The Foreign Intelligence Surveillance Act (FISA) is a chronic problem. The controversy over President Bush's decision to bypass FISA warrants in the electronic surveillance of al Qaeda operatives has highlighted the act's limitations. But FISA has been a problem ever since it became law in 1978.
Congress passed and President Carter signed the bill regulating electronic surveillance for foreign intelligence collection in the wake of an extended, post--Watergate debate about the so--called "imperial presidency." The debate was given added urgency by reports and official investigations of indiscriminate snooping in this country by elements of the U.S. intelligence community. However, like so much else from that period, the broad arguments about the president's role in the constitutional order were wrong, and the laws designed to correct real problems created a new set of problems.
One irony of today's debate is that so many liberals are now defending FISA. Previously, a common complaint from the ACLU and others was that the secret federal court that issues warrants for foreign intelligence surveillance in this country had become a "rubber stamp" for the executive branch. Out of the thousands of applications put forward by the Department of Justice to the panel over the years, only a handful had ever been rejected. Instead of a check on executive authority, the court had become complicit in its activities-or so it was said.
And to a certain extent that has been the case. Yet the reason for the high percentage of approvals has less to do with deference to executive judgment than with FISA's standard for obtaining a warrant when it involves surveillance of an American citizen or an alien residing legally in the United States. Before the government can get a warrant, the Justice Department must put together a case to present before the court stating the "facts and circumstances relied upon . . . to justify [the attorney general's] belief that the target is an agent of a foreign power" or "engages . . . in international terrorism." And the FISA judges can only grant the warrant when "there is probable cause to believe that the target" is engaged in espionage or terrorism. In short, before the government can collect intelligence on someone by breaking into his house or tapping his phones, it had better already have in hand pretty persuasive evidence that the person is probably up to no good. FISA is less about collecting intelligence than confirming intelligence.
This shouldn't be news. The inability of the FBI to aggressively pursue suspected foreign agents on U.S. soil was well aired in the 9/11 post--mortems, especially the notorious case of the "20th hijacker," French Moroccan Zacarias Moussaoui. His laptop was in the possession of FBI agents in Minneapolis in August 2001 but, for want of a warrant, was not searched before 9/11.
Though much of the court's workings are classified, there are known instances in which FISA's "probable cause" standard prevented the government from getting warrants where common sense made it perfectly clear surveillance was justified. Notably, there was the case of Wen Ho Lee, the Chinese--American scientist who worked at the Los Alamos nuclear weapons lab. Lee downloaded nuclear codes and databases from the lab's secure computers. "In the wrong hands," his boss noted, such information could "change the global strategic balance." Despite this, and the fact that Lee had access to a warhead design that had leaked to the Chinese, had visited China in the period when Beijing apparently acquired the data, and had obvious friendly ties to Chinese nuclear scientists, it was judged that a FISA warrant could not be obtained. It didn't matter how grave the damage might be if Lee was actually engaged in espionage; what mattered was the government had no real evidence that Lee was a likely spy.
In retrospect, the fastidiousness shown in the Lee and Moussaoui cases seems ludicrous, and some politicians and experts who have an interest in saving FISA have argued that the law was read too narrowly in both instances. Yet the law is what it is. And certainly it is these same politicians and experts who would be the first to complain about the government not abiding by FISA's standards should a case come to light involving dubious surveillance. Contrary to what many might think, the FBI, NSA, and the permanent bureaucracy at the Department of Justice have supported FISA primarily because it provided a line that they believed protected their organizations from the kinds of public trouble, political second--guessing, and congressional investigations that plagued them in the 1970s. And, for the most part, it has-although at an obvious cost.
More broadly, the law rests on a shallow understanding of the Constitution's system of separated powers. FISA's supporters believe that life without the law would lead to unfettered executive power and violate the system's guiding principle of "checks and balances." What the Constitution demands, in their view, is a two--key approach to public authority: No branch gets to act in key instances without concurrent approval from a second branch.
That approach, however, conflates the Constitution's scheme of "checks and balances" with its more fundamental system of separated powers. Although some checks do exist-like the president's qualified veto over legislation, or the Senate's role in confirming nominations-they are not the norms for government action but the exceptions. Far more common is for the president, Congress, and the courts to do their own thing, each interacting with the others indirectly and rarely concurrently.
And that is the way it was meant to be. Justice Brandeis's famous line that the "doctrine of separation of powers was adopted by the Convention of 1787, not to promote efficiency but to preclude the exercise of arbitrary power" is at best a half truth. After a decade of living under the Articles of Confederation, and seeing state legislatures run roughshod over weak executives, the Constitution's drafters wanted a system of separated powers that would free up the executive, not tie it down. By the time they convened in Philadelphia, the bias against the executive that arose from the fight with the British crown was pretty well gone. So much so that The Federalist would argue during the ratification debate that "energy in the executive is a leading character in the definition of good government" and that "decision, activity, secrecy and dispatch" were qualities only a unitary and independent executive could provide.
It's no surprise then that it is precisely these qualities that we see in President Bush's decision to go around FISA in the wake of 9/11 and to order the National Security Agency to conduct warrantless surveillance of emails and calls going back and forth from suspected al Qaeda operatives abroad to the United States, and vice versa.
Some critics of the president's actions have argued that he should have asked Congress to amend FISA to meet these new circumstances. Leave aside the practicalities of getting legislation of this sort passed in a timely manner, and without the underlying rationale leaking. This president (or any president worth his salt) would only accept legislation that either confirmed his discretionary authority, or reduced the standard for getting a warrant to some minimal requirement that the collection be "reasonably" connected to the country's foreign intelligence needs.
But if legislation of that type were passed, what role would the FISA court actually play? Either it really would become a rubber stamp, or it would become a surrogate executive confirming or denying a warrant based on reasoning that isn't, at bottom, judicial. Do we really want judges to play the role of second--guessers of executive branch decisions, substituting their own judgment on matters of national security for that of the president and his duly appointed subordinates?
So, then, what is to be done? Well, to start, we should have a serious debate about abolishing FISA and restoring the president's inherent constitutional authority to conduct warrantless searches for foreign intelligence purposes. And no, this wouldn't return us to the bad old days of a snooping J. Edgar Hoover. Within the executive branch there are now reams of guidelines and teams of inspector generals that make renegade operations improbable or, at least, difficult to keep hidden very long.
Just as important, there are now standing intelligence committees in both the House and the Senate. One of the odd effects of FISA has been to take serious and sustained congressional oversight of electronic surveillance off the table. The constitutional body that should be watching the executive's discretionary behavior is, after all, primarily Congress.
Here we reach the nub of the matter: The Founders, in the words of The Federalist, did not think it was wise or even possible to set a "limitation of that authority which is to provide for the defense and protection of the community." At the end of the day, a government has to do what is necessary to protect itself and its people. Yet, at the same time, the Founders believed in limited government. How did they square the circle? When it comes to the conduct of war, the history is pretty clear: They expected presidents to do what was required to secure the country's safety. But they did anticipate that Congress would play the role of Monday--morning quarterback: exposing malfeasance when called for, adding or cutting off funds when necessary, passing laws to regularize the exercise of executive discretion without undermining it, and, in the face of truly egregious behavior, being ready to impeach a president.
Obviously there is no neat solution to the problem of power and responsibility. However, as Winston Churchill said about democracy itself, the system of discretion and oversight the Constitution establishes is the worst possible solution-except for all others that have been tried.
Gary Schmitt is director of the program on advanced strategic studies at the American Enterprise Institute and former executive director of the President's Foreign Intelligence Advisory Board.