Rethinking the FBI?

Would the United States be better off with a new counterterrorism agency? Serious people have called for something like Britain's MI5, a real world precendent for Jack Bauer's omniscient CTU. Last month Judge Richard Posner ran a slightly reworked version of his standard FBI article ("Time to Rethink the FBI") in which he argues, as usual, that the FBI is incapable of effective counterterrorism and that the domestic antiterror functions should be taken from the FBI and given to a new agency modeled after Britain's MI5.

Judge Posner's most recent call for an American MI5 to replace the FBI's counterterror duties is occasioned, of course, by the recent news that the Bureau has been having difficulty crossing and dotting, respectively, all the administrative "t's" and "i's" with regard to the issuance of National Security Letters (NSLs). An NSL is a form of administrative subpoena that allows investigators in terror and counterintelligence cases to obtain certain categories of records, such as telephone records and financial records, if the agents are able to show that the records are relevant to a national security investigation. Apparently the biggest difficulty the Bureau has been having has been in counting how many NSLs were actually issued, although in some few cases there is at least some question whether a law may have been broken.

Let's leave aside for the time being the interesting question of whether better addition skills in the FBI would do much, if anything, to safeguard the nation. Instead let's focus on Judge Posner's by now familiar litany of reasons why we need that MI5 model here and ask: would such an agency even work if it were transplanted to the United States's legal and political environment?

Judge Posner's first complaint with the FBI is that it is primarily focused on prosecuting terrorists, rather than on preventing terrorist acts. This is a somewhat complicated topic, so let's take it in stages:

It is more than a decade since the then director of the FBI, Louis Freeh, tried to make the bureau take the terrorist threat to the United States seriously. He failed. His successor, the current director, Robert Mueller, has tried harder than Mr. Freeh, and has made some progress, but not enough. The cause lies deep in the bureau's organizational culture. The FBI is a detective bureau. Its business is not to prevent crime but to catch criminals. The Justice Department, of which the FBI is a part, knows only one way of dealing with terrorism, and that is prosecution. (Mr. Mueller is a former prosecutor.)

For prosecutors and detectives, success is measured by arrests, convictions and sentences. That is fine when the object is merely to keep the crime rate within tolerable limits. But the object of counterterrorism is prevention. Terrorist attacks are too calamitous for the punishment of the terrorists who survive the attack to be an adequate substitute for prevention.
Judge Posner makes some pretty broad assertions, but let's just ask: is it so bad to have an agency with a detective mentality on the terror beat? The detective mindset is to find out what's going on and, typically, to do so by talking to people. That, in fact, is what the FBI has always done best--it has, over the decades, demonstrated that its agents have a remarkable ability to elicit information from both average citizens as well as persons who are either on the margins of illegal or subversive activity or are actually involved in such activity. That would seem to be a pretty good start for a counter terror agency.

Again, is it necessarily such a bad thing to want to prosecute terrorists--even to plan to do so? I note that news reports show that Britain's MI5 regularly suspends its investigations of terrror plots in favor of arresting the plotters before they can pull off whatever terror act they were planning. That was certainly so in the case of the recent air liner bombing plot--MI5 decided at a certain point that it was too risky to continue in the investigative mode and switched to the prosecutive mode. But, of course, that means that from the very beginning they must have been documenting all their investigative steps and findings with just such an eventuality in mind. Is this approach only wrong when the FBI does it, but admirable when MI5 does it? Sauce for the goose should be sauce for the gander as well--right?

Judge Posner suggests, nay, positively asserts that for the FBI success is measured by "arrests, convictions and sentences." Now, this assertion contradicts one of the cardinal rules for any bureaucracy--and, yes, the Bureau is a bureaucracy. That rule is to take credit for anything for which credit can conceivably be taken. Does Judge Posner really want us to believe that the Bureau doesn't keep track of plots it has thwarted? I can certainly recall listening to newscasts about how many terror plots the FBI has thwarted every year since 9/11. Their math skills may be questionable, but there can be no doubt that the Bureau is quite capable of documenting success in terms of prevention and not just in terms of prosecution. That being the case, we should assume that the in true bureaucratic fashion the FBI not only claims public credit for thwarting terror plots but also gives credit within the organization as well--in the form of statistical achievements, performance awards, promotions, etc. But Judge Posner continues riffing on this theme, so lets follow:

Detecting terrorist plots in advance so that they can be thwarted is the business of intelligence agencies. The FBI is not an intelligence agency, and has a truncated conception of intelligence: gathering information that can be used to obtain a conviction. A crime is committed, having a definite time and place and usually witnesses and often physical evidence and even suspects. This enables a criminal investigation to be tightly focused. Prevention, in contrast, requires casting a very wide investigative net, chasing down ambiguous clues, and assembling tiny bits of information (hence the importance of information technology, which plays a limited role in criminal investigations).

The bureau lacks the tradition, the skills, the patience, the incentive structures, the recruitment criteria, the training methods, the languages, the cultural sensitivities and the career paths that national-security intelligence requires. All the bureau's intelligence operations officers undergo the full special-agent training. That training emphasizes firearms skills, arrest techniques and self-defense, and the legal rules governing criminal investigations. None of these proficiencies are germane to national-security intelligence. What could be more perverse than to train new employees for one kind of work and assign them to another for which they have not been trained?
Once again we are faced with a dense complex of topics, but lets try to follow some of the threads. Judge Posner writes that detecting terrorist plots in advance is the business of intelligence agencies, but that the FBI isn't an intelligence agency. From these premises it should logically follow that the Bureau does nothing to detect terrorist plots in advance--but is that true? Let's look at the mother of all terrorist plots: 9/11. Is it true that the Bureau was doing little or nothing to detect that plot in advance? To the contrary, what we have learned is that the Bureau's field investigators were busting their rear ends to pull the pieces of the plots together and came very close to doing so. Agents in New York, Phoenix and Minneapolis as well as other localities were tracing the plotters. The common denominator in their failures was that they were thwarted by the United States's own legal and political establishment, and especially by the notorious "Wall" that the Department of Justice's Office of Intelligence Policy and Review (OIPR) insisted upon interpreting in an unwarrantedly restrictive way.

Judge Posner's remedy for the failure of 9/11--the failure of our political and legal establishment that can be traced back to the Church Committee--is to form a new agency. Does this make sense? Would not that new agency be operating in the same unduly restrictive legal and political environment as the Bureau currently operates in? Should not Judge Posner be addressing root causes, rather than be recommending organizational band aids?

Then again, at some point the proof must be in the pudding. The fact is that, thank God, we have been spared further terror attacks since 9/11. From events around the world we know that the terror masters are persistent and ingenious, so we should be able to assume that our good fortune doesn't result from any change of heart among our terrorist enemies. What, then? Have we simply been lucky? While that is one possibility, I would like to suggest another.  It is also possible that the FBI has actually been doing some things right. Maybe not everything; maybe their paper counting skills could be improved upon, but doing enough things right to play a positive role in preserving the United States from the type of devestating terror attacks that we've seen around the world. That's not a suggestion that the FBI should rest on its laurels, but it is a sober reflection about rocking boats. If the mix of factors that has contributed to preventing additional terror attacks since 9/11 includes effective action by the FBI, then we should be very cautious about tampering with that mix of factors. Before we break up the FBI and start from scratch, it would be best to document that the Bureau really is "incapable of effective counterterrorism." So far, Judge Posner is less than persuasive.

What, then, of Judge Posner's contention that "[t]he bureau lacks the tradition, the skills, the patience, the incentive structures, the recruitment criteria, the training methods, the languages, the cultural sensitivities and the career paths that national-security intelligence requires.?" Is he on firmer ground here? Lets walk through this laundry list and see.

Does the FBI lack a tradition of national security intelligence work? Clearly the Bureau has a strong tradition in the national security field, dating back to near its founding. The Bureau was involved in investigating Soviet spy activities well before World War II, and has devoted significant resources to Counterintelligence up to the present day. This has been a steady, patient, long term effort that only draws public attention when a sensational case breaks publicly, but it is a consistent effort that has continued behind the scenes for well over half a century. The failures gain the most publicity, but there have been well documented successes, as well. Clearly, then, in light of this tradition, the Bureau must possess significant intelligence related skills, and must have had some success in recruiting talented national security investigators. As for incentive structures, why should arrests and convictions be the only statistical measure of success that the Bureau would recognize in this decades long involvement in national security? Judge Posner may lack imagination in this regard, but it is not difficult to imagine that the Bureau would afford internal career enhancing recognition to agents who have successfully identified intelligence officers and their agents, who may even have recruited informants or foreign spies who have provided sensitive information of great value to national security. Given the level of resources that the FBI has devoted to national security matters for so many decades it would be shocking--and contrary to well known behavior patterns of bureaucracies--if such incentive structures were not in place.

A further consideration that comes under the heading of "tradition" has to do with the Bureau's great expertise in investigating a wide spectrum of organized crime organizations, extending back to its earliest days. Terrorist organizations share many similarities with more traditional types of organized crime associations. Indeed terrorist groups are well known to both recruit among criminal elements as well as to engage in criminal activities in order to finance their terrorist structures and agenda--significantly blurring such seemingly hard and fast distinctions as "intelligence" and "criminal." Maintaining a unitary FBI facilitates and enhances communication between criminal investigative and intelligence investigative elements, each of which may come across activity of interest to one another and (now that the infamous "Wall" has been torn down) may find fruitful room for cooperation and exchange of information.

Judge Posner approaches the heart of the problem of an American MI5 when he states:

"All the bureau's intelligence operations officers undergo the full special-agent training. That training emphasizes firearms skills, arrest techniques and self-defense, and the legal rules governing criminal investigations. None of these proficiencies are germane to national-security intelligence. [My emphasis.] What could be more perverse than to train new employees for one kind of work and assign them to another for which they have not been trained?"
Is this confident assertion really true? Lets look at Judge Posners assertion that national security agents have no need to learn about "the legal rules governing criminal investigations." We have already noted that MI5 led investigations regularly result in arrests and convictions in Britain. The same is true in the United States. The fact is, the rules governing criminal investigations may indeed be highly relevant in national security investigations. Espionage, for example, is a criminal violation and the rules that apply to other criminal violations also apply to espionage investigations. The same is true of criminal terrorism violations. The fact that prosecutions of these violations may be the exception rather than the rule does not change the fact that investigators will need to understand criminal procedure or risk blowing important cases.

As for the supposed irrelevance of "firearms skills, arrest techniques and self-defense," exactly how does Judge Posner expect "American MI5" agents will deal with extraordinarily dangerous terrorists, many of them highly trained and motivated? Judge Posner is perfectly correct to note that "Every major nation (and many minor ones), except the United States, concluded long ago that domestic intelligence should be separated from its counterpart to the FBI." But what he fails to note is that the United States has a federal sytem of government. In Britain, MI5 can and routinely does draw on highly trained units of the country's police forces--it is not too much to say that they work hand in glove with the police, a matter that Judge Posner neglects to mention. Other countries have highly trained paramilitary police units at the ready. Would the "American MI5" be able to do the same? One may seriously doubt whether it would be wise for an "American MI5 to count on help from local police in a federal system like ours. Aside from complex jurisdictional issues in our federal sytem, we have already witnessed the police forces of major United States cities publicly stating that they will provide no assistance to the FBI in its national security investigations. We know that more than a few major cities refuse to assist in enforcing immigration laws--certainly an area that frequently intersects with national security investigations.

Having made these dubious assertions, Judge Posner provides an example that he clearly believes reveals FBI wrongheadedness, not to say blockheadedness:

MI5 and its counterparts in other nations...do not make the mistake that the FBI made last year in arresting seven Muslims in Miami on suspicion of plotting to blow up buildings there, along with the Sears Tower in Chicago.

The bureau had been able to plant an informant in the group. Yet as soon as it had enough evidence to prove a criminal conspiracy, it pounced. Because the group had no money or backers (except the FBI's informant!) and no skills or experience, and had been penetrated, it was not an imminent threat, so there was no urgency about arresting its members. The group had wanted to get in touch with foreign terrorists but had been unable to do so.

The informant might have helped them do so -- might even have helped them become part of a serious terrorist network, enabling the bureau to ascertain the network's scope and membership, methods and tradecraft, even goals and specific plans. The opportunity to exploit the penetration in this fashion was lost by the arrests -- about which the Attorney General boasted to an extent that, given the ineptness of the defendants, evoked ridicule.
News reports regarding these plotters revealed that the plotters--not to put too fine a point on the matter--were, in common parlance, a group of knuckleheads. Evil intentioned, but knuckleheads nevertheless. Are we to suppose that al Qaeda would be no more perceptive regarding these plotters than was the FBI? Would not al Qaeda, in all likelihood, steer clear of these clumsy amateurs? Any intelligence agency has to weigh its chances for success in any given case. Judge Posner has passed judgment in this case--the FBI and DoJ were guilty of great stupidity that merited the ridicule that great minds heaped upon them. But the resources that would have been involved in attempting to use these Miami plotters--who had "no money or backers...and no skills or experience" and had been unable to establish terrorist contacts--would have been considerable, and would have had to be committed for a period that would quite possibly extend for years, with no promise of success. If the plotters had failed in their efforts to contact known terror groups, we can also safely assume that the FBI informant had been unable to assist in that effort. Perhaps it was more prudent for the FBI to pull the plug on this case, take the knuckleheads off the street and devote always scarce resources to a more promising case. And not tell Judge Posner that that was what they were doing.

It will probably be best to pass over Judge Posner's reference to the 9/11 Commission in silence and proceed to this curious observation:

Civil libertarians worry about abuses of domestic intelligence. But an agency that had no powers of arrest or prosecution, and that conceived its primary role to be to prevent the alienation of Americans who have religious or family ties to nations that harbor terrorists, rather than to run up arrest statistics, would be less likely than the FBI to engage in the promiscuous issuance of administrative subpoenas.

Lets start from the end. My understanding is that the issue regarding the FBI's use of NSLs ("administrative subpoenas") doesn't revolve around whether or not there was justification for issuing the NSLs; rather, the issue is whether or not all administrative procedures were followed. And yet Judge Posner takes a fairly gratuitous swipe at the FBI's supposedly "promiscuous issuance of administrative subpoenas." Earlier, Judge Posner warned us that "arrest and prosecution should be postponed until a terrorist network has been fully traced and its methods, affiliates, financiers, suppliers and camp followers identified..." There is a problem here. The standard method for identifying terrorist "affiliates, financiers, suppliers and camp followers" involves examining their communications, then indentifying their contacts and examining their activities to determine the full extent of the terrorist network. This is typically accomplished through the use of NSLs to obtain communication and financial information, as we have learned from the continuing furor over the Patriot Act, FISA, the NSA and the monitoring of international financial transactions. How Judge Posner knows that the FBI has used NSLs "promiscuous[ly]" is anyone's guess, since that would presumably require a detailed knowledge of FBI terror investigations--as well as counterintelligence investigations, which also utilize NSLs. However, if indeed the the Bureau is issuing a large number of NSLs, that is actually an argument that the Bureau is aggressively pursuing terrorist leads--a good thing for the average citizen, but a blow to Judge Posner's contention that the Bureau takes no interest in such matters.

In a world with ever multiplying means for transferring money, data, text and voice communications, it is troubling that Judge Posner envisions an "American MI5" that makes less, rather than more, use of NSLs. Leaving the complexity of modern technology to one side, we should nevertheless expect the FBI to make constant use of NSLs, since terrorist organizations are well known to take precautions to elude our surveillance by continuously changing their methods of communicating and of transferring funds.

Judge Posner assures us that civil libertarians will cease to worry about abuses of domestic intelligence if the FBI's counterterror duties are replaced by an "American MI5" that has "no powers of arrest or prosecution." In fact, as should be well known to Judge Posner, no law enforcement organization in the United States has powers of prosecution, but even so this is a meaningless distinction. We have already noted that MI5 works hand in glove with British law enforcement organizations in their non-federal system--arrests and prosecutions regularly follow from MI5 investigations, just as they do from FBI investigations. Even more troubling, Judge Posner appears to equate the collection of intelligence information through what he calls the "promiscuous issuance of administrative subpoenas" with legitimate civil liberty concerns. How, in this day and age, our intelligence agencies are supposed to protect us without gathering and analyzing vast amounts of information is frankly beyond me, and yet Judge Posner assures us that that will be the case.

As for the tight integration of "local police and other information gatherers (such as border patrol police, customs officials and private security personnel) into a comprehensive national intelligence network, as MI5 has done in Britain -- and as the FBI has failed to do here," any failure in that regard is due as much to our federal system of government as to any FBI elitism. MI5 has been able to achieve that tight integration precisely because of these important differences in our constitutions and form of government. Transplanting the MI5 model to the United States political and legal environment would by no means assure us of achieving that type of integration--in fact, it is predictable that that would not occur but that the "American MI5" would find itself more, rather than less, isolated from local police. Why would this be so? The FBI has established a nation wide network of Joint Terrorism Task Forces (JTTFs) that include representatives from "local police and other information gatherers (such as border patrol police, customs officials and private security personnel)"--although Judge Posner strangely fails to take note of this fact. Why are local agencies of all sorts eager to cooperate with the FBI on terrorism matters, despite tensions based on our federal system? For the very good reason that participation in a JTTF can open the door for cooperation with the FBI in matters of far more immediate concern to local police forces than terrorism. But an "American MI5" separate from the FBI would have no such inducements to offer, and no way to compel cooperation. In our federal system the JTTF framework is probably as good as it gets, all Judge Posner's generalizing criticisms to the contrary notwithstanding. (An aside: Judge Posner's remark about "case stealing" tension and fears between the FBI and the local police is puzzling, given the clear jurisdictional lines that exist.)

Well, then, is everything right with the FBI? By no means, and strangely Judge Posner has failed to raise the strongest arguments that are available to him: he makes no mention of the FBI's counterintelligence functions. Now, MI5 handles not only counterterrorism but has also long been Britain's lead counterintelligence agency. Any "American MI5" would doubtless, like its British model, include both fields so it behooves us to ask how the Bureau measures up in the field of counterintelligence as well as in that of counterterrorism. I have defended the Bureau's recent record in counterterrorism and have noted its undoubted counterintelligence successes in the past. However, by far the most egregious recent Bureau failures in recent years have been precisely in the field of counterintelligence. What stronger argument could be advanced to prove that the FBI's National Security Branch is in need of overhaul than the mere mention of the Wen Ho Lee, Robert Hanssen and Katrina Leung cases? OIG executive summaries of these cases provide as much of the gory details as most of us can handle. Each of these cases involved compromises of national security of the most extreme types as well as systematic malfeasance, nonfeasance and mismanagement from beginning to end and from top to bottom of the FBI.

Judge Posner does make a passing reference to the fact that "the FBI director and his staff ... have no background in domestic intelligence." This reference, of course, brings to mind the Gary Bald affair. The appointment of Bald to a top intelligence position was strongly criticized because of Bald's lack of an intelligence background, as was famously evidenced by his public statement, under oath, that he neither knew nor cared about the distinction between Sunni and Shiite branches of Islam. According to Director Mueller and Bald himself, Bald's managerial skills made such (for them) arcane knowledge irrelevant. Bald ultimately received his appointment and six months later, having padded his resume, retired to take a high paying position in a field totally unrelated to intelligence. This is a shameful state of affairs and Director Mueller should be held to account for it. But how can this situation be remedied, short of establishment of an "American MI5."

Let me first make it clear that I am skeptical both as to the prospects of improving the situation by establishing an "American MI5" as well as to the prospects for reform of the FBI in the current legal and political environment. My preference for maintaining a unitary FBI is based on a belief that it is the lesser evil given our federal system, not on any illusion that it can be perfected. As I indicated above, an "American MI5" would almost certainly be unable to secure the type of cooperation from other law enforcement agencies that Britain's MI5 is able to command. On the other hand, an "American MI5" would be very much smaller than the current FBI. It would be even more subject to inter-agency jealousies and rivalries and even less able to defend its turf and have its concerns addressed by the rest of Government. (As an aside, it may be worth noting here that the law enforcement footprint in Britain, per capita, in relationship to the citizenry, is significantly larger than in the United States. Indeed, by European standards the United States is "under-policed." The FBI has only in the neighborhood of 12,000 agents scattered across an enourmous and populous country, handling duties of all sorts.)

Judge Posner hints that the Bureau's connection to the Department of Justice is problematic for an intelligence agency. His concern is undoubtedly well founded; study after study has demonstrated that OIPR involvement and legal interpretation has had an almost uniformly deleterious influence on national security and intelligence operations. Yet, while I believe that Judge Posner's observation has merit, it is difficult to see how it could be otherwise in our system of government; I do not believe that establishing an "American MI5" within the Department of Homeland Security would be an improvement, as it would be dwarfed in that mammoth bureaucracy. My best suggestion would also probably prove unpopular with the political class and civil liberties advocates. But here it is.

Judge Posner has identified lack of intelligence experience in the Bureau's top management levels as a significant problem. To that concern may be added the misuse of intelligence resources, an unresponsive command structure and a failure in devising a managerial system that adequately identifies and trains individuals for promotion who are truly talented, rather than interested in enhancing their resumes for cushy retirement jobs. While I have been critical of the details of Judge Posner's position, I believe he has made his case that these issues, at least, are significant concerns. In addition, they may be susceptible of correction.

The solution is to maintain the Bureau's unitary structure because of the great benefits it brings, but to inject a far greater degree of managerial oversight than currently exists in order to insure a professionalization of the Bureau's National Security Branch. That would mean that the FBI Director would not have a free hand in appointments to national security related positions and would be held accountable for funding, spending and training in the national security field. Theoretically, that type of oversight is supposed to be exercised, to some degree, by the newly created Director of National Intelligence, but we have yet to see any evidence that that position will be more than another self-aggrandizing federal bureaucracy. I suggest that the FBI's National Security Branch is small enough that the type of oversight that is needed could be adequately handled by the President's National Security Council. By requiring approval from the NSC for significant National Security Branch appointments the NSC would be in a position to enforce a professionalization of the Bureau's national security operations. More direct oversight for training and funding would also force the Bureau to clean house, as its problems would come more directly to view at a higher level than is currently the case. In addition, the National Security Branch would benefit because its own concerns would be more likely to receive a hearing, since the NSC would, reciprocally, share responsibility.

Finally, let me pose the question: what stands in the way of meaningful reform? In a word: Congress. With the advent of the income tax and the rise of the United States as an international economic power vast sums of money have flowed into the Treasury--and vast sums flow out. Control of the purse strings gives power, in many different ways. It has taken some time but, through the device of omnibus spending bills and oversight committees, an anomolous situation has arisen. While executive branch agencies are theoretically subject to the President, these agencies have come to recognize that their true master is Congress, which controls the purse strings. Agency heads know that Congress can tie the President in knots with oversight and investigations, and so they recongnize that their careers will be made or broken on the basis of their relationship with Congress as much as or more than their relationship with the President. Congressional intervention in the field of intelligence, since the time of the Church Committee, has reduced the field of national security to the status of a vehicle for the practice of demogoguery in pursuit of partisan political advantage. I believe that meaningful intelligence reform, as any reform, will ultimately be possible only if the President is given more control over the operation and management of the Executive Branch through a more significant say in the budget process. Until then, national security will likely remain a plaything for Congress. And perhaps it is here that an instructive comparison can be made to the British model, for while Governments have come and gone in Britain, as everywhere, national security debate in Britain has never, I believe, been reduced to the status of a political football to the extent that it has in the United States.

Al Johnson is a retired attorney with an interest in national security affairs.
Would the United States be better off with a new counterterrorism agency? Serious people have called for something like Britain's MI5, a real world precendent for Jack Bauer's omniscient CTU. Last month Judge Richard Posner ran a slightly reworked version of his standard FBI article ("Time to Rethink the FBI") in which he argues, as usual, that the FBI is incapable of effective counterterrorism and that the domestic antiterror functions should be taken from the FBI and given to a new agency modeled after Britain's MI5.

Judge Posner's most recent call for an American MI5 to replace the FBI's counterterror duties is occasioned, of course, by the recent news that the Bureau has been having difficulty crossing and dotting, respectively, all the administrative "t's" and "i's" with regard to the issuance of National Security Letters (NSLs). An NSL is a form of administrative subpoena that allows investigators in terror and counterintelligence cases to obtain certain categories of records, such as telephone records and financial records, if the agents are able to show that the records are relevant to a national security investigation. Apparently the biggest difficulty the Bureau has been having has been in counting how many NSLs were actually issued, although in some few cases there is at least some question whether a law may have been broken.

Let's leave aside for the time being the interesting question of whether better addition skills in the FBI would do much, if anything, to safeguard the nation. Instead let's focus on Judge Posner's by now familiar litany of reasons why we need that MI5 model here and ask: would such an agency even work if it were transplanted to the United States's legal and political environment?

Judge Posner's first complaint with the FBI is that it is primarily focused on prosecuting terrorists, rather than on preventing terrorist acts. This is a somewhat complicated topic, so let's take it in stages:

It is more than a decade since the then director of the FBI, Louis Freeh, tried to make the bureau take the terrorist threat to the United States seriously. He failed. His successor, the current director, Robert Mueller, has tried harder than Mr. Freeh, and has made some progress, but not enough. The cause lies deep in the bureau's organizational culture. The FBI is a detective bureau. Its business is not to prevent crime but to catch criminals. The Justice Department, of which the FBI is a part, knows only one way of dealing with terrorism, and that is prosecution. (Mr. Mueller is a former prosecutor.)

For prosecutors and detectives, success is measured by arrests, convictions and sentences. That is fine when the object is merely to keep the crime rate within tolerable limits. But the object of counterterrorism is prevention. Terrorist attacks are too calamitous for the punishment of the terrorists who survive the attack to be an adequate substitute for prevention.
Judge Posner makes some pretty broad assertions, but let's just ask: is it so bad to have an agency with a detective mentality on the terror beat? The detective mindset is to find out what's going on and, typically, to do so by talking to people. That, in fact, is what the FBI has always done best--it has, over the decades, demonstrated that its agents have a remarkable ability to elicit information from both average citizens as well as persons who are either on the margins of illegal or subversive activity or are actually involved in such activity. That would seem to be a pretty good start for a counter terror agency.

Again, is it necessarily such a bad thing to want to prosecute terrorists--even to plan to do so? I note that news reports show that Britain's MI5 regularly suspends its investigations of terrror plots in favor of arresting the plotters before they can pull off whatever terror act they were planning. That was certainly so in the case of the recent air liner bombing plot--MI5 decided at a certain point that it was too risky to continue in the investigative mode and switched to the prosecutive mode. But, of course, that means that from the very beginning they must have been documenting all their investigative steps and findings with just such an eventuality in mind. Is this approach only wrong when the FBI does it, but admirable when MI5 does it? Sauce for the goose should be sauce for the gander as well--right?

Judge Posner suggests, nay, positively asserts that for the FBI success is measured by "arrests, convictions and sentences." Now, this assertion contradicts one of the cardinal rules for any bureaucracy--and, yes, the Bureau is a bureaucracy. That rule is to take credit for anything for which credit can conceivably be taken. Does Judge Posner really want us to believe that the Bureau doesn't keep track of plots it has thwarted? I can certainly recall listening to newscasts about how many terror plots the FBI has thwarted every year since 9/11. Their math skills may be questionable, but there can be no doubt that the Bureau is quite capable of documenting success in terms of prevention and not just in terms of prosecution. That being the case, we should assume that the in true bureaucratic fashion the FBI not only claims public credit for thwarting terror plots but also gives credit within the organization as well--in the form of statistical achievements, performance awards, promotions, etc. But Judge Posner continues riffing on this theme, so lets follow:

Detecting terrorist plots in advance so that they can be thwarted is the business of intelligence agencies. The FBI is not an intelligence agency, and has a truncated conception of intelligence: gathering information that can be used to obtain a conviction. A crime is committed, having a definite time and place and usually witnesses and often physical evidence and even suspects. This enables a criminal investigation to be tightly focused. Prevention, in contrast, requires casting a very wide investigative net, chasing down ambiguous clues, and assembling tiny bits of information (hence the importance of information technology, which plays a limited role in criminal investigations).

The bureau lacks the tradition, the skills, the patience, the incentive structures, the recruitment criteria, the training methods, the languages, the cultural sensitivities and the career paths that national-security intelligence requires. All the bureau's intelligence operations officers undergo the full special-agent training. That training emphasizes firearms skills, arrest techniques and self-defense, and the legal rules governing criminal investigations. None of these proficiencies are germane to national-security intelligence. What could be more perverse than to train new employees for one kind of work and assign them to another for which they have not been trained?
Once again we are faced with a dense complex of topics, but lets try to follow some of the threads. Judge Posner writes that detecting terrorist plots in advance is the business of intelligence agencies, but that the FBI isn't an intelligence agency. From these premises it should logically follow that the Bureau does nothing to detect terrorist plots in advance--but is that true? Let's look at the mother of all terrorist plots: 9/11. Is it true that the Bureau was doing little or nothing to detect that plot in advance? To the contrary, what we have learned is that the Bureau's field investigators were busting their rear ends to pull the pieces of the plots together and came very close to doing so. Agents in New York, Phoenix and Minneapolis as well as other localities were tracing the plotters. The common denominator in their failures was that they were thwarted by the United States's own legal and political establishment, and especially by the notorious "Wall" that the Department of Justice's Office of Intelligence Policy and Review (OIPR) insisted upon interpreting in an unwarrantedly restrictive way.

Judge Posner's remedy for the failure of 9/11--the failure of our political and legal establishment that can be traced back to the Church Committee--is to form a new agency. Does this make sense? Would not that new agency be operating in the same unduly restrictive legal and political environment as the Bureau currently operates in? Should not Judge Posner be addressing root causes, rather than be recommending organizational band aids?

Then again, at some point the proof must be in the pudding. The fact is that, thank God, we have been spared further terror attacks since 9/11. From events around the world we know that the terror masters are persistent and ingenious, so we should be able to assume that our good fortune doesn't result from any change of heart among our terrorist enemies. What, then? Have we simply been lucky? While that is one possibility, I would like to suggest another.  It is also possible that the FBI has actually been doing some things right. Maybe not everything; maybe their paper counting skills could be improved upon, but doing enough things right to play a positive role in preserving the United States from the type of devestating terror attacks that we've seen around the world. That's not a suggestion that the FBI should rest on its laurels, but it is a sober reflection about rocking boats. If the mix of factors that has contributed to preventing additional terror attacks since 9/11 includes effective action by the FBI, then we should be very cautious about tampering with that mix of factors. Before we break up the FBI and start from scratch, it would be best to document that the Bureau really is "incapable of effective counterterrorism." So far, Judge Posner is less than persuasive.

What, then, of Judge Posner's contention that "[t]he bureau lacks the tradition, the skills, the patience, the incentive structures, the recruitment criteria, the training methods, the languages, the cultural sensitivities and the career paths that national-security intelligence requires.?" Is he on firmer ground here? Lets walk through this laundry list and see.

Does the FBI lack a tradition of national security intelligence work? Clearly the Bureau has a strong tradition in the national security field, dating back to near its founding. The Bureau was involved in investigating Soviet spy activities well before World War II, and has devoted significant resources to Counterintelligence up to the present day. This has been a steady, patient, long term effort that only draws public attention when a sensational case breaks publicly, but it is a consistent effort that has continued behind the scenes for well over half a century. The failures gain the most publicity, but there have been well documented successes, as well. Clearly, then, in light of this tradition, the Bureau must possess significant intelligence related skills, and must have had some success in recruiting talented national security investigators. As for incentive structures, why should arrests and convictions be the only statistical measure of success that the Bureau would recognize in this decades long involvement in national security? Judge Posner may lack imagination in this regard, but it is not difficult to imagine that the Bureau would afford internal career enhancing recognition to agents who have successfully identified intelligence officers and their agents, who may even have recruited informants or foreign spies who have provided sensitive information of great value to national security. Given the level of resources that the FBI has devoted to national security matters for so many decades it would be shocking--and contrary to well known behavior patterns of bureaucracies--if such incentive structures were not in place.

A further consideration that comes under the heading of "tradition" has to do with the Bureau's great expertise in investigating a wide spectrum of organized crime organizations, extending back to its earliest days. Terrorist organizations share many similarities with more traditional types of organized crime associations. Indeed terrorist groups are well known to both recruit among criminal elements as well as to engage in criminal activities in order to finance their terrorist structures and agenda--significantly blurring such seemingly hard and fast distinctions as "intelligence" and "criminal." Maintaining a unitary FBI facilitates and enhances communication between criminal investigative and intelligence investigative elements, each of which may come across activity of interest to one another and (now that the infamous "Wall" has been torn down) may find fruitful room for cooperation and exchange of information.

Judge Posner approaches the heart of the problem of an American MI5 when he states:

"All the bureau's intelligence operations officers undergo the full special-agent training. That training emphasizes firearms skills, arrest techniques and self-defense, and the legal rules governing criminal investigations. None of these proficiencies are germane to national-security intelligence. [My emphasis.] What could be more perverse than to train new employees for one kind of work and assign them to another for which they have not been trained?"
Is this confident assertion really true? Lets look at Judge Posners assertion that national security agents have no need to learn about "the legal rules governing criminal investigations." We have already noted that MI5 led investigations regularly result in arrests and convictions in Britain. The same is true in the United States. The fact is, the rules governing criminal investigations may indeed be highly relevant in national security investigations. Espionage, for example, is a criminal violation and the rules that apply to other criminal violations also apply to espionage investigations. The same is true of criminal terrorism violations. The fact that prosecutions of these violations may be the exception rather than the rule does not change the fact that investigators will need to understand criminal procedure or risk blowing important cases.

As for the supposed irrelevance of "firearms skills, arrest techniques and self-defense," exactly how does Judge Posner expect "American MI5" agents will deal with extraordinarily dangerous terrorists, many of them highly trained and motivated? Judge Posner is perfectly correct to note that "Every major nation (and many minor ones), except the United States, concluded long ago that domestic intelligence should be separated from its counterpart to the FBI." But what he fails to note is that the United States has a federal sytem of government. In Britain, MI5 can and routinely does draw on highly trained units of the country's police forces--it is not too much to say that they work hand in glove with the police, a matter that Judge Posner neglects to mention. Other countries have highly trained paramilitary police units at the ready. Would the "American MI5" be able to do the same? One may seriously doubt whether it would be wise for an "American MI5 to count on help from local police in a federal system like ours. Aside from complex jurisdictional issues in our federal sytem, we have already witnessed the police forces of major United States cities publicly stating that they will provide no assistance to the FBI in its national security investigations. We know that more than a few major cities refuse to assist in enforcing immigration laws--certainly an area that frequently intersects with national security investigations.

Having made these dubious assertions, Judge Posner provides an example that he clearly believes reveals FBI wrongheadedness, not to say blockheadedness:

MI5 and its counterparts in other nations...do not make the mistake that the FBI made last year in arresting seven Muslims in Miami on suspicion of plotting to blow up buildings there, along with the Sears Tower in Chicago.

The bureau had been able to plant an informant in the group. Yet as soon as it had enough evidence to prove a criminal conspiracy, it pounced. Because the group had no money or backers (except the FBI's informant!) and no skills or experience, and had been penetrated, it was not an imminent threat, so there was no urgency about arresting its members. The group had wanted to get in touch with foreign terrorists but had been unable to do so.

The informant might have helped them do so -- might even have helped them become part of a serious terrorist network, enabling the bureau to ascertain the network's scope and membership, methods and tradecraft, even goals and specific plans. The opportunity to exploit the penetration in this fashion was lost by the arrests -- about which the Attorney General boasted to an extent that, given the ineptness of the defendants, evoked ridicule.
News reports regarding these plotters revealed that the plotters--not to put too fine a point on the matter--were, in common parlance, a group of knuckleheads. Evil intentioned, but knuckleheads nevertheless. Are we to suppose that al Qaeda would be no more perceptive regarding these plotters than was the FBI? Would not al Qaeda, in all likelihood, steer clear of these clumsy amateurs? Any intelligence agency has to weigh its chances for success in any given case. Judge Posner has passed judgment in this case--the FBI and DoJ were guilty of great stupidity that merited the ridicule that great minds heaped upon them. But the resources that would have been involved in attempting to use these Miami plotters--who had "no money or backers...and no skills or experience" and had been unable to establish terrorist contacts--would have been considerable, and would have had to be committed for a period that would quite possibly extend for years, with no promise of success. If the plotters had failed in their efforts to contact known terror groups, we can also safely assume that the FBI informant had been unable to assist in that effort. Perhaps it was more prudent for the FBI to pull the plug on this case, take the knuckleheads off the street and devote always scarce resources to a more promising case. And not tell Judge Posner that that was what they were doing.

It will probably be best to pass over Judge Posner's reference to the 9/11 Commission in silence and proceed to this curious observation:

Civil libertarians worry about abuses of domestic intelligence. But an agency that had no powers of arrest or prosecution, and that conceived its primary role to be to prevent the alienation of Americans who have religious or family ties to nations that harbor terrorists, rather than to run up arrest statistics, would be less likely than the FBI to engage in the promiscuous issuance of administrative subpoenas.

Lets start from the end. My understanding is that the issue regarding the FBI's use of NSLs ("administrative subpoenas") doesn't revolve around whether or not there was justification for issuing the NSLs; rather, the issue is whether or not all administrative procedures were followed. And yet Judge Posner takes a fairly gratuitous swipe at the FBI's supposedly "promiscuous issuance of administrative subpoenas." Earlier, Judge Posner warned us that "arrest and prosecution should be postponed until a terrorist network has been fully traced and its methods, affiliates, financiers, suppliers and camp followers identified..." There is a problem here. The standard method for identifying terrorist "affiliates, financiers, suppliers and camp followers" involves examining their communications, then indentifying their contacts and examining their activities to determine the full extent of the terrorist network. This is typically accomplished through the use of NSLs to obtain communication and financial information, as we have learned from the continuing furor over the Patriot Act, FISA, the NSA and the monitoring of international financial transactions. How Judge Posner knows that the FBI has used NSLs "promiscuous[ly]" is anyone's guess, since that would presumably require a detailed knowledge of FBI terror investigations--as well as counterintelligence investigations, which also utilize NSLs. However, if indeed the the Bureau is issuing a large number of NSLs, that is actually an argument that the Bureau is aggressively pursuing terrorist leads--a good thing for the average citizen, but a blow to Judge Posner's contention that the Bureau takes no interest in such matters.

In a world with ever multiplying means for transferring money, data, text and voice communications, it is troubling that Judge Posner envisions an "American MI5" that makes less, rather than more, use of NSLs. Leaving the complexity of modern technology to one side, we should nevertheless expect the FBI to make constant use of NSLs, since terrorist organizations are well known to take precautions to elude our surveillance by continuously changing their methods of communicating and of transferring funds.

Judge Posner assures us that civil libertarians will cease to worry about abuses of domestic intelligence if the FBI's counterterror duties are replaced by an "American MI5" that has "no powers of arrest or prosecution." In fact, as should be well known to Judge Posner, no law enforcement organization in the United States has powers of prosecution, but even so this is a meaningless distinction. We have already noted that MI5 works hand in glove with British law enforcement organizations in their non-federal system--arrests and prosecutions regularly follow from MI5 investigations, just as they do from FBI investigations. Even more troubling, Judge Posner appears to equate the collection of intelligence information through what he calls the "promiscuous issuance of administrative subpoenas" with legitimate civil liberty concerns. How, in this day and age, our intelligence agencies are supposed to protect us without gathering and analyzing vast amounts of information is frankly beyond me, and yet Judge Posner assures us that that will be the case.

As for the tight integration of "local police and other information gatherers (such as border patrol police, customs officials and private security personnel) into a comprehensive national intelligence network, as MI5 has done in Britain -- and as the FBI has failed to do here," any failure in that regard is due as much to our federal system of government as to any FBI elitism. MI5 has been able to achieve that tight integration precisely because of these important differences in our constitutions and form of government. Transplanting the MI5 model to the United States political and legal environment would by no means assure us of achieving that type of integration--in fact, it is predictable that that would not occur but that the "American MI5" would find itself more, rather than less, isolated from local police. Why would this be so? The FBI has established a nation wide network of Joint Terrorism Task Forces (JTTFs) that include representatives from "local police and other information gatherers (such as border patrol police, customs officials and private security personnel)"--although Judge Posner strangely fails to take note of this fact. Why are local agencies of all sorts eager to cooperate with the FBI on terrorism matters, despite tensions based on our federal system? For the very good reason that participation in a JTTF can open the door for cooperation with the FBI in matters of far more immediate concern to local police forces than terrorism. But an "American MI5" separate from the FBI would have no such inducements to offer, and no way to compel cooperation. In our federal system the JTTF framework is probably as good as it gets, all Judge Posner's generalizing criticisms to the contrary notwithstanding. (An aside: Judge Posner's remark about "case stealing" tension and fears between the FBI and the local police is puzzling, given the clear jurisdictional lines that exist.)

Well, then, is everything right with the FBI? By no means, and strangely Judge Posner has failed to raise the strongest arguments that are available to him: he makes no mention of the FBI's counterintelligence functions. Now, MI5 handles not only counterterrorism but has also long been Britain's lead counterintelligence agency. Any "American MI5" would doubtless, like its British model, include both fields so it behooves us to ask how the Bureau measures up in the field of counterintelligence as well as in that of counterterrorism. I have defended the Bureau's recent record in counterterrorism and have noted its undoubted counterintelligence successes in the past. However, by far the most egregious recent Bureau failures in recent years have been precisely in the field of counterintelligence. What stronger argument could be advanced to prove that the FBI's National Security Branch is in need of overhaul than the mere mention of the Wen Ho Lee, Robert Hanssen and Katrina Leung cases? OIG executive summaries of these cases provide as much of the gory details as most of us can handle. Each of these cases involved compromises of national security of the most extreme types as well as systematic malfeasance, nonfeasance and mismanagement from beginning to end and from top to bottom of the FBI.

Judge Posner does make a passing reference to the fact that "the FBI director and his staff ... have no background in domestic intelligence." This reference, of course, brings to mind the Gary Bald affair. The appointment of Bald to a top intelligence position was strongly criticized because of Bald's lack of an intelligence background, as was famously evidenced by his public statement, under oath, that he neither knew nor cared about the distinction between Sunni and Shiite branches of Islam. According to Director Mueller and Bald himself, Bald's managerial skills made such (for them) arcane knowledge irrelevant. Bald ultimately received his appointment and six months later, having padded his resume, retired to take a high paying position in a field totally unrelated to intelligence. This is a shameful state of affairs and Director Mueller should be held to account for it. But how can this situation be remedied, short of establishment of an "American MI5."

Let me first make it clear that I am skeptical both as to the prospects of improving the situation by establishing an "American MI5" as well as to the prospects for reform of the FBI in the current legal and political environment. My preference for maintaining a unitary FBI is based on a belief that it is the lesser evil given our federal system, not on any illusion that it can be perfected. As I indicated above, an "American MI5" would almost certainly be unable to secure the type of cooperation from other law enforcement agencies that Britain's MI5 is able to command. On the other hand, an "American MI5" would be very much smaller than the current FBI. It would be even more subject to inter-agency jealousies and rivalries and even less able to defend its turf and have its concerns addressed by the rest of Government. (As an aside, it may be worth noting here that the law enforcement footprint in Britain, per capita, in relationship to the citizenry, is significantly larger than in the United States. Indeed, by European standards the United States is "under-policed." The FBI has only in the neighborhood of 12,000 agents scattered across an enourmous and populous country, handling duties of all sorts.)

Judge Posner hints that the Bureau's connection to the Department of Justice is problematic for an intelligence agency. His concern is undoubtedly well founded; study after study has demonstrated that OIPR involvement and legal interpretation has had an almost uniformly deleterious influence on national security and intelligence operations. Yet, while I believe that Judge Posner's observation has merit, it is difficult to see how it could be otherwise in our system of government; I do not believe that establishing an "American MI5" within the Department of Homeland Security would be an improvement, as it would be dwarfed in that mammoth bureaucracy. My best suggestion would also probably prove unpopular with the political class and civil liberties advocates. But here it is.

Judge Posner has identified lack of intelligence experience in the Bureau's top management levels as a significant problem. To that concern may be added the misuse of intelligence resources, an unresponsive command structure and a failure in devising a managerial system that adequately identifies and trains individuals for promotion who are truly talented, rather than interested in enhancing their resumes for cushy retirement jobs. While I have been critical of the details of Judge Posner's position, I believe he has made his case that these issues, at least, are significant concerns. In addition, they may be susceptible of correction.

The solution is to maintain the Bureau's unitary structure because of the great benefits it brings, but to inject a far greater degree of managerial oversight than currently exists in order to insure a professionalization of the Bureau's National Security Branch. That would mean that the FBI Director would not have a free hand in appointments to national security related positions and would be held accountable for funding, spending and training in the national security field. Theoretically, that type of oversight is supposed to be exercised, to some degree, by the newly created Director of National Intelligence, but we have yet to see any evidence that that position will be more than another self-aggrandizing federal bureaucracy. I suggest that the FBI's National Security Branch is small enough that the type of oversight that is needed could be adequately handled by the President's National Security Council. By requiring approval from the NSC for significant National Security Branch appointments the NSC would be in a position to enforce a professionalization of the Bureau's national security operations. More direct oversight for training and funding would also force the Bureau to clean house, as its problems would come more directly to view at a higher level than is currently the case. In addition, the National Security Branch would benefit because its own concerns would be more likely to receive a hearing, since the NSC would, reciprocally, share responsibility.

Finally, let me pose the question: what stands in the way of meaningful reform? In a word: Congress. With the advent of the income tax and the rise of the United States as an international economic power vast sums of money have flowed into the Treasury--and vast sums flow out. Control of the purse strings gives power, in many different ways. It has taken some time but, through the device of omnibus spending bills and oversight committees, an anomolous situation has arisen. While executive branch agencies are theoretically subject to the President, these agencies have come to recognize that their true master is Congress, which controls the purse strings. Agency heads know that Congress can tie the President in knots with oversight and investigations, and so they recongnize that their careers will be made or broken on the basis of their relationship with Congress as much as or more than their relationship with the President. Congressional intervention in the field of intelligence, since the time of the Church Committee, has reduced the field of national security to the status of a vehicle for the practice of demogoguery in pursuit of partisan political advantage. I believe that meaningful intelligence reform, as any reform, will ultimately be possible only if the President is given more control over the operation and management of the Executive Branch through a more significant say in the budget process. Until then, national security will likely remain a plaything for Congress. And perhaps it is here that an instructive comparison can be made to the British model, for while Governments have come and gone in Britain, as everywhere, national security debate in Britain has never, I believe, been reduced to the status of a political football to the extent that it has in the United States.

Al Johnson is a retired attorney with an interest in national security affairs.