Experts have long questioned whether the FBI should be the nation’s primary domestic counterterrorism agency. In spite of the absence of any major attacks on U.S. soil since 9/11, calls for reform within the U.S. intelligence community have amplified in recent months. Some advocates of reorganizing U.S. efforts to combat terrorism suggest that a domestic intelligence agency, like Britain’s MI5, would be more effective than the FBI.
Two of the country’s leading experts—Richard A. Posner, a circuit judge, University of Chicago law professor, and author of Preventing Surprise Attacks: Intelligence Reform in the Wake of 9/11 and Juliette Kayyem, a former member of the National Commission on Terrorism, Harvard professor, and coauthor of Protecting Liberty in an Age of Terror—debate whether the United States would benefit from a domestic intelligence agency similar to Britain’s MI5.
November 17, 2006
I want to first thank Judge Posner for a great series of discussions, as well as CFR.org for giving us the airspace.
In his final post, Posner provides a benign explanation for what the U.S. equivalent of an MI5 might be able to accomplish. Put that way, it seems like a no-brainer, the equivalent of a lean, mean, special operations-type force freed from the passiveness of prosecution.
That, of course, comes at a cost and the cost is something that Posner and I simply will disagree on. There is the cost of adding without subtracting, of overlaying a body over a system that, even at first blush, looks like no one is still in charge. There is the cost, though Posner does not mention it, of freeing a domestic intelligence agency from well-established rules regarding the kinds of “triggers” this democracy has chosen to put on the justification of such surveillance. Given Posner’s well-known ideas regarding the National Security Agency surveillance debate, those costs for him are outweighed by the threat we face. There is the cost of foiling a system that has, in many respects, begun to modify itself in light in 9/11, whether it be changes to investigations, or the law, or those involving the numerous local and state law enforcement officials who are invested in this challenge (Posner calls L.A. Police Chief Bratton’s anti-MI5 statements simple self-defense) and who are likely to know much more about local communities than any force the size of two- to three-thousand employees. (It is likely, given this nation’s size and geographic expanse, that our own MI5 would have to be much larger.)
These are costs, as are the risks that a MI5, freed from the prosecution priority, will become something more than the lean, mean counterterrorism force we wish it would be. It seems to me incompatible to say that a domestic MI5 would have no greater powers than the Federal Bureau of Investigation (FBI), but be freed from the system of accountability and oversight (however loose it is these days) that comes with a judicial presence in the system. In addition, in 2006, MI5 had files on 272,000 Brits—some serious, some not. Standing alone, that number means nothing. But, the British expectation of privacy is substantially different than our own, and that is a consequence that may alter, for readers, Posner’s benign representations. Again, we simply disagree about the balance of this cost. But, in a recent survey of thirty-seven countries by Privacy International, for example, Britain was ranked among Russia and China as practicing “endemic” surveillance against individual citizens.
In any event, I think Posner and I can agree on this. If we are to model ourselves after the British, perhaps what we can best learn is that their single greatest strength is the exceptional degree of coordination that extends throughout their national security hierarchy, a hierarchy that includes a number of entities, as does ours. That is a lesson, domestic MI5 or not, that we would all want for our nation.
Richard A. Posner
November 16, 2006
Professor Kayyem continues to ignore the principal argument for creating a U.S. counterpart to MI5 or CSIS (the Canadian Security Intelligence Service, in some ways a better model for us because Canada is more like the United States than Britain is). That is the incompatibility of a culture of criminal investigation (backward looking, preoccupied with arrests and prosecution, information hugging, etc.) and one of intelligence (preventive in orientation, casting a wide net for clues to impending attacks, laser-beam focused on terrorism and therefore highly sensitive to the need to maintain the loyalty of the U.S. Muslim community). This incompatibility is something that every other major democratic nation recognizes and that ten years of unsuccessful efforts to refocus the FBI confirm.
The arguments in her most recent post are unpersuasive:
That “no intelligence architecture is a perfect watchdog” is irrelevant to whether a new architecture would be an improvement.
The suggestion that noting the potential threat of homegrown terrorism is “fearmongering” is baseless, as is the veiled suggestion that MI5 may have contributed to the alienation of the British Muslim community (in fact MI5 has been criticized for underestimating the threat posed by that community). Also baseless is Kayyem’s suspicion that “part of the desire for an MI5 would be that it has powers beyond the FBI.” It would not. As for the statement by the head of MI5 that 100,000 British Muslims “believ[e] the July 7th  attacks [on the London transit system] to be justified,” does Kayyem doubt the accuracy of the figure, or its implications? Am I the fearmonger, or is she the ostrich?
She is correct that a new agency would not be functioning optimally from the first day. But the organizational problems involved in creating an agency of perhaps two to three thousand employees (the size of MI5 or CSIS—we probably don’t need a larger agency because of our vast number of police, which a domestic intelligence agency could knit into the effective national information-gathering network that we do not have) are not to be compared to the problems of welding 184,000 employees scattered across twenty-two agencies into a single department (DHS). And yes, “a cadre of people trained in such operations” can’t be created overnight, though in fact scattered across the vast U.S. intelligence community (almost 100,000) are enough such people to staff a small agency.
November 15, 2006
The batting average of any intelligence agency is never perfect; that is as obvious as apple pie. We are debating whether a domestic intelligence agency in America would be better, for a variety of reasons, than not having one. The reason to point out MI5’s failures in predicting serious attacks in Britain is to highlight that no intelligence architecture is a perfect watchdog; thus, Posner’s faith in an MI5 to correct for the deficiencies of the Federal Bureau of Investigation (FBI) should be met with a skepticism about its likely effectiveness that he does not display. As for the threat posed by our communities of interest here in America, Posner makes a sweeping claim based on numbers—not facts, not history; that is a claim that may be convincing, but also may just be fearmongering. In any event, Posner seems to acknowledge that part of Britain's problem arises from its failure to integrate its Muslim population as effectively as we have done. What is the likelihood that an MI5 would enable us to continue to be as successful as we have been on that score? Isn’t there a real risk that it might promote just the kind of alienation we see in Britain?
Posner now makes a new argument for why there ought to be a domestic intelligence agency—the competition it would spur in the FBI to finally make the changes that are necessary. Now, anyone who has seen the creation of the Department of Homeland Security, or indeed the Office of the Director of National Intelligence, knows that their birth, and even childhood, are difficult transformation periods. So the notion that we can’t wait for the FBI to change certainly can’t be defended on the grounds that an MI5 equivalent would be functioning immediately. And Posner never addresses the key difference between our country and others—most of those with domestic intelligence agencies have long histories of responding to internal terrorist threats, resulting in a cadre of people trained in such operations. We can’t manufacture personnel with such backgrounds just by assigning them to work in a new building with a new name.
Maybe we should get beyond the abstract debate over names and structures, therefore, and start focusing on just what it is that this MI5 could actually do that the FBI can’t or won’t. After all, post 9/11 we’re a lot closer to having a domestic intelligence agency—in light of the Patriot Act and changes in the Attorney General's Guidelines—than Posner acknowledges, including allowing a number of investigative techniques without any reasonable suspicion. Judge Posner said in his first post that such an MI5 would have no powers that the FBI does not already possess. If that is true, the only reason for having an MI5 is because our present agencies, in Posner's opinion, can’t perform. I suspect, however, that part of the desire for an MI5 would be that it has powers beyond the FBI, though Posner explicitly forecloses that.
Finally, Judge Posner reads the MI5 statement as being less alarmist than I do. The head of MI5 didn’t say there were just 1,600 suspects; she said there were 1,600 actively plotting, but she also said there were another 100,000 people who believed the July 7 attacks to be justified,and it was difficult to get a fix on who among them posed a threat.
Richard A. Posner
November 14, 2006
Professor Kayyem’s response is rather unresponsive. (Her reference to Chief Bratton’s puff piece about the LAPD [Los Angeles Police Department] is naive.) She does not discuss the systemic problem that bedevils the Federal Bureau of Investigation (FBI)—the conflict between criminal investigation and national security intelligence—or the empirical evidence: Ten years after Director Freeh first tried to reorient the Bureau to counterterrorism, the Bureau has taken only the first, halting steps (which is why its failure to uncover terrorist cells should not reassure us that there are no cells). Kayyem does not explain the anomaly that of all countries, only the United States (as far as I know) thinks it sufficient to commit domestic intelligence to a criminal-investigation agency. What’s special about us?
The fact that she takes MI5’s lack of a perfect record to demonstrate its unsuitability as a model for us reflects a fundamental misunderstanding. No intelligence agency has a perfect record, or even a good record. An intelligence agency can no more bat 1.000 than a baseball player can. Intelligence is inherently a highly imperfect undertaking; that is perhaps the most important thing there is to know about intelligence. Its radical imperfection is one of the reasons to have multiple agencies; our lack of a domestic intelligence agency has created a yawning gap in our defenses.
Like many, Kayyem believes we are safe because our Muslim communities are well integrated with the general American culture. Better integrated than Britain’s, yes, but there are two to three million American Muslims and it doesn’t take more than a handful to wreak havoc. There is growing concern not only with homegrown terrorism in general, but with the “second generation” question. May not a few of the teenagers and young adults, offspring of well-integrated American Muslims, perhaps look abroad to events in the Middle East that are helping to promote Muslim extremism and decide to join the fray?
She says the “real task at hand” is “bringing about the transformation of our FBI.” But how to do that? Director Mueller is trying, assisted by an able Central Intelligence Agency (CIA) transplant, Philip Mudd. It is not obvious what more can be done (maybe Kayyem has suggestions). It would be easier to create a new agency than to overcome the FBI culture, and a new agency might add the spur of competition to transformation endeavors within the Bureau.
MI5 undoubtedly owes its success to its longer history of fighting terrorism. We can learn from countries like Britain that have these longer histories. We can start now and hope that in five or ten years we have an agency as effective as MI5 to meet what may well be the greater dangers to national security down the road in this era of terrorism and proliferation. And by the way, the number of British suspects to which the head of MI5 referred recently is not “tens of thousands;” it is 1,600—out of some 1.6 million British Muslims. I don’t think MI5 is “find[ing] more enemies than actually exist.”
November 13, 2006
I would agree with Judge Posner on the following remark: “A domestic intelligence agency doesn't have to be lawless to be effective!” Those, like me, who wonder at the enthusiasm for an MI5 equivalent in the United States are often categorized as concerned solely for civil liberties, sometimes at the peril of security. I'll reserve the liberties discourse for a later post.
From a security perspective, if one were to look at the record of post-9/11 terrorist activity, Britain isn’t a terribly great model. There have been two serious homegrown terrorist attacks on their rail and bus lines that were not noticed by their domestic security agency. A third massive airline plot was disrupted, but its scope and imminence is now in question, even by our own counterterrorism officials, and it seems that even that evidence came from a human intelligence source who volunteered the information. This is a simple and obvious way to state that open democracies—whatever their intelligence architecture—will sometimes fail against internal threats, and the existence of a superagency along the lines of an MI5 is no guarantee against that.
So, I'm often left to wonder: What is the problem that proponents of MI5 want to fix?
First, like Posner, proponents of an MI5 are rightfully questioning of a Federal Bureau of Investigation (FBI) that has failed to deliver in many respects, though many local and state police departments—including, according to Chief Bratton, the Los Angeles Police Department—would argue with the representations regarding the sharing of information that Posner makes. But if the problem is often the sharing of information across agencies, it seems difficult to imagine that another agency—parallel yet distinct from the FBI—wouldn't muck it up even more.
Second, like Posner, proponents also argue that an MI5, freed from the duty to prosecute though with “no (more) powers” than the FBI already has, will be able to find and unearth homegrown terror more successfully. But the evidence that the post- 9/11 FBI is actually missing lots of plots that an MI5 would uncover is not apparent. Surely the fact that we are not aware of the FBI having unearthed a truly serious major plot is no evidence the agency has missed one. Commentators like Posner and myself could wax eloquent about the nature of the homegrown threat, its makeup, size, and complexity. But it is just as possible that the failure to disrupt any serious cells in America (like Posner, I tend to think of most of the terror arrests [to date] as ranging from silly to small-fry) might be because we have been successful at integrating communities that are so clearly not integrated in Britain, Germany, and France.
In fact, from a security perspective, the push to create an MI5 has some real downsides. Even if we assume that the Brits’ MI5 is “better” than our FBI at fighting terrorism, that may have little to do with its structure and much more to do with its methods of training, its long history of fighting a domestic terrorist threat, and the personnel it has attracted to an agency long been devoted to doing just that. Creating a whole new agency, then, may just be a way of ignoring the real task at hand—bringing about the transformation of our FBI. And there’s another reason to worry about creating an MI5. This Friday, the head of MI5 announced that tens of thousands of British citizens were suspected of being threats to the homeland. It was, for many, a rather transparent ploy for more resources and support. Indeed, such a number suggests that a domestic intelligence agency may, like any good bureaucracy, find more enemies than actually exist. And, in so doing, create an atmosphere where their suspected threats become very real. It’s not clear why that makes the MI5 an attractive model for us.
November 13, 2006
Richard A. Posner
At present, the Federal Bureau of Intelligence (FBI) is the federal agency primarily responsible for domestic intelligence. Unlike all other countries (as far as I know), the United States does not have a domestic intelligence agency comparable to Britain’s MI5 or Canada’s CSIS (Canadian Security Intelligence Service), which is to say an agency that has no powers of arrest or criminal law enforcement, but is purely a national-security intelligence service dedicated to detecting terrorist and other plots to attack or undermine (as by espionage) the national homeland.
The reason for placing such an intelligence function outside a criminal-investigation agency, such as the FBI, Scotland Yard, or the Royal Canadian Mounted Police, is that criminal investigators investigate completed crimes, whereas intelligence officers try to prevent terrorism, which, while criminal, can do so much harm that prevention is an imperative and punishment after the fact (for any terrorists who survive the attack) is an inadequate response. Because the FBI is dominated by criminal investigators, it has proved unable to transform itself into an intelligence agency. Conceiving intelligence as merely an adjunct to arrest and prosecution, and measuring success by number of arrests, the Bureau repeatedly jumps the gun, arresting terrorist suspects as soon as it has enough evidence to convict them of “material support” of terrorism or some other preparatory crime, rather than continuing the investigation until the full scope of the terrorist plot is revealed. The Bureau notoriously has failed to develop a computer system adequate to intelligence needs, because criminal investigations are normally handled by the Bureau’s field offices and do not require a sharing of information throughout the Bureau. It has notoriously failed to knit the nation’s 840,000 police into a nationwide network for information concerning potential terrorist threats because it fears having its cases stolen by local police. By requiring all Bureau intelligence operations officers to be trained as special agents (i.e. criminal investigators), the Bureau ensures that it will continue to be dominated by a culture of criminal investigation, not of intelligence. It is now a decade since the Bureau’s directors (Louis Freeh, and now Robert Mueller) vowed to make the Bureau effective against terrorism. Progress has been glacial.
Resistance to the creation of a U.S. counterpart to MI5 and CSIS is fed by misunderstandings concerning the effect of such an agency on civil liberties. MI5 used to operate rather lawlessly by U.S. standards because the United Kingdom (UK) had no Bill of Rights, but no longer, now that the UK has subscribed to the European Convention on Human Rights. Yet MI5 remains highly effective. A domestic intelligence agency doesn’t have to be lawless to be effective! A U.S. agency would operate within the attorney general’s guidelines for terrorist investigations. It would have no powers that the FBI does not have, and it would lack the power of arrest, which the FBI does have. It would be less ham-handed than the FBI is in investigations in the Muslim community, because unlike the FBI, its sole function would be to prevent terrorism—and the biggest factor in prevention is maintaining the loyalty of American Muslims.