here, as it happens, a paradigm case of intelligence failure. The parallels with the Pearl Harbor case are striking. In the months before Pearl Harbor, there was a clear and increasing danger of conflict with Japan. In the months before 11 September there was a clear and increasing danger of a major strike against America by AQ. Well before Pearl Harbor, Japan had demonstrated a propensity to start a war with a surprise attack on its enemy's main naval forces. Eight years before 11 September, AQ had demonstrated a propensity to attack major targets on the US mainland, specifically the WTC.

American naval exercises well before December 1941 had shown that a surprise carrier-based air attack on Pearl Harbor was feasible. Project Bojinka showed that AQ believed it feasible to use airliners as missiles to destroy major US public buildings. The scenario for a Japanese attack on the United States had been fictionalised years before, by Hector Bywater in The Great Pacific War. Project Bojinka and the 11 September attacks were prefigured by Tom Clancy, in Debt of Honour. Yet, when the attacks came, political leaders and intelligence officers, in both 1941 and 2001, were stunned and professed never to have imagined that this could have happened.

In a memorandum circulated following the 11 September attacks, CIA Director George Tenet is reported to have blamed inter-agency turf squabbles, bureaucratic rigidity, legal restraints on the recruitment of human intelligence sources abroad and poor coordination in intelligence sharing for the catastrophic intelligence failure that had just occurred. Yet the Central Intelligence Agency was created after the Second World War precisely in order to overcome these sorts of problems.

Failure of imagination is missing from George Tenet's list of problems, as is failure of institutional memory. Given the antecedents, the threats emanating from AQ and the resources available to CIA and FBI, surely it staggers the imagination to think that what happened on 11 September caught these agencies and those to whom they report by surprise? Why was Project Bojinka filed and forgotten after 1995-96?

To get this puzzle in deeper perspective, it is important to consider that the six decades between Pearl Harbor and 11 September 2001 were littered with intelligence failures of a similar nature. Not just in the United States and not just in matters of military security. In an essay published last year, Paul Dibb made the extraordinary observation that the Western intelligence services in general, specifically including Australia's, had found it "impossible to predict" the collapse of Soviet power in Eastern Europe, the reunification of Germany, the demise of the Soviet Union itself, "the Asian economic crisis and how it spread to Russia and Brazil; the collapse of the Suharto regime in Indonesia; the acquisition by India, Pakistan and North Korea of nuclear weapons and ballistic missiles" and, in a word, virtually every important geopolitical development of our time.

Dibb may not be wholly correct in making this claim, but he is well placed to know. He had been head of intelligence assessments, then deputy director and finally director of Australia's Joint Intelligence Organisation between 1975 and 1989, then Deputy Secretary of Defence for Strategy and Intelligence, in the early 1990s. His truly is an insider's of what makes intelligence organisations malfunction.

The thing to focus on here, however, is not the problems of Australian intelligence during Paul Dibb's career, but the problems endemic in the world of intelligence at large. One has only to read the memoirs of, say, East Germany's Markus Wolf or the KGB's Oleg Kalugin , for example, to realise that Paul Dibb's experience has a remarkable amount in common with that of Communist spymasters. It is the common themes that need to be understood - and rather urgently.

We can, however, widen the comparison further by considering the failures of intelligence in the marketplace alongside those in the world of government and international security. Such a comparison might begin with a simple list of similarities between the intelligence problem in the government and business fields. Here are nine possible points of comparison.

First. Most intelligence analysts fail to anticipate major geopolitical changes. Most institutional investors fail to beat the market.

Second. It is arguable that lack of critical thinking is largely responsible for geopolitical intelligence failures. It is also, I suggest, largely responsible for the poor performance of most professional investors. Logical fallacies too easily infect analysis in both domains and go unchallenged. The e-tech stocks bubble, for example, powerfully suggests that critical thinking was seriously laggard in all too many investment houses and entrepreneurial companies in the late 1990s.

Third. Both intelligence analysts and market analysts are dealing with complex systems, with hundreds of interrelated factors; making straightforward prediction all but impossible. Both commonly use this as an alibi for failure. In both fields fatalists argue that it is impossible to make accurate predictions, to "beat the market", as they say in Wall Street terminology. This claim would be more persuasive, in both fields, if there was good evidence that institutions actually try to do so in a systematic fashion. In reality, too many of them are guided by executive intuition, or what economic historian Graeme Snooks calls "strategic imitation" - ie running with the conventional wisdom.

Fourth. Given such complexity, one would expect both institutional systems to work according to procedures calculated to overcome complexity and minimise error, by cross-examining assumptions, weighing conflicting evidence and developing state of the art institutional memory. Certainly there are various efforts to do this, but the problems clearly run deep.

Fifth. In sombre fact, 'market outcomes' suggest that both the CIA and Wall Street (and their sister organisations around the world) far too often flounder in the cognitive quagmire in which they operate.

Sixth. In both fields, bureaucratic and corporate reward systems based on hierarchy and material incentives generate conflicts of interest and militate against challenging the conventional wisdom.

Seventh. This makes it more difficult than it should be for even the best professionals to stay on top of new developments. Learning is systematically undermined by policy or business pressure.

Eighth. Chronically and in both fields, decisions are reached based on unquantified or arbitrary criteria, with too few professionals covering too much of the field, too many walls between the professionals and, consequently, too little sharing of information.

Ninth. Intelligence success in both fields is achieved disproportionately by contrarian thinking and critical thinking. Warren Buffet, Peter Lynch and William Miller have managed to 'beat the market' over long periods of time. All three are outstanding contrarians and critical thinkers. All three have stressed the role of games and the liberal arts in keeping the mind flexible and inventive - ie alert and imaginative. It is far more difficult to achieve such flexibility and the freedom to act on one's insights within government intelligence agencies, policy-making bureaucracies or corporate hierarchies.

Ever since Sun-tzu's The Art of War (circa 500 BCE) , strategists and intelligence analysts have been urged to attend to subtleties, be imaginative and to think ahead. What is it that Sun-tzu wrote? "Unless someone has the wisdom of a sage, he cannot use spies; unless he is benevolent and righteous, he cannot employ spies; unless he is subtle and perspicacious, he cannot perceive the substance in intelligence reports. It is subtle, subtle!" But, alas, the vast bureaucracies and stultifying cognitive hierarchies that constitute intelligence, policy-making and corporate decision-making systems are anything but subtle - everywhere. They tend, rather, to be cumbersome, hidebound and reactive.

How, then, are these endemic problems to be overcome? The nine points of comparison would seem to suggest that a fundamental part of any solution must be institutionalising critical thinking. Easily said. How is to be achieved? We need to begin by recognising the common themes I've described and then creating the institutional innovations that will counteract them.

The sort of errors we are talking about do have identifiable causes. Those causes are, ultimately, the cognitive biases and blindspots that are hardwired into the human brain and which govern our thinking relentlessly. They cannot be removed, only corrected for by conscious and disciplined procedures - at every level of a cognitive hierarchy. If we want to overcome endemic tendencies to "irrational exuberance" or cluelessness; if we are to institutionalise critical thinking, then we must require that analysts and managers, policy-makers and decision-makers be skilled in the science of critical metacognition.

Critical metacognition means thinking about the way in which we think. It means being educated in the weaknesses and reflexive errors to which the human brain is naturally and universally prone. It means acknowledging that these weaknesses are heard-wired into the brain and do not melt away because you've got a good track record or because you've got a fancy university degree. It means correcting for these weaknesses and errors by disciplined processes which entail putting ones assumptions and preferences on the line. It means routinely subjecting the conventional wisdom and the apparent causes of one's success, not just one's failure, to ruthless critical review.

Too time consuming and expensive? Well, the dotcom crash wiped some $US2.5 trillion off American share market values in a matter of months. That's what I call too expensive. The cost of the destruction of the WTC and the war against terrorism has yet to be tallied up. Certainly the commitment to critical metacognition would require investment and serious institutional innovation. The US Government has just given the CIA $US1 billion to go after Osama bin Laden. Would it cost much more than that to overhaul cognitive practices at the CIA?

How much would it have cost to have had a system that would have modelled the bombing of the World Trade Centre in 1993 and critically examined, in its aftermath, assumptions blocking foresight about what AQ or similar groups might do next? A team doing such work in July this year might well have imported the Sato scenario from the team monitoring books like Clancy's , recalled the ominous details of Project Bojinka, noted Youseff's 1996 conviction date as a terrorist anniversary and AQ's escalating hostilities. Having done so, they might, just possibly, have pinpointed the threat of an airliner or more than one being used as missiles in an attack on the WTC on 11 September.

What happened in reality? Project Bojinka seems hardly to have registered with CIA or FBI analysts at all. The Philippine police officer who masterminded the operation in Manila has stated that Bojinka simply "was not given credibility. Otherwise [the CIA and FBI] could have prevented the destruction of the World Trade Centre". Murad, it seems, offered to cooperate with prosecutors for a more lenient sentence and may well have provided more information about the background to the plot and AQ's plans. His offer was refused and the matter was filed away and forgotten . Hardly a rigorous or metacognitive procedure. While what is called hindsight bias might well mislead us in cases such as this into thinking that it should have been obvious what was going to happen, the point here is not that it was obvious, only that it was certainly imaginable.

What, therefore, shall we learn from this fiasco? That reality is devilishly complex and unpredictable? That there's no knowing what Islamic extremists will do? That the CIA needs billions of extra dollars to do more of what it does already? I suggest not. I suggest that what it shows is the pressing need for sustained attention to cognitive practices in serious organisations. Indeed, just as we have long become accustomed to financial auditing in such organisations, we might do well to introduce cognitive auditing from now on.

 

© Copyright 2002

FAIR USE NOTICE: This site contains copyrighted material the use of which has not always been specifically authorized by the copyright owner. We are making such material available in our efforts to advance understanding of criminal justice, political, human rights, economic, democracy, scientific, and social justice issues, etc. We believe this constitutes a 'fair use' of any such copyrighted material as provided for in section 107 of the US Copyright Law. In accordance with Title 17 U.S.C. Section 107, the material on this site is distributed without profit to those who have expressed a prior interest in receiving the included information for research and educational purposes. For more information go to: http://www.law.cornell.edu/uscode/17/107.shtml. If you wish to use copyrighted material from this site for purposes of your own that go beyond 'fair use', you must obtain permission from the copyright owner.