Robert Jervis. 2010. Why Intelligence Fails: Lessons from the Iranian Revolution and the Iraq War. (Ithaca, N.Y., Cornell University Press, 248 pages).

When it comes to assessing the strength, the intention and the common sense of a foreign regime, why do intelligence analysts often get things quite so wrong and what can be done to reduce the likelihood of future intelligence failures? These are intriguing albeit not exactly novel questions. Next to official “postmortems” (p. 123) on intelligence failures (i.e. reports by judicial inquiries and oversight commissions), one also finds a growing number of intelligence scholars wrestling with the subject. Still, Robert Jervis, a long-time expert on deception in international relations and occasional CIA contractor, is uniquely positioned to expand on the practice, the challenges and the limitations to foreign intelligence analysis and his latest book, Why intelligence fails, does indeed stand out.

The author brings his privileged access to the intelligence community and his long-time experience with research design to bear on two remarkable case studies: the first on the inability of the U.S. intelligence community to predict the onslaught of the Islamic Revolution in Iran and the second on its flawed assessment regarding Saddam Hussein’s WMD programme. This adds much-needed texture to the arcane life-world of foreign intelligence. For example, in the Iraqi WMD intelligence failure case, Jervis reviews the entire intelligence production process from the perspective of the individual analysts who were tasked to make sense of conflicting reports on Saddam’s capabilities and intentions. We learn about the kind of information that they had to work with and the foregone opportunities to apply basic social science research methods to make their underlying assumptions more explicit and to embrace alternative explanations for their findings. We also learn about the analysts, their training, their resources, their work routines as well as the oftentimes difficult division of labour and information flows amongst different agencies as well as the taxing interactions with the various ‘clients’ in Washington. The constant referrals to the concrete case at hand makes his venture into the labyrinth of U.S. intelligence readable and provides authority to his recommendations for intelligence reform. Jervis rightly cautions against undue expectations on the intelligence services (arguing convincingly that excellent analysts could still have drawn the wrong conclusions) and his rebuttals of overzealous or merely cosmetic reform proposals follow logically from the empirical chapters.

It becomes more problematic, in my view, when Jervis uses his findings to reject the widespread notion that the intelligence community bowed to political pressure from the Bush administration. Jervis does not deny that the Bush administration pressured the intelligence services to produce suitable outputs for its predetermined decision to invade Iraq but he finds no reason to believe that the intelligence services succumbed to this pressure. This is, of course, quite a position to defend and the author is notably pleased to be at odds with mainstream analysis, which “so comforts to common sense that it has been a barrier to more careful thought” (p. 131). Yet it is here where Jervis fails to adhere to his own standards: Why must his well-argued position that the intelligence community failed to make sense of social science research methods be necessarily at odds with the finding that the intelligence services were also captured by an extremely hawkish administration? Jervis’ does not give much thought to this and his eagerness to rule out the latter is not in keeping with his basic advice for future generations of intelligence analysts.

(For a French version of this text, see: Poltique Etrangère, Vol. 76, Nr. 2.)

Dana Priest and William Arkin’s reporting on Top Secret America in the Washington Post was an important media event. Its various graphs and installments nicely illustrate the sheer magnitude of US intelligence privatization. Mind you, the size and the compartmentalization of the US intelligence community is difficult to grasp and a cause for many problems of its own. Add to that the even more elusive layers of contracting and sub-contracting (an estimated 265’000 contractors have top-secret clearances) and one gets a security state that has grown beyond that what anyone can call reasonable by a long stretch of imagination. (If only McNulty had a fraction of that cash for the fight against drugs in Baltimore…)

Here are a few more thoughts on both the content and the style of the WaPo story.

First off, the sprawling privatization of US intelligence is, of course, hardly news to the engaged reader. The WaPo must be criticized for not even referring to Tim Shorrock, the real authority on this subject. He was the first author to have studied post 9/11 intelligence contracting, and he did it in a much more systematic and problem-centered fashion (albeit with less colorful graphs and less floury language on Washington suburbia). His book, Spies for Hire, is a compelling read that tells us why, where and how the US arrived at a situation where an incredible 70 percent of its enormous intelligence budget ends up in the hands of private companies. One can also find his data set, here. Shorrock obtained an unclassified document from the office of the Director of National Intelligence (DNI) that corroborates his basic estimate. After his book release, the CIA and the main stream media (MSM) simply ignored Shorrock’s research. The former called his figures ‘way off the mark’ and the latter shunned an awfully gripping story. Why? Surely, it would be too simplistic to associate big advertising companies of MSM, such as Lockheed Martin and Northrop Grumman with this. Fact is, though, it took MSM a long time to pick it up.

Second, and more importantly, the WaPo piece is also short of analysis. Now that the MSM has picked up the story, it does not seem to know what to do with it. The WaPo story hardly raises the important conclusions that ought to be drawn from this incredible intelligence-industrial complex. More concretely, what are the strategic drawbacks when private companies influence public intelligence policy? All of these 1,913 companies pursue their shareholders’ rather than the public interest. Who can deny that the industry shapes intelligence policy making when the contractors are so deeply entwined with those who decide on the contracts, sometimes even overseeing the work of fellow contractors? Companies want more contracts for their business to flourish. To do so, they will pull all possible strings to convince the intelligence community (IC) that their service are not only top-notch quality but also indispensable for counter-terrorism. (One interesting part in the WaPo story was how some IT gadgets developed by corporate America for their intelligence services create envy amongst colleagues, branches etc, which, of course, leads to even more profit). It’s no longer just research and development but actually inherent government functions that are being sold and re-sold like commodities. Not only can private companies bid for such functions, they can also help define what a government function in the intelligence domain should look like, now and in the future. They continue to win large contracts (consider, for example, the latest 100 million contract awarded to Blackwater) despite their often-times very problematic, direct and unaccountable involvement in enhanced interrogation, torture and assassinations in the field or their exploitation of the federal budget from their plush new offices in suburbia. For an example of the former, grab a copy of Spies for hire. Shorrock gives the reader a much more contextual understanding of how the private company CACI directed torture at Abu Ghraib. All this is something, that the WaPo authors « soft-pedaled » in the output of their two-year investigation. (Marcy Wheeler is right on the mark on this and uses stronger words for their lack of critical reporting.)

[...]

Earlier this month, the UN Special Rapporteur on the promotion and protection of human rights and fundamental freedoms while countering terrorism submitted a “compilation of good practices and institutional frameworks and measures that ensure respect for human rights by intelligence agencies while countering terrorism, including on their oversight” to the UN Human Rights Council.

Despite its cumbersome title, this annotated enumeration of 35 examples of good practice is certainly worth reading. It contains a comprehensive and well-structured account of what national intelligence laws and institutional frameworks for intelligence services could look like. This compilation is the outcome of a consultation process where Governments, experts and practitioners provided input in various different ways. I’ve been asked to give comments on an earlier draft and hope that the following general thought on this kind of best practice advocacy might be of interest to Ultima Ratio readers.

What exactly is the objective of this report? It is interesting how A/HRC/14/46 fails to provide a clear statement of purpose. In the introduction, the Special Rapporteur states that « it is not the purpose of this compilation to promulgate a set of normative standards that should apply at all times and in all parts of the world ». Fine, this should not be read as a blueprint for all nations to follow. But what then is its purpose?

Possibly, this was a contentious issue for the various parties that have contributed to the consultation process. National governments, academic experts, the UN bureaucracy and professional consultants are likely to pursue different and potentially conflicting objectives. Perhaps, then, the author felt compelled not to spell this out in further detail. If this was the case, it would illustrate one of the numerous problems with (or better: constraints of) such mandated research. All too often it is the logic of the consultation process rather than the logic of an empirical enquiry that is decisive in the end. This is all the more deplorable because the latter can generate custom-tailored and sustainable policy recommendations for intelligence governance, too.

[...]

The US has 16 different intelligence agencies (See this graph in a recent FT article).  German tax payers also pay for the questionable luxury to entertain an equal number state-level intelligence agencies next to the three federal intelligence services. Both countries are – to say it diplomatically – organisationally and logistically challenged as a result.

Followers of the Bundestag’s BND-inquiry (2006-2009) might remember some of the testimonies by state-level intelligence officers. They were entirely embarassing for the intelligence community and the government at large. Through this rare keyhole of intelligence politics, the public learned about the duplication of investigative tasks, disorganisation, amenities, ill-communication and arbitrary work ethics. This sparked a debate whether or not to reduce the number of intelligence agencies in Germany. Yet, as ever so often with intelligence politics, despite a debate, the politics machine moved on without seriously addressing the issue.

In the US, the situation is very similar. Here, too, we find an inability/unwillingness to change the obviously sub-standard status quo. The 9/11 Commission called for swift changes but any real down-sizing of the US intelligence community has yet to see the light of day.

A common-sense resolution to the problem of intelligence compartmentalisation comes from the fringe — i.e. by two professors who rarely reflect on intelligence politics. Richard A. Posner and Luis Garicano compare the deplorable status quo of the US intelligence community with that of the US car giant GM and propose similar radical measures to get a grip on the problem.

« The domestic automakers’ organizational structures were notoriously complex and top-heavy. While Toyota had been selling the same car worldwide, Ford had insisted that American consumers would not buy the cars successfully produced by Ford for sale in Europe. As a result, every stage of production from R&D to actual manufacturing was duplicated in the two markets.

(Snip) [...]

In his inauguration speech, President Obama famously rejected as false the choice between America’s safety and its ideals. Has the new administration honoured this commendable position in political practice? After more than a year of steering American security, a fact-check appears to be in order.

I want to do this here with respect to intelligence governance. Examining first whether or not the new government has kept its promises, I then elaborate on  the driving forces behind the recent developments in this arcane domain.

Intelligence governance under Obama …

Unsurprisingly, the field of intelligence is replete with opportunities for the Obama administration to demonstrate its commitment to an « unprecedented level of openness » and its desire to « restore the standards of due process and core constitutional values that have made this country great ».

Under G.W. Bush, intelligence governance has severely damaged America’s international reputation. Key strategic decisions (such as the invasion of Iraq) were based on flawed and heavily politicised intelligence, US intelligence agents conducted torture and extraordinary renditions on suspected terrorists (often on the basis of equally flawed intelligence, see, for example, the el-Masri case), and the NSA engaged in illegal surveillance/wire-tapping of US citizens.

Unfortunately, this is only a selection of the most stinking turds that President Obama has found under the rugs of the White House upon his arrival. (For a fuller scope, see the ACLU’s Torture Report). On his first full day in office, Barack Obama has responded to this rather unpleasant dowry by promising the restoration of constitutional values. How did that go and what does the process reveal about the master and servants of US intelligence governance? [...]

For researchers interested in government secrecy, www.wikileaks.org can often be a valuable site. Sometimes, however, it presents information in an unbearably biased fashion. Consider, for example, this recent headline:

« US intelligence planned to destroy wikileaks ».

This sounded like quite a story. What would be the implications if that was true? Should Google pull out of the US? I was curious and started reading a bit in the leaked study commissioned by the US Department of Defense Intelligence Analysis Program (DIAP). It examined how wikileaks.org might jeopardise the interests of US forces by playing vital intelligence into the hands of its various opponents. Not surprisingly, it concluded that

« Wikileaks.org, a publicly accessible Internet Web site, represents a potential force protection, counterintelligence, operational security (OPSEC), and information security (INFOSEC) threat to the US Army. »

Looking then for signs that US intelligence actually planned to disrupt this site, I only stumbled across a number of rather common-sense recommendations:

« the identification, exposure, termination of employment, criminal prosecution, legal action against current or former insiders,
leakers, or whistlblowers could potentially damage or destroy this center of gravity and deter others considering similar actions from using the Wikileaks.org Web site ».

Don’t get me wrong. I admire and support the power of investigative sources on the net. Still, it is also understandable (and a calculated risk for leakers) that those who would like to keep the information secret will not shy away from using all available legal tricks against them. That’s not anything new under the sun nor has this threat become more serious for leakers than before.

What I find disturbing, then, is the sensationalist headline. Based on what I’ve gathered from DIAP study, this is a far cry from any plan to “destroy » Wikileaks. This kind of reporting is, in my view, entirely inappropriate. A more factual approach would be in order.

Last month, to quote Kings of War blogger, Rob Dover,

« a veritable poo-storm hit the British Court of Appeal (…) as a private exchange of correspondence – between the government’s lawyer Jonathan Sumption QC and the Court – became public. The government had asked that part of the judgment relating to the alleged torture of Binyam Mohammed be redacted (or removed) in the interests of protecting the reputation of our Security Service (known to nearly everyone as MI5). »

The case of Binyam Mohamed is rather complex (see this piece for starters, see also this piece for the various court rulings in pdfs, see Scott Horton’s feed for a more holistic appreciation of recent developments news on this). First and foremost, it entails a credible allegation of torture complicity by British intelligence. Yet, on closer inspection, it provides researchers with a rare opportunity to depict more general, systemic flaws with the British « art » of celebrating intelligence accountability. Amongst other things, it reveals how national security has become conflated with national embarrassment throughout the oversight proceedings on this particular accountability case.

While the official British intelligence watchdog (aka as the Intelligence and Security Committee (ISC)) has looked into this case (see its Renditions report (2007)), the most revealing information is to be found elsewhere. In Mohamed v. Foreign Secretary, the British High Court revealed (in 2008), how the ISC, which supposedly operates from within the ring of secrecy, has been kept outside of the loop in this (and arguably a number of other cases that will gradually come to the fore).

[...]