Summary  Why Section 218 Should be Retained  Imaginary Walls ...  A Response ...  The Last Word?


Section 218.
Amending the FISA Standard

A Summary by Mary DeRosa

Section 218 amends FISA by changing the certification requirement when the government seeks a FISA surveillance or search order. Previously, the government was required to certify that "the purpose" of the application was to obtain foreign intelligence information. After section 218, the government must certify that obtaining foreign intelligence information is "a significant purpose" of the application. This change was designed to promote information sharing between intelligence and law enforcement officials and to eliminate what has become known as the "wall" that separated law enforcement and intelligence investigations.

The FISA, passed in 1978, sets forth procedures for the conduct of electronic surveillance and physical searches for foreign intelligence purposes. Over the years, the Department of Justice interpreted FISA's requirement that "the purpose" of collection be foreign intelligence to restrict the use of FISA collection procedures when a law enforcement investigation was involved. The restriction was designed to ensure that prosecutors and criminal investigators did not use FISA to circumvent the more rigorous warrant requirements for criminal cases. But law enforcement and foreign intelligence investigations often overlap, and enforcing this separation between intelligence and law enforcement investigations—the "wall"—inhibited coordination of these investigations and the sharing of foreign intelligence information with law enforcement officials. The change to "significant purpose" was intended to clarify that no such separation is necessary.

It is not clear whether the change in section 218 was legally necessary to eliminate the "wall." The Department of Justice argued to the FISA Court of Review in 2002 that the original FISA standard did not require the restrictions that the Department of Justice imposed over the years, and the court appears to have agreed. This leaves the precise legal effect of a sunset of section 218 somewhat murky.

Why Section 218 Should be Retained
by Andrew C. McCarthy

No subordination of national security to hypothetical fears of civil liberties abuse was more emblematic of the pre-9/11 world than the metaphorical "wall' erected to obstruct the information flow between intelligence and criminal investigators.

Section 218 of the Patriot Act dismantled this construct by amending its literal underpinning—the basis for the ill-conceived "primary purpose' test by which FISA was misinterpreted for nearly a quarter-century, to disastrous effect. As the wall was founded on a skewed interpretation of law, Section 218 was theoretically unnecessary. Nevertheless, it was entirely appropriate and its enactment proved to be critical.

Post-9/11, discussions focus on explaining the genesis of the wall rather than defending it. Indeed, former Attorney General Janet Reno, on whose watch the wall was solidified in internal guidelines, testified to the 9/11 Commission that, more critical to national security than realigning the intelligence community would be " to knock down walls, to promote the sharing of information, and to enhance collaboration in the fight against terrorism.' And in 2002, the Foreign Intelligence Surveillance Court of Review, in its first ever opinion, provided a detailed explanation of the wall's fatal flaws.

The relevant history traces to the 1978 enactment of FISA (50 U.S.C. §§ 1801et seq. ((2000 ed.)). A reaction to Vietnam and Watergate era domestic-intelligence abuses, FISA authorizes a special federal FISA court to regulate and monitor the executive branch's conduct of electronic surveillance and physical searches in the context of national-security investigations. This is in contrast to ordinary investigations, where the use of those techniques is governed by the criminal law.

In the latter, agents must present probable cause of a crime to obtain a warrant. FISA, on the other hand, is not principally about rooting out crime; it is about national defense, targeting foreign enemies, including international terrorists. Thus, rather than requiring probable cause of a crime, FISA permitted government to "obtain foreign intelligence information' if "there is probable cause to believe that ... the target of the electronic surveillance is a foreign power or an agent of a foreign power[.]'

The difficulty here is that any theoretical divide between criminal and intelligence matters would not track reality. Espionage, for example, is both a dire national security issue and a felony. Similarly, terrorists commit many crimes (e.g., immigration fraud, identity theft, money laundering, seditious conspiracy, possession of precursor explosives, and bombing, to name just a few) in the course of plotting and attacking. Thus, whether an agent's investigative authority comes from FISA or the criminal law, what emerges is evidence that constitutes both national security intelligence and proof of quotidian crimes.

This should pose no problem. Agents conducting a proper investigation uncover information. Free to compare notes and study multiple options for dealing with threats to public safety, they can wisely choose the approach that makes the most sense in light of the entire informational mosaic. Prosecution of a crime will get a dangerous person off the street and, equally important, may motivate him to cooperate about the inner workings of a terror network. On the other hand, sustained monitoring might reveal the nature of a terror enterprise while allowing government to prevent attacks without triggering disclosure obligations that attend a prosecution (which educate terrorists about the state and sources of government's intelligence). Plainly, national security dictates a fully informed strategy, taking advantage of the tactics that best fit the circumstances. Prior to 9/11, however, development of such a strategy was hamstrung by a hypothetical and wrong-headed concern: viz., that permitting use in criminal cases of FISA-generated evidence might induce agents to resort to FISA when their "real' purpose was to conduct a criminal investigation.

This was irrational. First, the existence of a crime or national security threat is an objective reality, entirely independent of the investigators' subjective mindsets about why they are investigating. As for agent motivation, our concerns should be whether they have a good reason for investigating and whether the facts they present to a court are accurate. If those things are so, and agents happen to uncover evidence they did not anticipate finding, that is cause for celebration, not suppression. Thus, it has for decades been the law that (i) evidence of Crime A is admissible even if it was seized in the execution of warrant based on probable cause of Crime B; but (ii) evidence of a crime is suppressed if the probable cause predicating its seizure was based on intentional misstatements of material fact.

Second, it is not sensible to suspect systematically dishonest resort to FISA. FISA applications require a specialized and rigorous internal approval process before presentation to the court. Assuming arguendo an agent willing to act corruptly, it would be far easier and less detectable to fabricate the evidence necessary to get an ordinary criminal wiretap than to fabricate a national security reason to use FISA.

Finally, FISA as written posed no obstacle to the use of FISA evidence for criminal prosecution. From a national security perspective, this made eminent sense given the aforementioned propensity of terrorists to commit crimes and the consequent centrality of prosecution as a means to win cooperation and thus secure vital intelligence.

Regrettably, this common sense came unmoored over time. FISA required that a high executive branch official—typically, the FBI director—represent that "the purpose' of the investigation was to obtain foreign-intelligence information (as opposed to building a prosecution). This was simply intended to be a certification; it did not purport to restrict either the scope of the investigation or the permissible uses of any resulting evidence. Unfortunately, soon after FISA took effect, the Justice Department began construing the certification not as a mere announcement of purpose but as something more restrictive: a substantive limitation on the use of FISA evidence in criminal cases.

As the Review Court opinion elaborated, over time this erroneous interpretation of the certification requirement led to a "false dichotomy': a futile endeavor to sort FISA-derived information into the purportedly distinct categories of mere intelligence and criminal evidence. Moreover, given the government's apparent fear that there might be impropriety in the acquisition of criminal evidence via FISA, it should have come as no surprise that the federal courts, too, began fashioning safeguards not found in FISA's text. Thus was born the "primary purpose' test, under which FISA-derived evidence could not be used in criminal prosecutions unless the government demonstrated that its primary purpose had been to collect intelligence, not build a criminal case.

To the contrary, as the Review Court held in 2002, FISA as enacted "clearly did not preclude or limit the government's use ... of foreign intelligence information, which included evidence of certain kinds of criminal activity, in a criminal prosecution.' (Emphasis in original.) But rather than challenge the primary purpose test, the Justice Department bolstered it, by internal 1995 regulations, into what became known as "the wall.' This procedural edifice instructed "the FBI and Criminal Division [to] ensure that advice intended to preserve the option of a criminal prosecution does not inadvertently result in either the fact or the appearance of the Criminal Division's directing or controlling the [foreign intelligence (FI) or counterintelligence (FCI)] investigation toward law enforcement objectives.'

This directive, the Review Court found, was "narrowly interpreted' to " prevent the FBI intelligence officials from communicating with the Criminal Division regarding ongoing FI or FCI investigations.' This effectively cut intelligence investigators off not only from criminal agents but also from Assistant United States Attorneys who, by virtue of investigating and prosecuting several terrorism cases in the 1990's, were among the government's best resources regarding al Qaeda and its affiliates.

The best known pernicious consequence of all this occurred in August 2001. Relying on the wall, FBI headquarters declined to allow criminal investigators to assist an intelligence investigation seeking to locate probable terrorists Khalid al-Midhar and Nawaf al-Hazmi. A few weeks later, on 9/11, the pair helped hijack Flight 77 and pilot it into the Pentagon.

Section 218 makes a seemingly small but crucial adjustment: it guts the primary purpose test by requiring a government to certify that foreign intelligence is merely a significant purpose, rather than the purpose, for the FISA application. This strikes the correct balance: It recognizes that there is nothing inherently wrong with collecting criminal evidence by FISA, but ensures that FISA will not be employed unless there is some worthy national security purpose.

Section 218 was perhaps legally unnecessary. The Justice Department, after all, could, absent legislation, have changed its internal guidelines and argued that FISA had been misconstrued. Yet, it was certainly apt for Congress itself to address a key cog of pre-9/11 intelligence failure. Furthermore, given that the FISA court, post-9/11, improperly attempted to institute the wall procedures as an exercise of judicial supervision, it was no doubt immensely significant to the Court of Review—in reversing the FISA court in 2002—that the wall had been rejected not just by DOJ but by the force of law.

Section 218 is vital. The sunset should be removed, and the provision should otherwise remain as is.

Imaginary Walls and Unnecessary Fixes by David Cole

Supporters of the Patriot Act often complain that critics have perpetuated myths about the Act, blaming the Act for more than it actually deserves. But supporters are equally guilty of propagating competing myths in this debate, nowhere more so than with respect to Section 218 and the "wall." Justice Department officials regularly credit Section 218, as has Andrew McCarthy here, with bringing down a "wall" constructed by the Foreign Intelligence Surveillance Act (FISA) that barred information sharing between criminal law enforcement officials and intelligence agents. But as McCarthy himself concedes, FISA did not in fact require a "wall" before Section 218 was enacted. As such, the amendment made by Section 218 was neither necessary nor sufficient to eliminate the barriers to information sharing captured in the image of the "wall." There were many impediments to information sharing before 9/11, but their cause was and remains today largely bureaucratic, not statutory. Moreover, the reform that Section 218 did actually make is of questionable constitutionality, because it permits searches undertaken primarily for criminal law purposes on less than the criminal probable cause showing that the Fourth Amendment requires for criminal law enforcement searches.

Section 218 made a very simple change to pre-existing law. Prior to the Patriot Act, FISA required that foreign intelligence gathering be "the purpose" of a FISA wiretap or search; Section 218 requires that foreign intelligence gathering be only "a significant purpose" of a FISA wiretap or search. Courts had interpreted the pre-Patriot Act language to require the government to show that the "primary purpose" of a FISA application was intelligence gathering rather than criminal law enforcement. Critics often dismiss that distinction as silly, because, as McCarthy argues, many terrorist crimes are matters of both criminal law and foreign intelligence. But the pre-Patriot Act law recognized that, and in no way precluded FISA searches simply because they might also have a criminal law purpose. The "primary purpose" test simply sought to reduce the risk that FISA, which permits searches on less than criminal probable cause, would become an end run around the constitutional requirement of criminal probable cause for searches conducted for criminal law purposes. The law acknowledged that investigations might have dual purposes, but contemplated that if the government's primary purpose was criminal law enforcement, it should be required to seek a warrant under the criminal probable cause standard. As long as the government's primary purpose was foreign intelligence gathering, it could obtain a warrant under the less stringent FISA standards, even if the search also had criminal law enforcement purposes. .

Contrary to myth, before Section 218, FISA did not mandate a "wall" between law enforcement officials and intelligence agents. It did not bar prosecutors or law enforcement agents from turning over information to intelligence agents when the information might be relevant to foreign intelligence. Nor did it stop foreign intelligence agents from sharing with criminal prosecutors evidence of crime that they had discovered in their investigations, whether under FISA or otherwise. FISA never prohibited the government from using evidence obtained in a foreign intelligence investigation against a defendant in a criminal case. In fact, the government regularly relied upon FISA-obtained evidence in criminal trials before passage of the Patriot Act. Thus, claims that FISA created a "wall" are a myth.

Equally mythical are claims that Section 218 made possible terrorist prosecutions that were not possible prior to its enactment. The Justice Department often points to the indictment of Sami Al-Arian, a University of South Florida professor accused of providing material support to the Palestinian Islamic Jihad, as an example of a prosecution it could bring only after Section 218 was enacted. But the government's case against Al-Arian is predicated on years of FISA wiretaps undertaken before the Patriot Act was enacted, and thus the validity of those wiretaps stands or falls under the pre-Patriot Act law, and is unaffected by Section 218's relaxation of the standards for wiretaps conducted after the Patriot Act was enacted.

I do not mean to suggest that there were no impediments to information sharing before 9/11. There were plenty. But their principal source was not FISA, but administrative and bureaucratic culture. The FBI did not trust the CIA, and vice versa. Agencies were engaged in turf wars, and there were few if any mechanisms or incentives in place to break down the institutional boundaries between agencies. Legitimate concerns about not revealing sources make information sharing difficult even in the most well organized operations. But the blame for these problems cannot be laid at the foot of FISA, and Section 218 was not their solution.

It is true that the Justice Department had developed bureaucratic rules that erected further barriers to information sharing where FISA investigations were ongoing, in part out of concern that without such barriers, it would have difficulty showing that the primary purpose of a FISA investigation was foreign intelligence gathering. But those barriers were not required by statute, and appear to have been fed as much by the institutional barriers identified above as by any legitimate concerns about FISA's "primary purpose" test.

Critics of the wall sometimes suggest that before the Patriot Act, once a foreign intelligence investigation became primarily a criminal investigation, the government would have to take down the tap. But that is also not true. Once an investigation became primarily criminal in nature, government agents would simply have to satisfy the standards applicable to criminal investigations—namely, by showing that they had probable cause that the tap would reveal evidence of criminal conduct. The tap or the search could then continue. If an investigation has become primarily criminal in nature, it should not be too much to ask that the government show probable cause of criminal conduct to carry out a search or wiretap.

Indeed, the Constitution demands no less. FISA's constitutionality turns on an untested assumption (because the Supreme Court has never passed on the validity of FISA) that the government may engage in searches and wiretaps for foreign intelligence purposes on a lower showing of suspicion than is required for criminal law investigations. FISA does not require the government to show probable cause that evidence of a crime will be found, but only probable cause that the target of the search is an "agent of a foreign power." "Foreign power" is in turn defined so broadly that it encompasses any political organization comprised of a majority of noncitizens. Where "U.S. persons" are the target of a FISA search, the government must make additional showings, but to search the home of a foreign national here on a work permit, for example, the government need only show that he's an employee of an organization made up principally of noncitizens. It need not show that the individual be engaged in any criminal wrongdoing whatsoever, much less terrorism.

If FISA searches are constitutional, then, they must be justified on the basis of some application of the "administrative search" exception to the general Fourth Amendment rule requiring probable cause and a warrant for criminal law enforcement searches. That exception permits searches in limited settings on less than probable cause where the search serves some special need beyond criminal law enforcement. The FISA Court of Review relied on precisely this exception to find FISA searches valid. But the Supreme Court has carefully limited the "administrative search" exception to situations in which the government is pursuing a special need divorced from criminal law enforcement—e.g., highway or railroad safety, secondary school discipline, or enforcement of an administrative regime. It has refused to apply the exception where the government is engaged in criminal law enforcement, as in a checkpoint to search for cars carrying drugs. And the Court has also refused to apply the exception where the government has a "special need," but is using criminal law enforcement to further that need. Thus, it struck down a hospital program that subjected pregnant mothers to drug tests for the ultimate purpose of protecting the health of the fetus, where the hospital shared the test results with prosecutors in order to threaten the mothers with criminal prosecution if they did not seek drug treatment.

Where an investigation becomes primarily focused on criminal law enforcement, therefore, the "administrative search" exception no longer applies, and Supreme Court doctrine would compel the government to meet the traditional standards of criminal probable cause. Before the Patriot Act, FISA conformed to that requirement. By abandoning that distinction and allowing searches on less than probable cause where the government is primarily seeking criminal prosecution, Section 218 raises a serious constitutional question. Thus, Section 218 was not only unnecessary to bring down the wall, but may render FISA unconstitutional.

Finally, while claims that Section 218 brought down the "wall" are greatly exaggerated, its enactment and other reforms have made it increasingly likely that information obtained through FISA wiretaps and searches will be used against defendants in criminal cases. In light of these developments, a useful reform at this point would be a provision permitting criminal defendants—or their cleared counsel—an opportunity to review the initial application for the FISA wiretap or search when contesting the admissibility of evidence obtained through a FISA search. Under current law, they have no such opportunity. Without access to the warrant application, defendants and their attorneys cannot meaningfully challenge the legality of the tap or search in the first place. And when government officials know that their actions will never see the light of day, they are more likely to be tempted to cut corners. An amendment requiring disclosure of FISA applications where evidence is sought to be used in a criminal trial would encourage adherence to the law by putting federal officials on notice that at some point the legality of the FISA warrant would be subjected to adversarial testing. Concerns about confidentiality could be met by limiting access to cleared counsel where necessary, and/or by applying the protections of the Classified Information Procedures Act. But there is no good reason for the current blanket exemption against the production of all such applications in criminal cases. The presumption should be in favor of adversarial testing where evidence is to be used in a criminal case.

A Response to Professor Cole by Andrew C. McCarthy

It is apt that Professor David Cole begins the title of his response "Imaginary Walls[.]" His submission is largely imaginary, creating rather than relating "myths" about the structural impediments to good intelligence that plagued the pre-9/11 world.

Prof. Cole's response is thoroughly ill-conceived. His basic premise, Section 218 aside, is that FISA itself is unconstitutional. Why? "[B]ecause the Supreme Court has never passed on [its] validity." This is an unworthy claim. As an act of congress, FISA is presumptively valid. More to the point, it has been upheld repeatedly by federal appeals courts. The Supreme Court hasn't had to assess it lo these last 27 years because no challenge to it has been sufficiently colorable. By Cole's logic, we should fret that any makeweight claim may be of constitutional moment if its emptiness, unsurprisingly, has led the Supreme Court to eschew entertaining it.

Prof. Cole's problem with FISA appears to be that his imaginary Fourth Amendment says searches are inappropriate absent probable cause of a crime. Of course, the actual Fourth Amendment prohibits only "unreasonable" searches, and the commission of crime is not a sine qua non of reasonableness. The American people obviously have a right to determine if non-Americans are engaged in covert intelligence gathering, or worse. The FISA search standard is not, as Cole insists, a "lower showing of suspicion" than is required in the criminal context; it is a different showing of suspicion—still requiring probable cause, but of foreign power agency rather than traditional crime.

The suggestion that a "foreign power" under FISA could be any "political organization" comprised predominantly of non-citizens is overwrought. One isn't told what Prof. Cole means by a "political organization"—Hezbollah, Hamas and Sinn Fein, for example, describe themselves as such—but the foreign powers FISA targets, by statutory definition, are those engaged in clandestine intelligence gathering activities, sabotage or international terrorism. See 50 U.S.C. §§ 1801(b)(2)(A), (C).

Atop his faulty premise that FISA is suspect, Prof. Cole piles the additional myth that Section 218 is suspect because it permits FISA searches and surveillance "undertaken primarily for criminal law purposes." To the contrary, as the FISA Court of Review reasoned in its 2002 opinion, FISA as written never limited the government to searches whose primary purpose was intelligence gathering. By mandating that intelligence gathering be "a primary purpose," Section 218 actually constrains the government in a way that neither the Fourth Amendment nor FISA does. Given that FISA easily passes Fourth Amendment muster, a provision such as Section 218, which narrows it, a fortiori is not constitutionally suspect.

Finally, Professor Cole's mythical account skews the history of the wall and the purpose of Section 218. Nobody is saying FISA "mandated a 'wall' between law enforcement officials and intelligence agents"—and when I argued that FISA did not require a wall, that palpably was an assertion, not, as Professor Cole oddly spins it, a "conce[ssion]." Yes, the wall was unnecessarily erected by the courts and the Justice Department. But that it was a mistake did not make it any less real. And that there may have been cultural impediments to intelligence sharing does not mean the structural ones manufactured by the wall were not critical.

Thus, to assert, as Cole does, that Section 218 did not significantly contribute to prosecutions like the Sami al-Arian case because FISA wiretaps were valid under pre-Patriot law entirely misses the point. No one claims the wall tainted the propriety of intelligence gathering. It blocked sharing of the intelligence gathered. That is the bureaucratic monstrosity dismantled by Section 218.

As a practical matter, Section 218 was crucial because, had it not been enacted, the litigation ending in the FISA Court of Review decision would not have happened. Without that clarification of law, the disastrous primary purpose doctrine would be undisturbed, the unnecessary wall would still be in place, dots would remain unconnected, prosecutions like al-Arian would not have occurred, and the United States would be at considerably greater risk.

David Cole The Last Word?

I did not argue that FISA, as amended by Section 218, is unconstitutional simply because "the Supreme Court has never passed on [its] validity," as Andrew McCarthy claims, but because the very purpose of Section 218's revision deprives FISA of its constitutional justification. FISA searches are generally defended as constitutional without probable cause of criminal activity because they fall under the "administrative search" exception to the probable cause requirement. But that exception, the Supreme Court has held, does not apply where the government's purpose is criminal law enforcement. FISA's pre-Patriot Act requirement that the search's primary purpose be foreign intelligence gathering, not criminal law enforcement, may well have been constitutionally required to make FISA searches valid. After the Patriot Act, foreign intelligence gathering need only be "a significant purpose" of the search—the primary purpose can be criminal law enforcement. That raises a serious constitutional concern.

McCarthy's only response is predicated on a demonstrably false premise. He says Section 218 is constitutional because it "mandat[es] that intelligence gathering be 'a primary purpose'" of the investigation. In fact, the very purpose of Section 218 was to eliminate the "primary purpose" requirement.

McCarthy's response is also erroneous in other respects. He falsely claims that FISA targets only those "foreign powers" engaged in intelligence gathering, sabotage or international terrorism, citing 50 U.S.C. §§ 1801(b)(2)(A), (C)). In fact, FISA also defines "foreign power" as "a foreign-based political organization, not substantially composed of United States persons." 50 U.S.C. § 1801((a)(5). An "agent" of a foreign power need only be "an officer or employee of a foreign power." Neither agent nor power need be involved in any wrongdoing. Thus, a British citizen working here as an employee of Amnesty International is an "agent of a foreign power." Surely such a definition is a little overbroad.

Finally, McCarthy offers no response whatsoever to my proposal that, given FISA's increased use in criminal prosecutions, it should be amended to permit defendants in those prosecutions access to the FISA applications to challenge the warrant's validity. We may well need FISA, but FISA also needs to be held accountable to the adversary process, particularly as it becomes unmoored from its initial justification, and instead becomes an end-run around the Fourth Amendment in criminal investigations.