I. Introduction and Background
When civil litigation threatens to expose national security secrets, the government may attempt to quash any evidence involving sensitive national security matters through the state secrets privilege (“privilege”)1 whether it is the defendant in the suit or appearing as an intervener.2 The privilege goes as far as allowing complete dismissal of claims that cannot be untangled from disclosure of classified national security material.3 Although the privilege formulated by the Supreme Court has been utilized in American litigation for over sixty years4 and its role in American law dates back all the way to Aaron Burr’s treason trial,5 a circuit split has developed. The U.S. Courts of Appeal apply one of two theories to support the privilege—as the notable national security law professor Robert Chesney frames it, the privilege is either “a constitutional rule derived from the separation of powers” or “merely a common law rule of evidence of no greater stature than, for example, the spousal privilege[.]”6
The privilege in its modern form was set out by the Supreme Court in Reynolds, a Federal Tort Claims Act case concerning Air Force civilian deaths during equipment testing.7 The Court summarized an ex-parte, in-camera procedure and test for invoking the privilege.8
The privilege belongs to the Government and must be asserted by it; it can neither be claimed nor waived by a private party. It is not to be lightly invoked. There must be formal claim of privilege, lodged by the head of the department which has control over the matter, after actual personal consideration by that officer. The court itself must determine whether the circumstances are appropriate for the claim of privilege, and yet do so without forcing a disclosure of the very thing the privilege is designed to protect.9
Since Reynolds, and especially as the War on Terror has expanded, the government has liberally used the privilege to try to suppress suits against the United States involving national security operations, including controversial actions such as the rendition and torture program.10 Although Congress passed the Classified Information Procedures Act11 to handle evidentiary defense issues in criminal litigation that involves classified information12 the privilege continues to be the procedure for civil litigation.
This article discusses the split between the constitutional and common law theories of the state secrets privilege. It surveys the split by examining litigation in which each theory was adopted by one of the U.S. Courts of Appeal. It then summarizes arguments in favor of the constitutional basis for the privilege and ultimately suggests that constitutional law may be privilege’s appropriate home.
II. The circuit split over the basis of the privilege between the Fourth and Ninth Circuits.
In the Fourth Circuit, the privilege finds its home in the Constitution. El-Marsi, a German citizen, sued the government for wrongful subjection to the CIA’s rendition and torture program.13 The trial court dismissed his compliant after the government asserted the privilege.14 In holding that the privilege was applicable even though the rendition program was public, Judge Ellis surveyed the Constitution and found the privilege by looking to the “president’s constitutional authority over the conduct of this country’s diplomatic and military affairs.”15
On appeal, the Fourth Circuit upheld Judge Ellis’ order. The Fourth Circuit found that the privilege had a “a firm foundation in the Constitution, in addition to its basis in the common law of evidence by reviewing cases concerning national security, foreign affairs, and separation of powers.16 Relying on dicta in United States v. Nixon and cases on intelligence activity, the court stated that the executive has a to collect intelligence and protect that information in furtherance of his role as Commander in Chief.17 The Supreme Court denied a petition for certiorari after the Fourth Circuit’s decision.18
The Ninth circuit takes a different approach, holding that the privilege is situated solely in the common law.19 The Ninth Circuit specifically declined to adopt the constitutional view of the state secrets privilege in Fazaga, and asserted that it is a common law privilege only.20 The Fazaga litigation concerns a suit by three Muslims who claim the FBI targeted them for surveillance solely based on their religion.21 In response, the government moved to dismiss the various claims Fazaga raised, including an argument that for some, the privilege required dismissal.22
On appeal, Fazaga, like El-Marsi, was a review of a civil liberties suit dismissal after the state secrets privilege was invoked, but with a twist: the plaintiffs were suing under Foreign Intelligence Surveillance Act (FISA), with its own cause of action and accompanying secrecy procedures.23 More plainly stated, the issue was whether FISA preempted the state secrets privilege.24 The court did not mince words in rejecting the constitutional theory of the privilege, stating explicitly that “it is an evidentiary rule rooted in common law, not constitutional law.”25 It additionally cited to its earlier precedent that “the state secrets privilege is an evidentiary privilege rooted in federal common law.”26 The court repudiated the Fourth Circuit and the El-Marsi holding, stating that “El-Masri d[id] not specify a clear statement rule; it sp[oke] generally about the constitutional significance of the state secrets privilege.”27 The Ninth Circuit found that the privilege had “a constitutional ‘core’ or constitutional ‘overtones,’”28 yet insisted that it was still a common law rule. It therefore held that “in enacting FISA, Congress displaced the common law dismissal remedy created by the Reynolds state secrets privilege as applied to electronic surveillance within FISA’s purview.”29
Thus, two highly divergent views on the nature of the state secrets privilege have arisen in American jurisprudence. In trying to reconcile them prior to the Fazaga case, Professor Chesney has suggested “[a] careful review of the origin and evolution of the privilege suggests that both explanations are true to some extent”30 and that the privilege might exist in a hybrid state.31 Though the origins of the privilege may be traceable through both theories, as a matter of first principles the privilege can either be based in the executive power or not–there is no true middle ground in this arena according to the Fazaga court.32 Only the Supreme Court will be able to answer the question as to whether the state secrets privilege is a component of executive privilege.
III. The privilege is most likely an outgrowth of the executive privilege when considering the structure of the Constitution, its historical origins, and separation of powers concerns.
Is the Fazaga court correct in its argument that the privilege is rooted in the common law? As discussed above, the answer matters a great deal in terms of the scope of Congress’ power. There are three veins of argument that cut against the common law basis for the privilege. First, the structure of Article II of the Constitution points towards a privilege stemming from the executive power. Second, the history behind the privilege similarly points to a more constitutional basis. Finally, concerns over the current status of separation of powers supports policy arguments behind a constitutional state secrets privilege. This section summarizes each of those contentions in turn.
The Fazaga Court was too dismissive of the Fourth Circuit’s constitutional approach.
The Fourth Circuit held in El-Marsi that the common law origins on the state secrets privilege did not change the fact that it was constitutional in nature.33 The Fazaga Court was bold in its rejection of the constitutional theory of the state secrets privilege, attributing the Constitution’s role to providing mere overtones.34 The court did not fully explain why those overtones failed to persuade it that the privilege was a part of the broader Article II executive privilege.
Instead, the Ninth Circuit argued that the privilege belonged in the common law of evidence based on dicta in another Supreme Court case: “[t]he Supreme Court . . . has [explained] that Reynolds ‘decided a purely evidentiary dispute by applying evidentiary rules.’”35 This argument is flawed, however, as evidentiary rules on privilege would of course include constitutionally based privileges.36 To say that Reynolds involved evidentiary rules does nothing to illuminate whether the evidentiary rule it involved was based on the Constitution or another source. Thus, the crucial analysis is whether the “overtones” of the privilege grant it constitutional status.
Supreme Court jurisprudence heavily utilizes such “overtones” in deciding questions of constitutional law. The most famous instance is Justice Douglas’ opinion in Griswold, relying on “penumbras” in parts of the Constitution and “emanations” from those penumbras to read a right to privacy into the document.37 This structuralist approach to constitutional law has been employed time and again to shed light on different areas of constitutional law, including issues beyond the scope of the Bill of Rights, ranging from interstate travel to federalism.38 This constitutional approach has also been used in national security jurisprudence. For example, in Zivotofksy v. Kerry, Justice Kennedy held that the president has the implied power to recognize foreign countries based on the express presidential powers in Article II, Section 2 of the Constitution to make treaties and to nominate and receive ambassadors.39
If one accepts that reliance on overtones of the Constitution’s text is a valid method of resolving constitutional questions, the Fourth Circuit and trial court’s survey of Article II in El-Marsi to find the state secrets privilege makes a great deal of sense. The trial court in El-Marsi found the privilege by looking to the “president’s constitutional authority over the conduct of this country’s diplomatic and military affairs.”40 In doing so it turned to the structural overtones of the Constitution, the “emanations” from the “penumbras” of the Commander-in-Chief clause.41 Those overtones lead to a logical conclusion: if the president is empowered to conduct the military and diplomatic affairs of the nation, he must also be empowered to protect his ability to conduct those affairs in the federal courts. The Fourth Circuit’s analysis looked to similar structural analysis that find intelligence responsibilities in the commander-in-chief, and thus a privilege to protect that aspect of the role.42 Neither of these opinions strayed from the routine examination employed by functional structuralism.
The analysis and conclusion of the both the district court and Fourth Circuit are in line with the structuralist tradition of interpretation. The Ninth Circuit may have had compelling reasons to think these overtones did not support a constitutional basis, but its opinion rested on a flawed interpretation of dicta. The privilege, to embody the Fourth Circuit’s holding, does indeed have a foundation in the Constitution itself.
The historical context of the common law points to an executive state secrets privilege.
During the first half of the 1600s, the early English common law developed the state secrets privilege out of the broader crown privilege.43 When the United States became independent, it adopted the English common law as the basis for its own non-statutory law.44 This would at first seem to bolster the argument that the state secrets privilege is a common law rule.
However, the country was adopting a common law that did not totally stem from an independent judiciary. The courts that made the decision that the crown privilege protected state secrets were not independent Article III courts as we have today, but part of an English judiciary that was premised on the Monarch’s powers—the judiciary in England at the time was not yet independent.45 Considering that the common law courts that recognized the privilege were not independent from the crown, but rather part and parcel, the argument that the privilege stems from the common law alone begins to muddy. The origin of the common law rule that national security secrets should be excluded from the courts is actually rooted in a recognition of an executive power and its interests.
The Supreme Court has at other times recognized that law is limited by the sovereignty of the United States in similar ways as the state secrets privilege by relying on older English and political concepts. One prime example is the federal sovereign immunity, never mentioned in the Constitution and, like the state secrets privilege, derived from the English monarch’s ownership of the courts.46 Despite its roots in monarchical power, sovereign immunity has been upheld by the Supreme Court as “[A]n axiom of our jurisprudence. The government is not liable to suit unless it consents thereto.”47
The Supreme Court often relies on pre-revolutionary English legal history in determining constitutional questions.48 Examining the logic and structure of English government as outlined above, it seems that the privilege as the United States adopted it was an artifact of executive government, not a mere common law rule of evidence. The acceptance of sovereign immunity in a democratic republic, even in light of its undemocratic origins, suggests that monarchical executive artifacts are accepted by the Supreme Court as sound grounding for American law.
Separation of powers concerns in the current political environment suggest that strengthening the privilege is a prudent policy consideration.
In an age that has increasingly been marked by constitutional hardball and vicious partisan approaches to handling the intelligence community, protecting the separation of powers with regard to national security may be more important than ever. Political scenarios that neither the founders nor the court in Reynolds could have imagined are becoming more commonplace. The common law interpretation of the privilege presents a dangerous double-edged sword of legislative flexibility in which an extreme political situation undercuts the separation of powers. As academics49 and the Fazaga court have directly recognized,50 a common law state secrets privilege would be vulnerable to congressional modification. In an extreme scenario, a hostile, partisan Congress that has decided to make political hay of the executive branch’s national security arms could aggressively claw back the privilege if it so desired. On the other hand, an intelligence community that lost the trust of the American public might need to be reined in through a modification of the privilege, even with the judicial review acting as a check on privilege assertions.
Congressional modification of the privilege may not be hostile—it could be done with the best of intentions. Congress could attempt to reform the state secrets privilege in order to expand access to justice or ensure it continues to serve the ideals of justice. For example, both Democratic and Republican members51 originally cosponsored the State Secrets Protection Act, which would have attempted to keep litigation alive despite the privilege using special procedures that balanced justice and national security.52 Congress could also pass supplemental rules for the privilege, such as the “silent witness rule,” which might allow for more flexibility in national security litigation by usage of code words by trial participants to refer to classified information.53 Such rule modifications would better keep the federal courts in their role of deciding cases and controversies and ensure litigation is kept alive beyond the discovery stage.
A constitutional, executive-based theory would shield the privilege from Congress; it would further entrust any modification of it to the courts, meaning that reform would be possible but only through the judicial review mechanism with the merits and pitfalls that accompany it. More simply, reformation of the Reynolds test and procedure would solely be possible through appellate litigation. Congress would be cut out completely but would be unable to use reform to play partisan games with the executive branch. The privilege would, ironically, be safeguarded in the courts by the traditional common law deference to precedent.
The common law theory of the privilege gives Congress and the courts the classic flexibility and avenues for reform that has allowed the common law to flourish as a legal system alongside legislatures in the first place. The constitutional theory furthers separation of powers in a political era when the foundations of republican government are potentially eroding. Given the concerning breakdown of political norms in this country, the Court may seek to ensure that separation of powers is further enhanced at the cost of more accessible paths to reform.
Though the average American has likely never heard of the state secrets privilege, it presents plaintiffs with a severe and at times uncompromising reality. In ruling that El-Marsi’s complaint must be dismissed due to the government’s invocation of the privilege, Judge Ellis acknowledged that “El–Masri has suffered injuries as a result of our country’s mistake and deserves a remedy. Yet, it is also clear from the result reached here that the only sources of that remedy must be the Executive Branch or the Legislative Branch”54 Just this year, an American journalist working in Syria that sought court intervention after alleging the government had mistakenly placed him on a “kill list” of terrorists and bombed him five times found he could not maintain his suit after the government invoked the privilege.55 Judge Collyer wrote that despite “the serious nature of the [p]lanitiff’s allegations,” she was powerless to allow the suit to proceed given that the privilege is absolute when properly asserted.56 In light of the outcomes the privilege sometimes produces, Professor Sudha Setty, another notable national security law professor, has suggested the privilege be reformed to add another element to its test, consideration of human rights interests.57
This article has argued, by examination of the structure of the Constitution, historical context, and policy considerations, that the most appropriate basis for the state secrets privilege is in constitutional law. The most important consequence to that basis is that Congress cannot lawfully modify the privilege, as the Fazaga court believed it could. This grounding in the Constitution makes reform difficult, but not impossible. Whether the privilege’s constitutional nature is truly desirable or not is another matter, as is whether the privilege in its modern form does more harm than good. Only the Supreme Court will be able to settle that question. The status quo, a world in which the American people do not know why exactly it is that the government may suppress evidence in some of the most morally difficult cases heard by the judiciary by an ex-parte, in-camera process, is certainly undesirable. The Supreme Court, whatever it may rule in the future, should take up this split as soon as the opportunity next presents itself.
1. See 98 C.J.S. Witnesses § 424 (2019).
2. See, e.g., Firth Sterling Steel Co. v. Bethlehem Steel Co., 199 F. 353, 356 (E.D. Pa., 1912) (allowing the government to prevent artillery blueprints from being introduced into commercial litigation).
3. See, e.g., Mohamed v. Jeppsen Dataplan, Inc., 614 F.3d 1070, 1087 (9th Cir. 2010) (en banc) (allowing for a complete litigation bar to a case involving a Central Intelligence Agency program).
4. See United States v. Reynolds, 345 U.S. 1, 7–8 (1953) (setting out the privilege in modern terms).
5. See, e.g., Jonathan M. Lamb, Comment, The Muted Rise of Rise of the Silent Witness Rule in National Security Litigation: The Eastern District of Virginia’s Answer to the Fight Over Classified Information at Trial, 36 Pepp. L. Rev. 213, 221–22 (2008) (“The Court revisited the issue during an evidentiary matter in the United States v. Burr treason trial of 1807”).
6. Robert M. Chesney, State Secrets and the Limits of National Security Litigation, 75 Geo. Wash. L. Rev. 1249, 1270–71 (2007).
7. United States v. Reynolds, 345 U.S. 1 (1949).
8. Id. at 10.
9. Id. at 7–8.
10. See, e.g., Benjamin Bernstein, Comment, Over Before it Even Began: Mohamed v. Jeppesen Dataplan and the Use of the State Secrets Privilege in Extraordinary Rendition Cases, 34 Fordham Int’l L.J. 1400, 1410–12 (2011) (documenting governmental use of the privilege to dismiss cases over the rendition program).
11. 18 U.S.C. Appendix §§ 1–16 (2018).
12. See Lamb, supra note 5, at 236.
13. El-Marsi v. Tenet, 437 F. Supp. 2d 530, 532–34 (E.D. Va. 2006).
14. Id. at 541.
15. Id. at 535, 538.
16. El-Marsi v. United States, 479 F.3d 296, 30304 (4th Cir. 2007).
17. Id. (citing to Dep’t of the Navy v. Egan, 484 U.S. 518, 527 (1988); United States v. Marchetti, 466 F.2d 1309, 1315 (4th Cir.1972)).
18. El-Marsi v. United States, 552 U.S. 947 (2007).
19. Fazaga v. Fed. Bureau of Investigation, 916 F.3d 1202, 1231 (9th Cir. 2019).
21. Id. at 1210.
22. Id. at 1211.
23. Id. at 1210–11 (pointing out the separate procedures in 18 U.S.C. § 1806(f)).
24. Id. at 1230.
25. Id. at 1231 (emphasis original).
26. Id. at 1230 (citing to Kasza v. Browner, 133 F.3d 1159, 1167 (9th Cir. 1998)).
28. Id. at 1231.
29. Id. at 1230.
30. Chesney, supra note 6.
31. Id. at 130910.
32. See Fazaga v. Fed. Bureau of Investigation, 916 F.3d 1202, 1231 (9th Cir. 2019) (“at bottom, [the privilege] is an evidentiary rule rooted in common law, not constitutional law.”) (emphasis original).
33. El-Marsi v. United States, 479 F.3d 296, 304.
34. Fazaga v. Fed. Bureau of Investigation, 916 F.3d 1202, 1231 (9th Cir. 2019).
35. Id. at 1231 (citing to General Dynamics Corp. v. United States, 536 U.S. 478, 485 (2011)).
36. United States v. Nixon, 418 U.S. 683, 709–10 (1974) (“Thus, the Fifth Amendment to the Constitution provides that no man ‘shall be compelled in any criminal case to be a witness against himself.’ And, generally, an attorney or a priest may not be required to disclose what has been revealed in professional confidence. These and other interests are recognized in law by privileges against forced disclosure, established in the Constitution, by statute, or at common law.”).
37. Griswold v. Connecticut, 381 U.S. 479, 484 (1965) (“[S]pecific guarantees in the Bill of Rights have penumbras, formed by emanations from those guarantees that help give them life and substance.”).
38. Printz v. United States, 521 U.S. 898, 923 n.13 (1997) (“Our system of dual sovereignty is reflected in numerous constitutional provisions… and not only those, like the Tenth Amendment, that speak to the point explicitly. It is not at all unusual for our resolution of a significant constitutional question to rest upon reasonable implications.”); Crandall v. State of Nevada, 73 U.S. 35, 43, 49 (1867) (Finding a right to travel between states based on the fact that the country is one whole nation and was intended to be such in the constitution).
39. Zivotofksy v. Kerry, 135 S.Ct. 2076, 2085 (2015) (“President has the sole power to negotiate treaties . . . . The President, too, nominates the Nation’s ambassadors and dispatches other diplomatic agents . . . . [T]he President himself has the power to open diplomatic channels simply by engaging in direct diplomacy with foreign heads of state and their ministers . . . . [These] specific Clauses confer the recognition power on the President . . .[t]he text and structure of the Constitution grant the President the power to recognize foreign nations and governments . . . . Recognition is a topic on which the Nation must [speak with one voice].”) (internal citations and quotation marks omitted.).
40. El-Marsi v. Tenet, 437 F. Supp. 2d 530, 535 (E.D. Va. 2006).
41. See Griswold v. Connecticut, 381 U.S. 479, 484 (1965).
42. El-Marsi v. United States, 479 F.3d 296, 303–04 (4th Cir. 2007).
43. Sudha Setty, Litigating Secrets: Comparative Perspectives on the State Secrets Privilege, 75 Brook. L. Rev. 201, 227–28 (2009) (pointing to Blackstone and sources discussing the time of Charles I).
44. See 15 Am. Jur. 2d § 4 (“The common law of England is the basic component of the common law as adopted by American courts.”).
45. Judith Resnik & Lane Dilg, Responding to a Democratic Deficit: Limiting the Powers and the Term of the Chief Justice of the United States 154 U. Pa. L. Rev. 1575, 1638 (2006) (“Only in 1701 had England rejected the practice that judges’ commissions ended with that of a given individual’s kingship; the Act of Settlement of 1701 created judges’ independence from the Crown by providing them with terms of office that could be concluded only through a formal request by the two Houses of Parliament.”).
46. Victor E. Schwartz et al., Prosser, Wade, and Schwartz’s Torts, Cases and Materials 681 (13th ed. 2015).
47. Prince v. United States, 174 U.S. 373, 375–76 (1899).
48. See, e.g., Dist. of Columbia. v. Heller, 554 U.S. 570, 582, 592–93 (2008) (utilizing early English legal history to interpret constitutional questions about the Second Amendment).
49. Chesney, supra note 6.
50. Fazaga v. Fed. Bureau of Investigation, 916 F.3d 1202, 1230 (9th Cir. 2019).
51. S.2533, 110th Cong. (2008) (“Mr. Kennedy (for himself, Mr. Specter, and Mr. Leahy) introduced the following bill”).
53. Lamb, supra note 5, at 217–218, 245.
54. El-Marsi v. Tenet, 437 F. Supp. 2d 530, 541 (E.D. Va. 2006).
55. Kareem v. Haspel et al, No. 17-581, 2019 WL 4645155 at *1 (D.D.C. Sept. 24, 2019)
56. Id. at *1, *5.
57. Setty, supra note 43, at 204 (arguing that Congress should reform the privilege).
The full text of this Article is available to download as a PDF.