Chairman, Ranking Member Leahy, and Members of the Committee, thank you
for inviting me to testify this morning on a subject of vital
importance to our Nation. The Chairman's bill and these hearings are,
in my view, key steps toward modernizing the Foreign Intelligence
Surveillance Act in a way that allows the President to protect our
people from attack while preserving the cherished liberties and
separation-of-powers that our Constitution demands. As a national
security and information security and privacy lawyer for most of my
career (serving six years in the Clinton Administration, and two years
in the George W. Bush Administration, as well as being a Member of the
Markle Foundation Task Force on National Security in the Information
Age and concentrating on these issues in private practice), I am
In addition to responding to your questions, I will address today three key issues. First, I will discuss the constitutional and legal analysis, based on numerous court decisions and Executive Branch legal opinions, leading me to the conclusion that -- depending upon the precise facts and circumstances of the Terrorist Surveillance Program (TSP) -- the program is constitutional and, therefore, legal, notwithstanding the current Foreign Intelligence Surveillance Act (FISA). This analysis, as I will explain, is not affected by the United States Supreme Court's recent decision in Hamdan v. Rumsfeld. Second, I will explain why I believe that this constitutional conclusion, as well as the fact that FISA, as currently written, cannot be carried out in the post-9/11 world in a way that protects our people, compels the need for legislation like the Chairman's bill, sooner rather than later. Third, I will provide some observations on the Chairman's draft bill itself.
At the outset, a brief word about bi-partisanship. In my career, I have served longer under a Democratic President than under Republicans. I commend Chairman Specter and other Members of Congress who have sought reasonable solutions to the problems we discuss today without regard to partisan concerns. I also commend the President and Attorney General for their willingness to work with this Committee and others to craft reasonable legislation. To state the obvious, your work today and in the future on this issue will affect the activities of future Presidents, of both parties, far more than President Bush. More importantly, our people expect their lives and their liberties to be protected regardless of which party is in power. Bipartisan consensus on the constitutional aspects of the TSP is represented by the common ground I suspect I will find today with former Clinton Administration Associate Attorney General John Schmidt, and have found with former Associate Deputy Attorney General David Kris, who served in senior FISA-related positions in both the Clinton and Bush Administrations, and who has testified recently before this Committee.
The President's Constitutional Foreign Affairs/Foreign Intelligence Authority to Authorize the Terrorist Surveillance Program
The hearing today, quite appropriately, is about the merits of proposed FISA reform legislation. I support the Chairman's bill, S. 2453, and will provide observations about the proposal later in my statement. It is first necessary, however, to summarize what I believe to be the proper constitutional framework for analyzing the TSP, as it is crucial to address squarely the President's constitutional authority to conduct the program. I reach this conclusion not because I believe the Chairman's bill is necessary to provide the President with legal authority to conduct the program. To the contrary, a proper understanding of the President's existing constitutional (and, therefore, legal) authority is necessary, I believe, precisely because the current and all future presidents do have the constitutional authority to carry out such a program, and our Supreme Court would uphold that authority, notwithstanding the Hamdan decision.
Because presidents have not only this authority, but the unique constitutional responsibility to protect us from attack, I believe that all presidents, of whatever party, would be using that authority today, will use it in the future, at least if we are still at war, and, indeed, would be horribly negligent in carrying out their responsibilities not to conduct such a program. Therefore, because I also believe that our security and liberty is best protected with the active involvement of all three branches of government, I support the Chairman's bill to help enhance the involvement of Congress and the Judiciary in the TSP, a program to protect us from attack from attack that clearly is going to continue whether or not FISA is amended.
Both the Administration and its opponents have discussed the President's authority to authorize the TSP principally in the context of Article II, Section 2, of the Constitution, in which the President is given authority as "the Commander in Chief of the Army and Navy of the United States." This focus is understandable, given the apparent recognition by all sides of the debate that the TSP is part of the ongoing global military campaign against terror announced by the President in response to al Qaeda's attacks on our homeland, and recognized by Congress in the AUMF.
The focus to date on the Commander-in-Chief power is, at best, incomplete, however, because it fails to give due weight to a critical series of United States Supreme Court and other federal court decisions and Executive Branch legal opinions analyzing the President's constitutional authority to control foreign intelligence operations such as the TSP. Based on this ample and significant precedent, it is clear that programs such as the TSP fall principally under the President's constitutional foreign affairs authority. In my view, the perplexing failure to recognize this fact, or even to address this extensive, and directly applicable, precedent (or only cursorily to do so), fatally undermines much of the published constitutional analysis of the TSP to date, including by the small task force selected by the President of the American Bar Association, the Congressional Research Service, and by a number of self-described constitutional scholars.
The United States Constitution, in its text, places the duty on the President -- and only the President -- to "preserve, protect, and defend the Constitution." At least since 1898, Supreme Court authority, and Executive Branch opinions under both political parties relying on such authority, has recognized that the President has the first, strongest, and most direct authority and responsibility for the protection of our national security, and that this authority and responsibility flows, at least in significant part, from the President's "plenary" authority over the conduct of our foreign affairs.
Supreme Court decisions over many decades strongly support this view, including, importantly, a number of cases decided both before and after Justice Jackson's famous concurring opinion in Youngstown Sheet and Tube v. Sawyer, the touchstone for most opponents of the TSP. For example, in Department of the Navy v. Egan, Justice Harry Blackmun, writing for the majority, reiterated that the "Court . . . has recognized ‘the generally accepted view that foreign policy was the province and responsibility of the Executive.'"
Perhaps most famously, the Supreme Court forcefully affirmed, in its 1936 decision in Ú.S. v. Curtiss-Wright Export, the:
delicate, plenary and exclusive power of the President as the sole organ of the federal government in the field of international relations--a power which does not require as a basis for its exercise an act of Congress, but which, of course, like every other governmental power, must be exercised in subordination to the applicable provisions of the Constitution.
Not only has the Supreme Court never reversed its holding in Curtiss-Wright, but a Westlaw search indicates it has been cited by our courts well over 150 times, including in numerous cases decided well after Youngstown.
To cite but two additional examples, both from the so-called Pentagon Papers case, Justice Stewart, in his concurring opinion joined by Justice White, noted that:
In the governmental structure created by our Constitution, the Executive is endowed with enormous power in the two related areas of national defense and international relations. This power [is] largely unchecked by the Legislative and Judicial branches.
Similarly, Justice Thurgood Marshall, concurring, stated that it: "is beyond cavil that the President has broad powers by virtue of his primary responsibility for the conduct of our foreign affairs and his position as Commander in Chief."
While Congress has, of course, certain enumerated powers under Article I, to declare war, to raise and support the Army, provide for a Navy, and the like, it is the President, then, who -- in addition to his express powers to make treaties, appoint and receive ambassadors, and serve as Commander-in-Chief -- has the "plenary and exclusive" power to conduct foreign affairs, as intended by the framers of our Constitution.
As firmly established is the President's plenary constitutional position in foreign affairs generally, however, it is even stronger in the conduct of foreign intelligence operations, such as the TSP. In a 1988 decision upholding the authority of the President's senior intelligence official to terminate the employment of a CIA officer's employment, on sexual preference grounds, Justice O'Connor stated:
The functions performed by the Central Intelligence Agency and the Director of Central Intelligence lie at the core of "the very delicate, plenary and exclusive power of the President as the sole organ of the federal government in the field of international relations."
Federal appeals courts ruling on the President's authority to conduct foreign intelligence electronic surveillance operations have recognized the President's constitutional preeminence in the collection of foreign intelligence to protect our national security. These decisions, all decided well after Justice Jackson's concurring opinion in Youngstown, recognize the President's core authority in the conduct of foreign intelligence operations such as the TSP. Notably, they also stress the fundamental difference between purely, or primarily, domestic cases, such as Youngstown, and those involving the collection of intelligence regarding foreign threats to our nation's security, such as the TSP, and have looked with disfavor on legislative restrictions on the latter.
More than 20 years after Youngstown, the Fifth Circuit Court of Appeals, in United States v. Brown, upheld the President's inherent constitutional authority to authorize warrantless wiretaps for foreign intelligence purposes, explaining that:
[B]ecause of the President's constitutional duty to act for the United States in the field of foreign relations, and his inherent power to protect national security in the context of foreign affairs, we reaffirm. . . that the President may constitutionally authorize warrantless wiretaps for the purpose of gathering foreign intelligence. Restrictions upon the President's power which are appropriate in cases of domestic security become artificial in the context of the international sphere. Our holding . . . is buttressed by a thread which runs through the Federalist Papers: that the President must take care to safeguard the nation from possible foreign encroachment, whether in its existence as a nation or in its intercourse with other nations.
Similarly, in United States v. Truong Dinh Hung, a case cited with approval in 2002 by the Foreign Intelligence Surveillance Court of Review, the Court of Appeals for the Fourth Circuit, in approving warrantless electronic surveillance for foreign intelligence purposes, stated the matter plainly:
Perhaps most crucially, the executive branch . . . is . . . constitutionally designated as the pre-eminent authority in foreign affairs . . . . Just as the separation of powers in Keith forced the executive to recognize a judicial role when the President conducts domestic surveillance, so the separation of powers requires us to acknowledge the principal responsibility of the President for foreign affairs and concomitantly for foreign intelligence surveillance.
The special Foreign Intelligence Surveillance Court of Review apparently agreed, as it recognized, in a post-FISA, and 9/11, case, that:
[t]he Truong court, as did all the other courts to have decided the issue, held that the President did have the inherent authority to conduct warrantless searches to obtain foreign intelligence information. . . . We take for granted that the President does have that authority and, assuming that is so, FISA could not encroach on the President's constitutional power.
Separation of Powers: The Decisive Constitutional Principle
Virtually everyone who has taken a position on the TSP's constitutionality agrees that the fundamental constitutional principle of separation of powers is a decisive part of the analysis, though there is strong disagreement as to which way that principle cuts. Conceding, for the sake of argument, that Congress has attempted to occupy the field for intercepting international communications through FISA as the "exclusive means," and the President's exercise of any power in this area is at Youngstown's "lowest ebb" from a separation-of-powers perspective, if FISA impermissibly interferes with the proper execution of the President's own authorities and responsibilities as Chief Executive, Commander-in-Chief, and "sole organ" of foreign relations, FISA is unconstitutional as applied.
Our Supreme Court, long after Youngstown, has repeatedly rejected the approach many have taken in applying Justice Jackson's concurrence to the TSP, namely, that, once a President is at his "lowest ebb," Congress' will must prevail regardless of the constitutionality of a statute as applied to a particular set of facts. Rather, the Supreme Court has made clear that a balancing of powers approach must be used. In Nixon v. Adm'r of General Services, for example, the Court held that "in determining whether [a legislative] Act disrupts the proper balance [of power] between the coordinate branches, the proper inquiry focuses on the extent to which it prevents the executive from accomplishing its constitutionally assigned functions."
In Public Citizen v. United States, the Supreme Court was faced with the issue of whether the Federal Advisory Committee Act (FACA), which established procedures by which the Executive branch utilizes private advisory committees, was constitutional as applied to a putative private advisory committee formed by the American Bar Association (ABA) that advised the President on the qualifications of potential federal judicial nominees. Under Article II, the President has the power to appoint judges, but the Senate also has a clear and important advise and consent power. The majority in Public Citizen recognized that applying FACA in this context raised serious separation-of-powers questions, and, invoking the constitutional avoidance doctrine, ruled that Congress intended to have FACA apply only to advisory committees that were established or controlled by the Executive. Id. at 461.
Accordingly, because the ABA committee was not established or controlled by the Executive, FACA did not apply in this case. Justice Kennedy, however, joined by then-Chief Justice Rehnquist and Justice O'Connor, all concurring in the result, found it necessary to reach the constitutional question, and stated that applying FACA to the manner in which the President obtains advice on potential nominees would violate the separation of powers:
In some of our more recent cases involving the powers and prerogatives of the President, we have employed something of a balancing approach, asking whether the statute at issue prevents the President 'from accomplishing [his] constitutionally assigned functions' . . . . and whether the extent of the intrusion on the President's powers "is justified by an overriding need to promote objectives within the constitutional authority of Congress."
Applying FACA to the appointments process surely would not prevent the President from nominating whomever he chose to be a federal judge. Nevertheless, Justice Kennedy recognized that FACA's impairment of the exercise of even a small part of that presidential power -- namely, the ability to receive unfettered advice from the private sector in the aid of his appointment power -- was sufficient to disable the legislative branch from regulating the exercise of that power.
It cannot be seriously doubted that applying FISA to preclude the TSP would impair the execution of a core constitutional duty of the President to a much greater degree than would have been the case in applying FACA to the ABA advisory committee.
Presidents of both political parties -- and their senior most legal advisers -- have recognized not only the necessity of such a balancing analysis, but the constitutional authority and, perhaps, responsibility, of a president faced with a statute unconstitutional as applied to particular facts and circumstances. To cite just one example, albeit a highly pertinent one, the constitutional power of the President to gather and share intelligence information was recognized by the Clinton Administration with regard to a statutory preclusion on the sharing of intelligence information gleaned from a criminal wiretap carried out up under Title
III. As aptly summarized in a 2000 Office of Legal Counsel Opinion for President Clinton's Administration:
In extraordinary circumstances electronic surveillance conducted pursuant to Title III may yield information of such importance to national security or foreign relations that the President's constitutional powers will permit disclosure of the information to the intelligence community notwithstanding the restrictions of Title III. . . . Where the President's authority concerning national security or foreign relations is in tension with a statutory rather than a constitutional rule, the statute cannot displace the President's constitutional authority and should be read to be "subject to an implied exception in deference to such presidential powers." Rainbow Navigation, Inc. v. Department of the Navy, 783 F.2d 1072, 1078 (D.C. Cir. 1986) (Scalia, J.). We believe that, if Title III limited the access of the President and his aides to information critical to national security or foreign relations, it would be unconstitutional as applied in those circumstances.
As discussed above, the conduct of foreign intelligence operations is a core constitutional function of the President, which Congress may not constitutionally impair. Recognition of this constitutional fact may have led the Congress that passed FISA to state, even as it was passing the law, that:
The conferees agree that the establishment by this act of exclusive means by which the President may conduct electronic surveillance does not foreclose a different decision by the Supreme Court. The intent of the conferees is to apply the standard set forth in Justice Jackson's concurring opinion in the [Youngstown] case.
This statement, by the Congress that passed FISA, is significant for several reasons. First, it acknowledges that Congress itself had some doubt about the constitutionality of FISA's attempt to completely control the President's authority to conduct electronic surveillance for foreign intelligence purposes. Second, it suggests that Congress understood that, even within Justice Jackson's zone of "lowest ebb," there are limits to the degree to which Congress may constitutionally restrict the President in the area of foreign intelligence collection. Finally, the statement indicates that Congress specifically contemplated that the degree to which FISA might constitutionally tie the President's hands could one day reach the Supreme Court. This makes sense if, but only if, Congress contemplated that then-President Carter, or a future President, might be required to act outside the FISA statute, exercising the very inherent authority that Congress was attempting to limit.
How, then, to balance "core" presidential authorities with Congress' far weaker ones, but in the context of Congress' clear intent to "occupy the field" and foreclose the President's options? One solid framework is that articulated by former Clinton and Bush FISA-expert David S. Kris in his March 28, 2006 testimony before this Committee:
[R]eal and hypothetical examples illustrate what Professor Corwin famously called the Constitution's "invitation to struggle" for dominance in foreign affairs. Depending on the vigor of the struggling parties, I believe the constitutional (and perhaps political) validity of the NSA Program will depend on two operational questions. The first question concerns the need to obtain the information sought (and the importance of the information as compared to the invasion of privacy involved in obtaining it). To take a variant on the standard example as an illustration of this point, if the government had probable cause that a terrorist possessed a nuclear bomb somewhere in Georgetown, and was awaiting telephone instructions on how to arm it for detonation, and if FISA were interpreted not to allow surveillance of every telephone in Georgetown in those circumstances, the President's assertion of Article II power to do so would be quite persuasive and attractive to most judges and probably most citizens. The Constitution is not a suicide pact. . . . The second question concerns the reasons for eschewing the use of FISA in obtaining the information.
This is, in my view, the correct analysis. The final answer, of course, is unknowable without many more facts than currently available, at least to me. Based on the factual assumptions I have made, the considerations discussed below, and the weight of authority by our Supreme Court and other courts, as well as more than 200 years of Executive practice, the TSP easily satisfies this test and a proper constitutional analysis leads to but one conclusion. FISA is unconstitutional as applied to the TSP to the extent it impermissibly impedes the President's ability to carry out his constitutional responsibilities to collect foreign intelligence and protect our Nation from attack.
Hamdan v. Rumsfeld Does Not Alter This Conclusion
Within days of publication of the Supreme Court's voluminous, and somewhat baroque, opinion in Hamdan v. Rumsfeld, opponents (and some supporters) of the TSP announced that this decision undermines, perhaps fatally, the statutory and constitutional underpinnings of the TSP. This conclusion is remarkable and, at least with regard to a proper constitutional analysis is, in my view, flatly mistaken, for several reasons.
First, the majority opinion in Hamdan is overwhelming a statutory one, delving deeply into the details of the Detainee Treatment Act, the Authorization to Use Military Force, and the like. The Hamdan majority clearly does not to reach significant separation-of-powers questions. By contrast, at least under the analysis I put forward today, the question of the legality of the TSP is almost entirely a constitutional one.
In the few paragraphs in which the majority opinion discusses issues arguably related to separation of powers, if anything, Justice Stevens, speaking for the majority, reinforces the analysis discussed in my testimony today, as he quotes from Ex parte Milligan:
[N]either can the President, in war more than in peace, intrude upon the proper authority of Congress, nor Congress upon the proper authority of the President.
This, of course, is precisely the position I take. In the context of the TSP, however, based on the facts assumed herein, it is FISA, as applied, that "intrudes" upon the "proper authority of the President." As such, as taught more than two centuries ago by legendary Chief Justice John Marshall, any attempt by Congress to so "intrude," through FISA, on the President's constitutional authority, is "entirely void." Justice Stevens appears to acknowledge this constitutional truism by noting that the President may not disregard statutory limitations placed "in the proper exercise" of Congress' own powers.
Even Justice Kennedy's plurality concurring opinion, most often cited as arguably undercutting the constitutional basis for the TSP, clearly supports the separation-of-powers analysis discussed above. Justice Kennedy declares, again based on Ex parte Milligan, that:
Subject to constitutional limitations . . . Congress has the power and responsibility to determine the necessity for military courts, and to provide the jurisdiction and procedures applicable to them.
It is worth noting that this quoted passage by Justice Kennedy appears in the very same paragraph as the language cited for the proposition that Hamdan undermines the TSP's constitutionality, though it is not quoted by them. Also worth noting is that Chief Justice John Roberts did not participate in Hamdan because he had ruled (with the government, as it happens) on that very case as a D.C. Circuit Court of Appeals judge. It seems unlikely, at best, that, in a challenge to the TSP, on which the Chief Justice would vote, Justice Kennedy's Hamdan concurrence would gather a fifth vote.
Far more importantly, even if one were to assume that Hamdan intended to reach the constitutional separation-of-powers question, it would no more undercut the constitutionality of the TSP than does Youngstown itself, for both the Hamdan majority and the Kennedy plurality clearly rest their analysis on the considerable independent authorities explicitly committed to Congress in the text of the Constitution itself, in the specific context of providing law governing the conduct of military justice. Thus, Justice Kennedy relies on "a long tradition of legislative involvement in matters of military justice." Id. Justice Stevens, for the majority, relies on Congress' express constitutional authorities to: "declare War . . . and make Rules concerning Captures on Land and Water . . . [and] make Rules for the Government and Regulation of the land and naval Forces," Congressional authorities all directly relevant to the activities at issue in Hamdan.
In stark contrast, no such textual power or "long history" exists for Congress in the areas of foreign intelligence collection programs such as the TSP (or even in foreign affairs generally). Quite the contrary. As demonstrated by even the cursory discussion in my testimony of a few of the many relevant cases, the "core" constitutional authority for such activities is, and must be, vested almost exclusively in the President.
Thus, were the Supreme Court (even, or perhaps especially, in the wake of Hamdan) to conduct a separation-of-powers analysis, based on its own balancing test, of facts and circumstances in which FISA, as applied to the post-9/11 world, impairs or impedes the President's ability to carry out such a "core" function, it is likely the Court would find FISA itself, as applied to the facts and circumstances of the TSP, and not the TSP, unconstitutional and, as such, "void."
The Need for S. 2453 and Some Observations
Precisely because I believe the TSP already is constitutional (and, therefore, legal), and critical for this and future Presidents in preventing attacks on our people, I support Chairman Specter's draft legislation, S. 2453. The bill, or one very much like it, is critical, in my view, to ensure that constitutional activities necessary to protect our people from attack, which will continue with or without FISA reform, are afforded the meaningful participation of all three branches of our government.
Programmatic Judicial Review and Approval
On February 5, 2005, along with former Democratic House of Representatives staff member Daniel Prieto, I published an Op-Ed recommending, among other measures to protect civil liberties, that Congress and the President, in reforming FISA:
Ensure a role for the courts. To preserve and promote appropriate judicial oversight, new methods of court involvement must be considered. As one example, courts could pre-approve categories of electronic surveillance. This would allow the government to apply strict, pre-determined criteria to particular communications without the need for case-by-case court approvals. Categories, criteria and eavesdropping activity would be subject to regular re-examination, with approvals subject to periodic court renewals.
Based on my review of S. 2453, creating clear jurisdiction for the Foreign Intelligence Surveillance Court (FISC) to conduct just such "programmatic" review, approval, and oversight is the central function of the legislation. As such, I strongly support the bill, and believe it is absolutely necessary.
S. 2453 clearly establishes jurisdiction in the FISC to approve an "electronic surveillance program," upon application of the Executive Branch, including information required by the statute, and if the FISC makes a series of required judicial findings. This type of programmatic approval -- that is, a system of prior judicial approval of a collection program (rather than individual communications intercepts) based on criteria articulated in statute, will help ensure the participation of our Judiciary, and, through articulating the required criteria, of Congress. With clearly established criteria, I believe -- although there is room for reasonable lawyers and judges to differ on this -- that FISC orders approving (or denying) TSP applications likely would satisfy the Constitution's "case and controversy" requirement.
S. 2453 avoids what I believe to be one of the fatal flaws of the 1978-era FISA, namely the requirement for individualized, target-by-target approval, based on known facts which often, in the post-9/11 world, will be unknown in any timely fashion, and perhaps unknowable given the technology and enemies we now face.
Selected Reasons FISA is Unworkable Today
As has been discussed publicly by technically and legally knowledgeable experts, there are a host of technological developments which have rendered FISA, as currently drafted, unworkable against the post-9/11 terrorist threat to our Nation, including the development of "packet-based" communications, the use of proxy servers and Internet-based, encrypted, highly mobile telephone communications and PDAs, and the routing of vast amounts of purely overseas Internet communications through the United States.
In my view, equally fatal to the ability of any President to comply with all of the substantive and procedural requirements of the 1978 FISA is the current statute's target-by-target dependence upon two principal factors for determining the predicates necessary for approval of intercepts: (1) whether or not a potential target is a known or presumed United States Person (a citizen or Permanent Resident Aliens); and (2) whether the collection of information takes place within the territory of the United States or overseas. Based on my personal experience across multiple decades and Administrations, these two criteria are no longer workable in the post-9/11 world. Because these two pieces of information often will be unknowable given today's (and tomorrow's) technology -- or at least unknowable in a timely enough way to secure FISA warrants to capture brief but crucial terrorist attack warning information -- FISA, as written, easily could impede (indeed, prevent entirely) the President from carrying out his constitutional duty to prevent attacks.
I am not alone in my view as to the feasibility of using these factors in the post-9/11 world. For example, the non-partisan Markle Foundation Task Force on National Security in the Information Age, on which I was proud to serve with two of my fellow panelists today, concluded, after several years of factual research and legal and policy debate, that these two standards are "outdated." I must be clear that the Task Force, in reaching these conclusions about the unworkability of the U.S. Person and place-of-collection rules in the post-9/11 world, did so only with regard to access to, and sharing of, information. The Task Force explicitly did not reach a conclusion on this issue with regard to the collection of information. But to me, not purporting to speak for the Task Force, the logic is inescapable. If technology has made it unworkable, in a timely fashion, to determine U.S. Person status or place of collection for sharing of information, after the fact of collection, these problems are, if anything, far more clearly unworkable with regard to collection, which must, of necessity, be done in a far more rapid, time-sensitive fashion than the sharing of similar information.
In my view, programmatic approval of the type to be established by S. 2453 would go far towards ameliorating these, and other, fatal technical and legal problems in the current FISA by not mandating case-by-case substantive requirements and procedures inconsistent with today's technology and the enemies we face.
Some Members of Congress and others have argued that there is no information on the public record suggesting that FISA, as currently drafted, is unworkable. Having served as Assistant General Counsel for the Central Intelligence Agency in the Clinton Administration, in peacetime, involved in delicate negotiations with Congress as to how to gain needed legal change without exposing to our enemies the precise vulnerabilities we were attempting to cure, I can well understand the reticence of the Administration to discuss in public its precise problems with FISA.
It doesn't take classified information, however, to readily envision the problems, as many already have. In addition to the issues identified above, I would suggest one hypothetical, among many potential ways FISA almost certainly is unworkable as currently written. Imagine our government intercepts a communication from al Qaeda senior operational planner Ayman al-Zawahiri, to Bryan Cunningham in Denver. Zawahiri tells me ten nuclear devices are to be set off in five minutes and I'm to call the ten presumed-U.S. Person al Qaeda operatives in the United States and order then to do so. Even under the emergency Attorney General approval provisions of FISA, it would be literally impossible to gather enough information, process it, and submit it for the Attorney General's approval in time to intercept those following ten telephone calls.
The failure to recognize this obvious scenario in many quarters, I fear, results from a serious misunderstanding about the emergency approval provisions of the current FISA. What many apparently fail to understand is that the NSA may not lawfully listen to a single syllable of an "electronic communication" under FISA, until the Attorney General approves (most often in writing) an emergency order. My example obviously is extreme, but it is easy to imagine that the approval process, in many cases, would not be possible even with hours available for approval (e.g., a CIA officer finds a laptop in an al Qaeda camp with hundreds of U.S. telephone numbers, and evidence of an imminent attack that might be triggered by calls between or among any of them).
Based on my experience, and my current private practice focus on information security and technology, I also can imagine a number of purely technological advances possibly made by the government which could enable collection activity outside the strictest interpretation of the 1978-era FISA, but which could be cured by S. 2453. I do not feel it wise, however, to discuss these in open testimony because I do not want to risk alerting our enemies to capabilities of which they may not be aware.
S. 2453 wisely, in my view, recognizes the concept of "electronic tracking," as "the acquisition by an electronic, mechanical, or other surveillance device" of certain electronic communications. The draft legislation appears to recognize such tracking as an integral part of an electronic surveillance program eventually leading to the access to the contents by human beings of a far fewer number of selected communications than those triaged by computer. This is a crucial distinction, and one that, as technology has evolved, clearly needs recognition in our electronic surveillance laws.
I believe that the use of machines to triage communications content and other sensitive, i.e., personally identifiable, information prior to human review will be crucial over the coming years in balancing privacy and civil liberties and our national security. Depending upon one's interpretation of the current FISA, such "machine triage" -- the use of which bi-partisan experts, including the Markle Task Force, have recommended -- might today require individual FISA applications. Such a situation, obviously, would present an insurmountable obstacle to the use of machine triage that could enhance civil liberties and operational capabilities by reducing dramatically the volume of information that must be reviewed by our perennially resource-starved law enforcement and intelligence agencies.
Necessity of Section 801
Section 801 of S. 2453 states: "Nothing in this Act shall be construed to limit the constitutional authority of the President to collect intelligence with respect to foreign powers and agents of foreign powers." The bill also includes several other provisions that appear intended to make clear that electronic communications of foreign powers, or their agents, may be lawfully intercepted under specific statutory authorities or under the "Constitution of the United States."
These amendments -- as well as any others necessary to conclusively reverse any implication of FISA being the "exclusive means" for such collection -- are, in my judgment, critical to the passage and effective implementation of any FISA reform legislation. This is so for at least four reasons.
First, restoring the pre-FISA Congressional statement of the law of foreign intelligence electronic surveillance would recognize, in our public statutes, what I believe, as outlined earlier in my testimony, to be the most accurate statement of the balance of legal authorities in this "core" area of Presidential authority. Second, given the view of the current Administration (and, I believe, of any future President) concerning the constitutional prerogatives and obligations in the area of foreign intelligence collection to prevent attacks on our Nation, it is unlikely any FISA reform legislation not clearly restoring this proper constitutional balance would survive a veto. Thus, inclusion of such language likely is necessary to gain passage of legislation achieving the vitally important goal of enhancing involvement of all three co-equal branches of government in the TSP.
Third, this language will ensure that no future President, of either party, in the event that technology or the nature of our enemies change again significantly enough that FISA, as modified by S. 2453, unconstitutionally impedes or impairs that future President's ability to carry out his, or her, constitutional responsibilities, will face the Hobson's choice of deliberately failing to protect us from attack or being accused of violating the law. Finally, and importantly, codifying Congressional support for constitutionally permissible measures to protect our Nation from attack will provide our career intelligence officers with the assurance they need that no branch of government will second guess their actions or punish them after the fact for acting lawfully. This is crucial to reduce the "risk aversion" for which the Clinton and pre-9/11 Bush Administrations were properly criticized by members of both political parties and several independent Commissions.
Document Management System
S. 2453 also calls for the creation of a "document management system" for processing applications for FISA orders. This is an important step to expedite the processing not only of applications, if S. 2453 is enacted into law, under the TSP, but traditional FISA applications, which will continue to be crucial elements of protecting our national security and carrying out our Nation's other foreign policy interests. Any reasonable steps to streamline what remains a far-too-slow approval process are welcome. No combination of such steps, alone, however, can remedy the fatal flaws, outlined above, in the 1978-era Foreign Intelligence Surveillance Act.
Some Issues for Further Deliberation
Obviously, prior to, and after, approval by this Committee, S. 2453 will face a series of additional deliberations, including in the Intelligence Committee and on the floor of this body, in the other body, and in conference. As that process unfolds, I encourage Members to consider the following steps, to be addressed in the language of the legislation, and/or by way of explanation in its legislative history:
While it is axiomatic that a "state of war is not a blank check for the President," it is also true, as Justice Jackson himself warned, that the courts should not "convert the constitutional Bill of Rights into a suicide pact." From the limited information available about the TSP, it appears that the TSP is within the President's constitutional authority to carry out limited electronic surveillance of suspected terrorists who would attack this Nation and kill our people. This President and, I dare say, his successors, will continue with this program, or programs like it, if they believe their constitutional duties require it. Therefore, and for a host of other reasons related to the failure of FISA to keep pace with the evolution of technology and the threats to our people, rapid amendment to FISA is vital, and I support the efforts of the Chairman, other Members of Congress, and the Administration, to do so.