College Athletics, Coercion, and the Establishment Clause: The Case of Clemson Football

Once a person turns eighteen and goes to college, do they immediately become less susceptible to the influences of those in power and their peers? The Supreme Court tells us that they do. While consistently willing to find that prayers at middle school graduations and high school football games are violations of the Establishment Clause under the coercion test, the Court has stated that adults are more mature and “presumably” less susceptible to religious coercion. Scholars and the circuit courts of appeals have taken varying approaches and arrived at different outcomes when considering adult claimants. None, however, have articulated a uniform test for adults to establish coercion. Using indicative language from the Supreme Court, this Note argues for the first time that adult claimants must show that a State action has a “real and substantial likelihood” of coercion in order to bring a successful Establishment Clause challenge. It further proposes that a spectrum of susceptibility to coercion exists under the Establishment Clause based on certain populations’ ages and respective environments.

After articulating the standard of coercion for adults and the spectrum of susceptibility to coercion, this Note applies both to a prominent example of overt incorporation of religion into a public university—the Clemson University football program. The Clemson football coaching staff unabashedly integrates religion into many aspects of the program, from Bible studies led and organized by staff to baptisms of players on the practice field. Using psychological and educational research about the effects of coaches and teammates on a college student-athlete’s values, beliefs, and behaviors, this Note argues that college student-athletes are uniquely prone to coercion and places them on the spectrum of susceptibility to coercion. Finally, it applies the standard of coercion for adults to conclude that religious aspects of Clemson’s football program are unconstitutional under the Establishment Clause.

Introduction

Situated in the small college town of Clemson, South Carolina, the Clemson University (“Clemson”) football team boasts quite a record. With NCAA College Football Playoff (“CFP”) National Championships in 2016 and 2018, consecutive CFP appearances and Atlantic Coast Conference Championships from 2015 to 2019, and at least ten wins in each season from 2011 to 2019, head coach Dabo Swinney has built a culture of success in his program around his slogan “all in.”1.National Champions, https://daboswinney.com/championships/ [https://perma.cc/9S7H-SPP4] (last visited Sept. 28, 2020); Greg Wallace, Inside a Top 10 College Football Team’s Summer Conditioning Program, Bleacher Rep. (June 6, 2014), https://bleacher­report.com/articles/2088395-inside-a-top-10-college-football-teams-summer-conditioning-program [https://perma.cc/274C-3ZG5] (internal quotation marks omitted).Show More While the students, alumni, and fans of Clemson football may consider football their religion, there is a tenet of actual sectarian religion deeply ingrained and woven into the program’s culture.2.Tim Rohan, Faith, Football and the Fervent Religious Culture at Dabo Swinney’s Clemson, Sports Illustrated (Sept. 4, 2019), https://www.si.com/college-football/2019/09/04/clemson-dabo-swinney-religion-culture [https://perma.cc/46JS-8ZGA].Show More As documented by the Freedom from Religion Foundation’s (“FFRF”) 2014 letter sent to Clemson’s Senior Associate General Counsel, several of the program’s practices—which originate from the coaching staff’s conduct and are not student-led or organized—indicate not only an “endorsement of religion over nonreligion,” but also a preference for “Christian worship.”3.Letter from Patrick C. Elliott, Staff Att’y, Freedom from Religion Found., to Erin Swan Lauderdale, Senior Assoc. Gen. Couns., Clemson Univ. 1 (Apr. 10, 2014) [hereinafter FFRF Letter], https://ffrf.org/images/clemson_letter.pdf [https://perma.cc/Y26K-4HUA]. FFRF sent this letter to Clemson after they reviewed records obtained through a FOIA request. Id.Show More Coach Swinney has maintained an “an outwardly religious program.”4.Kevin Trahan, Freedom from Religion Foundation Complains About Clemson Football Program, SBNation (Apr. 15, 2014), https://www.sbnation.com/college-football/2014/4/15/5616602/clemson-football-dabo-swinney-religious-freedom-complaint.Show More Quite simply, “[a]t Clemson, God is everywhere.”5.Brad Wolverton, With God on Our Side, Chron. of Higher Educ. (Nov. 24, 2013), https://www.chronicle.com/article/With-God-on-Our-Side/143231 [https://perma.cc/7Q3G-M4PG].Show More

In 2011, James Trapp became the official chaplain of the football team at Coach Swinney’s personal invitation and insistence.6.FFRF Letter, supra note 3, at 1.Show More In his paid role as chaplain, Mr. Trapp went beyond simply leading team prayers. He “was regularly given access to the entire football team in between drills for the purpose of bible study,” maintained an office in the Jervey Athletic Center where he kept Bibles for distribution and displayed Bible quotes, and planned and facilitated sessions on “being baptized” in the athletic center.7.Id. at 2–3.Show More Mr. Trapp also organized more than eighty devotionals for the football team between March 2012 and April 2013, which were approved by Coach Swinney and led by members of the coaching staff.8.Id. at 4.Show More Further, he organized the team’s transportation via coach buses to local churches for annual “Church Day[s]” during training camp.9.Id.Show More

Journalists have reported other instances of the coaching staff’s endorsement of religion. In the fall of 2012, star wide receiver DeAndre Hopkins was baptized on the field in his uniform and pads at the conclusion of practice.10 10.Rohan, supra note 2; see also Wolverton, supra note 5 (describing DeAndre Hopkins’s baptism in a livestock trough on the practice field).Show More Then-Assistant Coach Jeff Scott even tweeted a photo that captured the scene.11 11.Coach Jeff Scott (@coach_jeffscott), Twitter (Sept. 2, 2012, 10:14 PM), https://twitter.com/coach_jeffscott/status/24244533830313984.Show More Following Hopkins’s baptism, it is estimated that between ten and fifteen player baptisms occurred over the next two seasons—many of which took place during camp in a pond by the practice field.12 12.Rohan, supra note 2.Show More Coach Swinney tells recruits and their families that he is a Christian and, if they “have a problem with that, [they] don’t have to be [there].”13 13.Wolverton, supra note 5.Show More One recruit’s mother distinctly remembers Coach Swinney’s guarantee “that every single player that comes through this program will hear about the Gospel of Christ.”14 14.Rohan, supra note 2.Show More

If Clemson were a public high school instead of a public university, this situation would present a clear violation of the Establishment Clause of the First Amendment.15 15.See Lee v. Weisman, 505 U.S. 577, 592 (1992) (“[T]here are heightened concerns with protecting freedom of conscience from subtle coercive pressure in the elementary and secondary public schools.”).Show More The Supreme Court has stated, however, that college students “are . . . young adults” and are therefore “less impressionable than younger students.”16 16.Widmar v. Vincent, 454 U.S. 263, 274 n.14 (1981). Widmar was decided on free speech grounds as the Court did not find the State’s interest in creating more separation between church and state than required by the Establishment Clause “sufficiently ‘compelling’ to justify content-based discrimination against respondents’ religious speech.” Id. at 276. For further discussion of circuit courts of appeals’ application of Widmar to the coercion test in cases involving higher education, see also infra notes 110–12 and accompanying text.Show More Yet, the Court has not spoken directly on the issue of religious coercion with respect to adult college students at a public university. While the religious nature of Clemson’s football program presents only one example of overt incorporation of religion at a public university, the initial, more important, and unanswered question is what must adult claimants show in order to litigate a successful Establishment Clause challenge.17 17.By “adult,” I mean individuals who have reached the age of majority in their respective states.Show More

Using the prominent example of Clemson football,18 18.Clemson’s football program presents one of many examples of the incorporation of religion into college football. For examples of other football programs that have hired chaplains, see Freedom from Religion Found., Pray to Play: Christian Coaches and Chaplains Are Converting Football Fields into Mission Fields 13–15 (2015) [hereinafter Pray to Play], https://ffrf.org/images/Pray_To_Play_FINAL_REPORT1.pdf [https://perma.cc/A4QB-T8V­H]. Examples of the incorporation of religion can also be found in other college sports, such as basketball. See, e.g., Whitelaw Reid, Man of Faith: How Tony Bennett’s Religion Has Shaped His UVa Tenure, Daily Progress (Nov. 24, 2010), https://www.dailyprogress.com/­sports/man-of-faith-how-tony-bennett-s-religion-has-shaped/article_de7b70b5-54f2-5f94-bc4f-dc4449748bb8.html [https://perma.cc/CM7C-K9JR] (“As a number of recruits have signed to play for [Tony] Bennett, the first thing they’ve talked about . . . . [is] the connection they’ve felt with Bennett through God.”).Show More this Note answers that question by articulating a coercion standard for adults, arguing that a spectrum of susceptibility to coercion exists under the Establishment Clause, and suggesting where college student-athletes fit along that spectrum. While some scholars have written about the Establishment Clause and college athletics, they do not apply the modern coercion test,19 19.The modern coercion test focuses on psychological coercion as articulated in Lee v. Weisman. See infra Section I.B.Show More articulate a coercion standard for adults, or advance a theory regarding a range of susceptibility to coercion. For example, Clayton Adams, emphasizing a need to protect a “government employee’s right to speak on matters of public concern,” applied a “modified coercion test” to religious aspects of various college football programs—including the Clemson football program.20 20.Clayton D. Adams, Personal Foul, Roughing the Speaker: The Illusory War Between the Establishment Clause & College Football, 84 Miss. L.J. Supra 167, 183, 194–202 (2015).Show More Kris Bryant suggested that the Court should adopt a “Coercion/Endorsement Test ‘with teeth’” when analyzing the Establishment Clause claims of public university students.21 21.Kris Bryant, Take a Knee: Applying the First Amendment to Locker Room Prayers and Religion in College Sports, 36 J. Coll. & U.L. 329, 359–60 (2009).Show More Gil Fried and Lisa Bradley briefly suggested that there is an “Establishment Clause case law scale from elementary school prayer to prayer opening legislative sessions cases” and recognized that “college prayer cases” fall “in between these two ends of the continuum” without theorizing further.22 22.Gil Fried & Lisa Bradley, Applying the First Amendment to Prayer in a Public University Locker Room: An Athlete’s and Coach’s Perspective, 4 Marq. Sports L.J. 301, 303, 310–13 (1994).Show More Fried and Bradley, however, then applied the now disfavored Lemon test to analyze college locker room prayers.23 23.Id.Show More In a similar vein, other scholars have addressed the Establishment Clause’s application to students at public universities, or adults in general, without applying the psychological coercion test, articulating a coercion standard for adults, or proposing a theory regarding a range of susceptibility to coercion.24 24.See, e.g., Phillip E. Marbury, Comment, Audience Maturity and the Object of the Establishment Clause, 6 Liberty U. L. Rev.565, 579 (2012) (arguing that “audience maturity is a significant factor” in the Court’s modern Establishment Clause jurisprudence); Elizabeth B. Halligan, Coercing Adults? The Fourth Circuit and the Acceptability of Religious Expression in Government Settings, 57 S.C. L. Rev.923, 924–26 (2006) (analyzing Mellen v. Bunting, a Fourth Circuit case that struck down a prayer at a public military university because of public prayer’s potential impact on adult audience members); Deanna N. Pihos, Assuming Maturity Matters: The Limited Reach of the Establishment Clause at Public Universities, 90 Cornell L. Rev.1349, 1373 (2005) (arguing that the respective ages of high school and college students is a “questionable distinction on which to create two different standards of Establishment Clause protection” under the coercion, Lemon, and endorsement tests).Show More

This Note addresses the gap in the literature regarding how to treat adult claimants under the coercion test of the Establishment Clause. Instead of suggesting a new or modified coercion test or using a now disfavored test, this Note articulates a practical coercion standard for adults that is rooted in the current jurisprudence. Part I of this Note traces the development of the modern coercion test in the Supreme Court and the test’s application to cases involving higher education in the circuit courts of appeals. Then, Part II proposes a coercion standard for adults and, based on their respective environments, places various populations along a spectrum according to their level of susceptibility to coercion. Finally, Part III applies the coercion standard for adults and coercion spectrum to college student-athletes. It argues that college student-athletes should be seen as more susceptible to coercion than typical college students and that various religious-oriented aspects of the Clemson football program violate the Establishment Clause. A conclusion follows.

  1. * J.D., University of Virginia School of Law, 2021; M.Ed., Clemson University, 2015. I am especially grateful to Professor Micah Schwartzman for sparking this research and for supporting me in the development of this Note. I also sincerely appreciate the members of the Virginia Law Review who assisted in the editing and preparation of the Note—especially Olivia Roat.
  2. National Champions, https://daboswinney.com/championships/ [https://perma.cc/9S7H-SPP4] (last visited Sept. 28, 2020); Greg Wallace, Inside a Top 10 College Football Team’s Summer Conditioning Program, Bleacher Rep. (June 6, 2014), https://bleacher­report.com/articles/2088395-inside-a-top-10-college-football-teams-summer-conditioning-program [https://perma.cc/274C-3ZG5] (internal quotation marks omitted).
  3.  Tim Rohan, Faith, Football and the Fervent Religious Culture at Dabo Swinney’s Clemson, Sports Illustrated (Sept. 4, 2019), https://www.si.com/college-football/2019/09/04/
    clemson-dabo-swinney-religion-culture [https://perma.cc/46JS-8ZGA].
  4. Letter from Patrick C. Elliott, Staff Att’y, Freedom from Religion Found., to Erin Swan Lauderdale, Senior Assoc. Gen. Couns., Clemson Univ. 1 (Apr. 10, 2014) [hereinafter FFRF Letter], https://ffrf.org/images/clemson_letter.pdf [https://perma.cc/Y26K-4HUA]. FFRF sent this letter to Clemson after they reviewed records obtained through a FOIA request. Id.
  5. Kevin Trahan, Freedom from Religion Foundation Complains About Clemson Football Program, SBNation (Apr. 15, 2014), https://www.sbnation.com/college-football/2014/4/15/
    5616602/clemson-football-dabo-swinney-religious-freedom-complaint.
  6. Brad Wolverton, With God on Our Side, Chron. of Higher Educ. (Nov. 24, 2013), https://www.chronicle.com/article/With-God-on-Our-Side/143231 [https://perma.cc/7Q3G-M4PG].
  7. FFRF Letter, supra note 3, at 1.
  8. Id. at 2–3.
  9. Id. at 4.
  10. Id.
  11. Rohan, supra note 2; see also Wolverton, supra note 5 (describing DeAndre Hopkins’s baptism in a livestock trough on the practice field).
  12.  Coach Jeff Scott (@coach_jeffscott), Twitter (Sept. 2, 2012, 10:14 PM), https://twitter.com/coach_jeffscott/status/24244533830313984.
  13. Rohan, supra note 2.
  14. Wolverton, supra note 5.
  15. Rohan, supra note 2.
  16. See Lee v. Weisman, 505 U.S. 577, 592 (1992) (“[T]here are heightened concerns with protecting freedom of conscience from subtle coercive pressure in the elementary and secondary public schools.”).
  17. Widmar v. Vincent, 454 U.S. 263, 274 n.14 (1981). Widmar was decided on free speech grounds as the Court did not find the State’s interest in creating more separation between church and state than required by the Establishment Clause “sufficiently ‘compelling’ to justify content-based discrimination against respondents’ religious speech.” Id. at 276. For further discussion of circuit courts of appeals’ application of Widmar to the coercion test in cases involving higher education, see also infra notes 110–12 and accompanying text.
  18. By “adult,” I mean individuals who have reached the age of majority in their respective states.
  19. Clemson’s football program presents one of many examples of the incorporation of religion into college football. For examples of other football programs that have hired chaplains, see Freedom from Religion Found., Pray to Play: Christian Coaches and Chaplains Are Converting Football Fields into Mission Fields 13–15 (2015) [hereinafter Pray to Play], https://ffrf.org/images/Pray_To_Play_FINAL_REPORT1.pdf [https://perma.cc/A4QB-T8V­H]. Examples of the incorporation of religion can also be found in other college sports, such as basketball. See, e.g., Whitelaw Reid, Man of Faith: How Tony Bennett’s Religion Has Shaped His UVa Tenure, Daily Progress (Nov. 24, 2010), https://www.dailyprogress.com/­sports/man-of-faith-how-tony-bennett-s-religion-has-shaped/article_de7b70b5-54f2-5f94-bc4f-dc4449748bb8.html [https://perma.cc/CM7C-K9JR] (“As a number of recruits have signed to play for [Tony] Bennett, the first thing they’ve talked about . . . . [is] the connection they’ve felt with Bennett through God.”).
  20. The modern coercion test focuses on psychological coercion as articulated in Lee v. Weisman. See infra Section I.B.
  21. Clayton D. Adams, Personal Foul, Roughing the Speaker: The Illusory War Between the Establishment Clause & College Football,
    84

    Miss. L.J. Supra

    167

    , 183, 194–202 (2015).

  22. Kris Bryant, Take a Knee: Applying the First Amendment to Locker Room Prayers and Religion in College Sports, 36 J. Coll. & U.L
    .

    329, 359–60 (2009).

  23. Gil Fried & Lisa Bradley, Applying the First Amendment to Prayer in a Public University Locker Room: An Athlete’s and Coach’s Perspective, 4 Marq. Sports L.J. 301, 303, 310–13 (1994).
  24. Id.
  25.  See, e.g., Phillip E. Marbury, Comment, Audience Maturity and the Object of the Establishment Clause, 6 Liberty U. L. Rev.

    565, 579 (2012) (arguing that “audience maturity is a significant factor” in the Court’s modern Establishment Clause jurisprudence); Elizabeth B. Halligan, Coercing Adults? The Fourth Circuit and the Acceptability of Religious Expression in Government Settings, 57 S.C. L. Rev.

    923, 924–26 (2006) (analyzing Mellen v. Bunting, a Fourth Circuit case that struck down a prayer at a public military university because of public prayer’s potential impact on adult audience members); Deanna N. Pihos, Assuming Maturity Matters: The Limited Reach of the Establishment Clause at Public Universities, 90 Cornell L. Rev.

    1349, 1373 (2005) (arguing that the respective ages of high school and college students is a “questionable distinction on which to create two different standards of Establishment Clause protection” under the coercion, Lemon, and endorsement tests).

Against Fiduciary Constitutionalism

A growing body of scholarship draws connections between fiduciary law and the Constitution. In much of this literature, the Constitution is described as a fiduciary instrument that establishes fiduciary duties, not least for the President of the United States.

This Article examines and critiques the claims of fiduciary constitutionalism. Although a range of arguments are made in this literature, there are common failings. Some of these involve a literalistic misreading of the works of leading political philosophers (e.g., Plato and Locke). Other failings involve fiduciary law, such as mistakes about how to identify fiduciary relationships and about the content and enforcement of fiduciary duties. Still other failings sound in constitutional law, including the attempt to locate the genre of the Constitution in the categories of private fiduciary law. These criticisms suggest weaknesses in the new and increasingly influential attempt to develop fiduciary constitutionalism.

Introduction

Recently, a number of scholars have argued that the U.S. Constitution resembles a fiduciary document, and that it imposes fiduciary duties on various actors, including the President of the United States.1.See, e.g., Gary Lawson & Guy Seidman, “A Great Power of Attorney”: Understanding the Fiduciary Constitution (2017); Randy E. Barnett & Evan D. Bernick, The Letter and the Spirit: A Unified Theory of Originalism, 107 Geo. L.J. 1, 18 (2018); Andrew Kent, Ethan J. Leib & Jed Handelsman Shugerman, Faithful Execution and Article II, 132 Harv. L. Rev. 2111 (2019); Ethan J. Leib & Stephen R. Galoob, Fiduciary Principles and Public Offices, in The Oxford Handbook of Fiduciary Law 303 (Evan J. Criddle, Paul B. Miller & Robert H. Sitkoff eds., 2019); Ethan J. Leib & Andrew Kent, Fiduciary Law and the Law of Public Office: Suggestions for a Research Agenda, Wm. & Mary L. Rev. (forthcoming 2021) (on file with authors); Ethan J. Leib & Jed Handelsman Shugerman, Fiduciary Constitutionalism: Implications for Self-Pardons and Non-Delegation, 17 Geo. J.L. & Pub. Pol’y 463 (2019); Ethan J. Leib, David L. Ponet & Michael Serota, A Fiduciary Theory of Judging, 101 Calif. L. Rev. 699 (2013); Robert G. Natelson, The Constitution and the Public Trust, 52 Buff. L. Rev. 1077, 1140 (2004) [hereinafter Natelson, The Constitution and the Public Trust].Show More When Barack Obama was President, some found in the “fiduciary Constitution”2.Different scholars writing about what we call the “fiduciary Constitution” might have their own nomenclature. Our choice of terms, however, is not original and is not meant to be pejorative. See Lawson & Seidman, supra note 1 (title); Kent et al., supra note 1, at 2182; Leib & Shugerman, supra note 1 (title); id. at 464.Show More a means by which a court could hold unconstitutional the signature achievement of his administration, the Affordable Care Act.3.E.g., Lawson & Seidman, supra note 1, at 91–98.Show More Now others find in fiduciary constitutionalism a means by which a court could find the present incumbent to have violated a proscription on self-dealing.4.See Leib & Shugerman, supra note 1, at 468–69, 475.Show More The fiduciary Constitution contains multitudes—everything from a handy way for an originalist to justify Bolling v. Sharpe5.See Lawson & Seidman, supra note 1, at 151–53, 170–71.Show More to a way for skeptics of the non-delegation doctrine to rein it in.6.See Leib & Shugerman, supra note 1, at 479–85.Show More The convenience and malleability of this new constitutional argument should make us wary.

This Article offers a critique of fiduciary constitutionalism, finding it bad fiduciary law and bad constitutional law. We are not the first to criticize fiduciary constitutionalism, and our work therefore builds on that of others, especially Seth Davis and Richard Primus.7.See generally Seth Davis, The False Promise of Fiduciary Government, 89 Notre Dame L. Rev. 1145 (2014) (analyzing differences between private fiduciaries and public officials and critiquing the lack of authority for fiduciary constitutionalism); Richard Primus, The Elephant Problem, 17 Geo. J.L. & Pub. Pol’y 373 (2019) (critically reviewing the evidence and methodology in Lawson & Seidman, supra note 1). Others have also criticized the over-extension of fiduciary concepts, albeit in different contexts. E.g., Lina M. Khan & David E. Pozen, A Skeptical View of Information Fiduciaries, 133 Harv. L. Rev. 497 (2019).Show More

It is important at the outset to note that the burgeoning literature in favor of a fiduciary reading of the Constitution means that different scholars make different claims. Some analogize the Constitution to a trust,8.See Leib & Shugerman, supra note 1, at 477–79.Show More some to an agency relationship,9.Lawson & Seidman, supra note 1, at 112.Show More some to a power of attorney,10 10.Id. at 75 (referring to James Iredell’s quote regarding power of attorney).Show More and some to an attractively all-purpose generic “fiduciary” construct.11 11.Barnett & Bernick, supra note 1, at 18–21; see also id. at 19 nn.81 & 83 (discussing common fiduciary relationships and duties); cf. Lawson & Seidman, supra note 1, at 76 (“[W]e think it close to obvious that the Constitution, as a legal document, is best understood as some kind of agency or fiduciary instrument, whereas the case for viewing it specifically as (or as like) a power of attorney is more attenuated.”); Leib et al., supra note 1, at 708, 712–13 (recognizing differences among different kinds of fiduciaries, as well as differences between public and private law, but relying on a general “fiduciary principle”).Show More Some draw a straight line between a fiduciary law of the eighteenth century and the present.12 12.See, e.g., Barnett & Bernick, supra note 1, at 19–21.Show More Others admit there are differences between past and present, and thus offer a bevy of qualifications and hedges13 13.E.g., Lawson & Seidman, supra note 1, at 6 (“[W]e are making no claims about the extent to which the meaning we uncover should or must contribute to legal decision making.”); id. at 11 (“Again, we frame these interpretative conclusions in hypothetical form: to the extent that the Constitution can be seen as a fiduciary instrument, or in some cases as a fiduciary instrument of a particular kind, certain conclusions about the document’s meaning follow from that identification.”). The hedges and qualifications are more pronounced—and thus the conclusions more circumspect—in the article by Kent, Leib, and Shugerman than in some of the other literature. See, e.g., Kent et al., supra note 1, at 2190 (“We do not opine here on the way the framers envisioned enforcing the President’s duty of loyalty and avoiding self-dealing.”).Show More—but those tend to fall away as soon as the scholars turn to spelling out the “enforceable duties” that they want to see the courts administering.14 14.E.g., Lawson & Seidman, supra note 1, at 7 (“Now to our affirmative project: understanding the fiduciary character of the Constitution is important not simply as a historical matter but also for its contribution to constitutional interpretation.”); id. at 11; Leib & Shugerman, supra note 1, at 465 (“[T]his article delves further into this language’s [i.e. the Take Care Clause’s] likely meaning, indicating how it can establish enforceable duties for public officials.”); see also Kent et al., supra note 1, at 2119 (claiming “that the best historical understanding of the meaning of the Faithful Execution Clauses is that they impose duties that we today—and some in the eighteenth century as well—would call fiduciary”). Richard Primus and Suzanna Sherry note this oscillation in Lawson and Seidman’s argument. Primus, supra note 7, at 400, 402–04; Suzanna Sherry, The Imaginary Constitution, 17 Geo. J.L. & Pub. Pol’y 441, 447–49 (2019) (reviewing Lawson & Seidman, supra note 1). For discussion of a similar tendency in Kent et al., supra note 1, see infra note 87.Show More Some of the literature starts out cautious and then gallops away; some of it gallops from the start. But it all tends to be ambitious and thoroughly presentist.

This Article offers three main critiques of fiduciary constitutionalism:

The first is about contemporary fiduciary law. The fiduciary constitutionalists pervasively treat concepts as “fiduciary” that are in fact not limited to fiduciary law and instead have much a broader application, such as good faith.15 15.See infra Part II.Show More This matters because a point of connection between the Constitution and some area that can be characterized as fiduciary (e.g., trust, agency), as thin as it is, cannot sustain the fiduciary constitutionalist project if it is a point of connection with many areas of law.

The second is about historical anachronism. Some fiduciary constitutionalists rely on a “fiduciary law” of 1789 when there was in fact no such law.16 16.E.g., Barnett & Bernick, supra note 1, at 20–21 (asserting that the Constitution’s “organization and language sounds in eighteenth-century fiduciary law”). More cautious on this point is Kent et al., supra note 1, at 2179–80 (“[W]hat the three meanings we can attribute to the [Faithful Execution] Clauses have in common is that they are all part of the basic ways the private law constrains fiduciary discretion and power.”).Show More There were quite specific legal regimes for trust, agency, bailment, and so on. Some of these were at law, and some were in equity. Their standards of liability were different; their remedies were different; they differed in respect to defenses and the availability of a jury.17 17.Contra Lawson & Seidman, supra note 1, at 62 (“The fiduciary responsibilities of a trustee and an attorney do not differ in any way material to our project.”).Show More As other fiduciary constitutionalists concede, there was no pan-subject “fiduciary law” in 1789.18 18.See, e.g., Kent et al., supra note 1, 2180–81 (describing the “crystalliz[ing]” of “the ‘private’ fiduciary law we would recognize today” as not occurring in America until late in the eighteenth century, and concluding that “a fiduciary law of ‘private’ offices was unlikely to have been plucked off-the-rack by the Philadelphia Convention drafters and applied to public offices”); id. at 2179 (“Our historical findings about the original meaning of the Faithful Execution Clauses align with core features of modern fiduciary law . . . . ”); Leib & Shugerman, supra note 1, at 468 (suggesting doubt about whether “private fiduciary law was itself fully formed at the time of the founding”).Show More This abstraction is fatal to fiduciary constitutionalism.

The third is about how some fiduciary constitutionalists evade the anachronism problem. They try to escape this problem by tying the Constitution not to a fictional eighteenth-century “fiduciary law” but rather to fiduciary law of the twenty-first century.19 19.Leib & Shugerman, supra note 1, at 468–69 (“[Article II of the U.S. Constitution] uses the language of faith and care to signal to courts and to executive officials that the President was supposed to be held to the same kinds of fiduciary obligations to which corporate officers, trustees, and lawyers are routinely held today in the private sector.”).Show More That interpretive move is certainly available on some theories of constitutional interpretation—theories that would to a substantial degree integrate political morality and constitutional law, or theories that would allow the courts to legitimately turn what was originally a non-fiduciary Constitution into a fiduciary Constitution, creating and developing new fiduciary duties for government officers.20 20.But cf. Davis, supra note 7, at 1182–95 (discussing whether that is advisable).Show More But that is not the argument the fiduciary constitutionalists have made to date (though they are of course free to reshape their arguments going forward). Instead, the fiduciary constitutionalists have relied on text and structure as understood at the Founding.21 21.See infra notes 161 and 164 and accompanying text. We do recognize that there is a spectrum of views about how much the interpretation of the constitutional text should be informed by the legal categories at the time of ratification (as part of context). The more one separates the text from its legal context, the more room there is to integrate constitutional law and political morality, and the less constraint there is from the original understanding (with the familiar virtues and vices of that interpretive equilibrium).Show More And that more-or-less originalist argument does not hold if the Constitution can only be understood as fiduciary in light of present-day fiduciary law.

One thing, however, does need to be said in favor of the fiduciary constitutionalists’ claims. It is true that there is a long history in political thought of the use of trust and agency metaphors for governance. This figurative language appears in Plato and Cicero, in Locke and Hume.22 22.See infra Section I.A.Show More But this language offers moral guidance and political wisdom, not enforceable duties with remedies that can be awarded by courts. And mere metaphor is not the big game the fiduciary constitutionalists are pursuing. Against this long history of a figurative and legally thin understanding of public office as a trust, it becomes easier to recognize the fiduciary constitutionalist project for what it is: an earnest and literalistic misreading of the tradition and an insistence on taking figurative language that works across thousands of years of political theory and treating it as if it were an invocation of an inevitably more particular body of legal or equitable claims and remedies. There is no such body of claims and remedies that can support this move—not a fiduciary law of 1789 (once it is recognized as a fiction), and not present-day fiduciary law (once it is correctly described).

The remainder of this Article proceeds as follows. Part I analyzes the supposed antecedents of fiduciary constitutionalism—both the classical and early modern tradition in political theory and also more recent advances in fiduciary theory with which the fiduciary constitutionalists align themselves. Part II critiques the fiduciary law of fiduciary constitutionalism. Part III critiques the constitutional law of fiduciary constitutionalism. The idea that the Constitution ought to be understood as a fiduciary instrument is a well-meant and seemingly timely entrant into the constitutional discourse of the United States. But taken together, these critiques show that the historical, philosophical, and legal foundations of fiduciary constitutionalism are weak.

  1. * Professor of Law, Notre Dame Law School.
  2. ** Professor of Law, Associate Dean for International and Graduate Programs, and Director of the Notre Dame Program on Private Law, Notre Dame Law School. The authors are grateful for comments from Randy Barnett, Evan Bernick, Nathan Chapman, Evan Criddle, Evan Fox-Decent, Joshua Getzler, Andrew Gold, Albert Horsting, Andrew Kent, Sung Hui Kim, Gary Lawson, Ethan Leib, Ben McFarlane, Michael McConnell, David Pozen, Richard Primus, Richard Re, Lionel Smith, and Larry Solum, as well as attendees at the Third Annual International Fiduciary Law Conference held at Trinity College, Cambridge.
  3. See, e.g., Gary Lawson & Guy Seidman, “A Great Power of Attorney”: Understanding the Fiduciary Constitution (2017); Randy E. Barnett & Evan D. Bernick, The Letter and the Spirit: A Unified Theory of Originalism, 107 Geo. L.J. 1, 18 (2018); Andrew Kent, Ethan J. Leib & Jed Handelsman Shugerman, Faithful Execution and Article II, 132 Harv. L. Rev. 2111 (2019); Ethan J. Leib & Stephen R. Galoob, Fiduciary Principles and Public Offices, in The Oxford Handbook of Fiduciary Law 303 (Evan J. Criddle, Paul B. Miller & Robert H. Sitkoff eds., 2019); Ethan J. Leib & Andrew Kent, Fiduciary Law and the Law of Public Office: Suggestions for a Research Agenda, Wm. & Mary L. Rev. (forthcoming 2021) (on file with authors); Ethan J. Leib & Jed Handelsman Shugerman, Fiduciary Constitutionalism: Implications for Self-Pardons and Non-Delegation, 17 Geo. J.L. & Pub. Pol’y 463 (2019); Ethan J. Leib, David L. Ponet & Michael Serota, A Fiduciary Theory of Judging, 101 Calif. L. Rev. 699 (2013); Robert G. Natelson, The Constitution and the Public Trust, 52 Buff. L. Rev. 1077, 1140 (2004) [hereinafter Natelson, The Constitution and the Public Trust].
  4. Different scholars writing about what we call the “fiduciary Constitution” might have their own nomenclature. Our choice of terms, however, is not original and is not meant to be pejorative. See Lawson & Seidman, supra note 1 (title); Kent et al., supra note 1, at 2182; Leib & Shugerman, supra note 1 (title); id. at 464.
  5. E.g., Lawson & Seidman, supra note 1, at 91–98.
  6. See Leib & Shugerman, supra note 1, at 468–69, 475.
  7. See Lawson & Seidman, supra note 1, at 151–53, 170–71.
  8. See Leib & Shugerman, supra note 1, at 479–85.
  9. See generally Seth Davis, The False Promise of Fiduciary Government, 89 Notre Dame L. Rev. 1145 (2014) (analyzing differences between private fiduciaries and public officials and critiquing the lack of authority for fiduciary constitutionalism); Richard Primus, The Elephant Problem, 17 Geo. J.L. & Pub. Pol’y 373 (2019) (critically reviewing the evidence and methodology in Lawson & Seidman, supra note 1). Others have also criticized the over-extension of fiduciary concepts, albeit in different contexts. E.g., Lina M. Khan & David E. Pozen, A Skeptical View of Information Fiduciaries, 133 Harv. L. Rev. 497 (2019).
  10. See Leib & Shugerman, supra note 1, at 477–79.
  11. Lawson & Seidman, supra note 1, at 112.
  12. Id. at 75 (referring to James Iredell’s quote regarding power of attorney).
  13. Barnett & Bernick, supra note 1, at 18–21; see also id. at 19 nn.81 & 83 (discussing common fiduciary relationships and duties); cf. Lawson & Seidman, supra note 1, at 76 (“[W]e think it close to obvious that the Constitution, as a legal document, is best understood as some kind of agency or fiduciary instrument, whereas the case for viewing it specifically as (or as like) a power of attorney is more attenuated.”); Leib et al., supra note 1, at 708, 712–13 (recognizing differences among different kinds of fiduciaries, as well as differences between public and private law, but relying on a general “fiduciary principle”).
  14. See, e.g., Barnett & Bernick, supra note 1, at 19–21.
  15. E.g., Lawson & Seidman, supra note 1, at 6 (“[W]e are making no claims about the extent to which the meaning we uncover should or must contribute to legal decision making.”); id. at 11 (“Again, we frame these interpretative conclusions in hypothetical form: to the extent that the Constitution can be seen as a fiduciary instrument, or in some cases as a fiduciary instrument of a particular kind, certain conclusions about the document’s meaning follow from that identification.”). The hedges and qualifications are more pronounced—and thus the conclusions more circumspect—in the article by Kent, Leib, and Shugerman than in some of the other literature. See, e.g., Kent et al., supra note 1, at 2190 (“We do not opine here on the way the framers envisioned enforcing the President’s duty of loyalty and avoiding self-dealing.”).
  16. E.g., Lawson & Seidman, supra note 1, at 7 (“Now to our affirmative project: understanding the fiduciary character of the Constitution is important not simply as a historical matter but also for its contribution to constitutional interpretation.”); id. at 11; Leib & Shugerman, supra note 1, at 465 (“[T]his article delves further into this language’s [i.e. the Take Care Clause’s] likely meaning, indicating how it can establish enforceable duties for public officials.”); see also Kent et al., supra note 1, at 2119 (claiming “that the best historical understanding of the meaning of the Faithful Execution Clauses is that they impose duties that we today—and some in the eighteenth century as well—would call fiduciary”). Richard Primus and Suzanna Sherry note this oscillation in Lawson and Seidman’s argument. Primus, supra note 7, at 400, 402–04; Suzanna Sherry, The Imaginary Constitution, 17 Geo. J.L. & Pub. Pol’y 441, 447–49 (2019) (reviewing Lawson & Seidman, supra note 1). For discussion of a similar tendency in Kent et al., supra note 1, see infra note 87.
  17. See infra Part II.
  18. E.g., Barnett & Bernick, supra note 1, at 20–21 (asserting that the Constitution’s “organization and language sounds in eighteenth-century fiduciary law”). More cautious on this point is Kent et al., supra note 1, at 2179–80 (“[W]hat the three meanings we can attribute to the [Faithful Execution] Clauses have in common is that they are all part of the basic ways the private law constrains fiduciary discretion and power.”).
  19. Contra Lawson & Seidman, supra note 1, at 62 (“The fiduciary responsibilities of a trustee and an attorney do not differ in any way material to our project.”).
  20. See, e.g., Kent et al., supra note 1, 2180–81 (describing the “crystalliz[ing]” of “the ‘private’ fiduciary law we would recognize today” as not occurring in America until late in the eighteenth century, and concluding that “a fiduciary law of ‘private’ offices was unlikely to have been plucked off-the-rack by the Philadelphia Convention drafters and applied to public offices”); id. at 2179 (“Our historical findings about the original meaning of the Faithful Execution Clauses align with core features of modern fiduciary law . . . . ”); Leib & Shugerman, supra note 1, at 468 (suggesting doubt about whether “private fiduciary law was itself fully formed at the time of the founding”).
  21. Leib & Shugerman, supra note 1, at 468–69 (“[Article II of the U.S. Constitution] uses the language of faith and care to signal to courts and to executive officials that the President was supposed to be held to the same kinds of fiduciary obligations to which corporate officers, trustees, and lawyers are routinely held today in the private sector.”).
  22. But cf. Davis, supra note 7, at 1182–95 (discussing whether that is advisable).
  23. See infra notes 161 and 164 and accompanying text. We do recognize that there is a spectrum of views about how much the interpretation of the constitutional text should be informed by the legal categories at the time of ratification (as part of context). The more one separates the text from its legal context, the more room there is to integrate constitutional law and political morality, and the less constraint there is from the original understanding (with the familiar virtues and vices of that interpretive equilibrium).
  24. See infra Section I.A.

Secrecy Surrogates

Debates about how best to check executive branch abuses of secrecy focus on three sets of actors that have access to classified information and that traditionally have served—in one way or another—as our surrogates: congressional committees, federal courts, and leakers or whistleblowers. These actors provide only limited checks on the Executive’s misuse of secrecy, however. Most legal scholarship bemoans their flaws but concludes that the status quo is the best that we can do. This Article challenges this account, arguing that there is a different set of actors—a set of unsung “secrecy surrogates”—that can provide additional checks on the quality and legality of the Executive’s classified operations in the cyber, election, and counter-terrorism settings.

Technology companies, states and localities, and foreign allies have become an integral part of U.S. national security operations and enjoy some critical advantages over our traditional surrogates. These actors possess expertise about—and in some cases control—national security-related targets, making them essential partners for the Executive. Further, these surrogates have incentives to check the Executive in ways that advance the public law values of accuracy, accountability, and legality. Finally, unlike leakers, these unsung secrecy surrogates can challenge the Executive without revealing government secrets. These surrogates can only check government abuses of secrecy as long as the Executive requires their cooperation, but they have begun to supplement our traditional surrogates in important ways.

This Article maps the growing role of these unsung secrecy surrogates, argues that they are well-situated to address some persistent secrecy problems, and proposes ways to preserve and enhance the surrogates’ position in the secrecy ecosystem in the future.

Introduction

Misuse of government secrecy is in the headlines. Consider the revelation that White House officials transferred the transcript of President Trump’s call with Ukraine’s President to a highly classified stand-alone computer system to prevent leaks.1.Julian E. Barnes, Michael Crowley, Matthew Rosenberg & Mark Mazzetti, White House Classified Computer System Is Used To Hold Transcripts of Sensitive Calls, N.Y. Times (Sept. 29, 2019), https://www.nytimes.com/2019/09/27/us/politics/nsc-ukraine-call.html [https://perma.cc/F9ZC-68Z3]; seeDustin Volz, Andrew Duehren & Natalie Andrews, White House Official Feared Trump Transcript Leak Could Be Politically Damaging, Wall St. J. (Nov. 17, 2019, 5:52 AM), https://www.wsj.com/articles/white-house-official-feared-trump-transcript-leak-could-be-politically-damaging-11573942481 [https://perma.cc/LV53-PL42].Show More For many, this incident reflects a paradigmatic problem with government secrecy: actors in the Executive can employ it as a tool to avoid politically embarrassing or legally problematic revelations.2.See, e.g., Ayesha Rascoe & Franco Ordoñez, Former Officials Say White House’s Use of Secret System Is Unusual, ‘Disturbing,’ NPR (Sept. 27, 2019, 5:40 AM), https://www.npr.org/2019/09/27/764759182/former-officials-say-white-houses-use-of-secret-system-is-unusual-disturbing [https://perma.cc/7PBJ-5MF8] (quoting former White House official as stating that “[t]his seems to be nothing more than an abuse of the classification and the information security system to safeguard not the information, but to effect a cover-up” (internal quotation marks omitted)).Show More This episode proved to be a success story. A government whistleblower carefully followed statutory procedures, and the Intelligence Community Inspector General shared the whistleblower’s complaint with Congress, which held impeachment hearings to judge the President’s actions. Yet the case may be just as notable for its uniqueness, given how infrequently the whistleblowing process works as intended.

Indeed, legal scholars and political scientists have long decried the current state of affairs, in which the Executive exercises near total control over secret government information with few external checks. A substantial literature wrestles with how to manage the genuine need for secrecy about many national security operations in a democracy whose government should be accountable to the people. Government secrecy can foster four types of problems. First, the Executive can employ secrecy to conceal unlawful acts, such as spying on political enemies. Second, the Executive can use secrecy to conceal poor or controversial judgments or policies. Third, the Executive can use secrecy to conceal incompetent, empirically wrong, or insufficient intelligence and analysis.3.See Loch K. Johnson, Governing in the Absence of Angels: On the Practice of Intelligence Accountability in the United States, in Who’s Watching the Spies? Establishing Intelligence Service Accountability 57, 61 (Hans Born, Loch K. Johnson & Ian Leigh eds., 2005) (quoting intelligence scholar as stating that the major problem facing U.S. intelligence in 2005 was that the “CIA [had] not been gathering enough quality data”).Show More Each of these three types of missteps is embarrassing to the Executive and creates incentives to conceal the underlying action.4.See Frederick A.O. Schwarz Jr., Democracy in the Dark: The Seduction of Government Secrecy 2 (2015) (“[T]oo much is kept secret not to protect America but to keep embarrassing or illegal conduct from Americans.”).Show More Fourth, making decisions in secret insulates the Executive from having to justify and defend those decisions in public.5.Robert M. Pallitto & William G. Weaver, Presidential Secrecy and the Law 3 (2007) (noting executive interest in “maintain[ing] presidential prerogative against congressional inquiries and judicial orders”); id. at 6 (“Where a president may do what is desired in secret, there is no reason to withstand the ordeal of a political battle to achieve the same ends.”); H.R. Select Comm. To Investigate Covert Arms Transactions with Iran & S. Select Comm. on Secret Military Assistance to Iran and the Nicaraguan Opposition, 100th Cong., Rep. of the Congressional Committees Investigating the Iran-Contra Affair, Minority Report 450, 515 (Comm. Print 1987) (stating that the President should undertake “democratic persuasion” to develop support for his policies and that he will not succeed “unless the public is exposed to and persuaded by a clear, sustained, and principled debate on the merits”); David E. Pozen, Deep Secrecy, 62 Stan. L. Rev. 257, 279 (2010) (discussing how secrecy provides “insulation from scrutiny”).Show More All of these possible abuses of secrecy engender public skepticism about the government and make it harder for the public—and U.S. allies—to trust the Executive in cases in which secrecy truly is necessary.6.See Pozen, supra note 5, at 280; S. Select Comm. To Study Governmental Operations with Respect to Intel. Activities, Final Report, S. Rep. No. 94-755, at 4 (1976) (“[T]here are many necessary and proper governmental activities that must be conducted in secrecy. . . . [However,] intelligence activities conducted outside the framework of the Constitution and statutes can undermine the treasured values guaranteed in the Bill of Rights. Further, if the intelligence agencies act in ways inimical to declared national purposes, they damage the reputation, power, and influence of the United States abroad.”).Show More

Legal scholarship about government secrecy usually focuses on three sets of actors that check and balance executive branch secrecy to reduce abuse.7.See, e.g., Rahul Sagar, Secrets and Leaks: The Dilemma of State Secrecy (2013) (considering role of Congress, the courts, whistleblowers, and leakers in managing government secrecy); Pallitto & Weaver, supra note 5 (considering Congress, the courts, and leaks); Michael P. Colaresi, Democracy Declassified: The Secrecy Dilemma in National Security 181 (2014) (discussing role of legislatures, freedom of information laws, and press freedom laws in checking executive secrecy); Mark Fenster, The Transparency Fix: Secrets, Leaks, and Uncontrollable Government Information (2017) (discussing the role of the press, freedom of information laws, anti-corruption non-governmental organizations, and leaks); Pozen, supra note 5, at 269, 274 (focusing on deep secret keeping by executive officials, treating Congress and the courts as the primary recipients of shallow secrets, and assuming that very few private actors generally will be aware of government secrets); Johnson, supra note 3, at 58 (noting that the “most consistent and serious manifestation of intelligence oversight has come not from presidential commissions, but from the media and the Congress”); Josh Chafetz, Whose Secrets?, 127 Harv. L. Rev. F. 86, 87 (2013) (emphasizing the role of Congress in the secrecy regime).Show More Two sets of actors lie in the Executive’s co-equal branches of government. In the 1970s, Congress created two intelligence committees—the Senate Select Committee on Intelligence (“SSCI”) and the House Permanent Select Committee on Intelligence (“HPSCI”)—in the wake of the Church and Pike Committees’ reports. The intelligence committees, which conduct much of their work in secret, directly oversee the intelligence community and its activities.8.Members of the committees that oversee military and foreign relations issues also have access to classified information relevant to their legislative and oversight tasks.Show More The 1970s reforms also produced the Foreign Intelligence Surveillance Court, in which Article III judges authorize executive surveillance for foreign intelligence purposes. Article III courts also review classified information and activities in cases involving state secrets, Freedom of Information Act litigation, and certain criminal cases.

Although not constitutionally linked to the public in the way that Congress and federal courts are, whistleblowers and leakers constitute a third set of actors who attempt to bring abuses (or alleged abuses) to the attention of actors outside the Executive. In the national security setting, leakers emerge from within the executive branch itself and usually remain anonymous. They reveal classified information to the public, often by sharing it with journalists who report on the programs or intelligence contained in the leak. Whistleblowers, on the other hand, follow a statutory process of revealing abuses to their agency’s inspector general and then potentially to members of Congress.

Some view these three sets of actors as surrogates for the broader public, which does not and often should not have access to government secrets. As the literature makes clear, however, all three groups are imperfect surrogates. Congressional committees lack the robust incentives and sometimes the deep experience necessary to check the Executive’s national security activities. Federal courts often doubt their own competence to evaluate secret government programs and so accord substantial deference to executive claims that certain disclosures will harm national security. Leakers reveal information sporadically and can harm genuine national security equities when they do so. Leaking classified information also generally violates criminal law. And whistleblowing is fraught: those who blow the whistle often are subject to retaliation, even though statutes prohibit such responses.

Notwithstanding the flaws in the capabilities and performance of these three groups, many have concluded that the current state of affairs is the best we can do. This Article challenges that conclusion, arguing that this model overlooks at least three other sets of actors who increasingly can and do play a role in curbing misuses of executive secrecy.9.I have argued elsewhere that other actors and mechanisms constrain the Executive in its classified operations, including executive branch lawyers and norms of reason giving. See Ashley Deeks, Checks and Balances from Abroad, 83 U. Chi. L. Rev. 65 (2016) [hereinafter Deeks, Checks and Balances]; Ashley S. Deeks, Secret Reason-Giving, 129 Yale L.J. 612 (2020) [hereinafter Deeks, Reason-Giving]; Ashley S. Deeks, The Substance of Secret Agreements and the Role of Government Lawyers, 111 AJIL Unbound 474 (2018).Show More In recent decades, the national security threat landscape has shifted from one of overt, kinetic state-to-state conflict to a landscape dominated by non-state actors and clandestine, hostile operations by foreign governments using new technologies. In light of these new threats, which manifest themselves in the form of pernicious cyber operations, election interference, and terrorist acts, three groups have assumed critical—though underappreciated—roles in the U.S. national security ecosystem.10 10.Additional actors play a role in what this Article terms the “secrecy ecosystem.” These actors include the Privacy and Civil Liberties Oversight Board, the Presidential Intelligence Advisory Board (“PIAB”), and the Defense Advisory Board. See, e.g., Kenneth Michael Absher, Michael C. Desch & Roman Popadiuk, Privileged and Confidential: The Secret History of the President’s Intelligence Advisory Board (2012) (discussing the PIAB); Johnson, supra note 3, at 73 (noting that when Brent Scowcroft was Chair of the PIAB, he produced a “hard-hitting review” of the intelligence organization). For reasons of space, however, this Article does not address the role of these other secrecy surrogates.Show More U.S. technology companies,11 11.By “technology companies,” I mean large social media, software, and computer technology companies such as Google, Microsoft, and Facebook, as well as companies that provide cybersecurity services, such as FireEye/Mandiant, Crowdstrike, and IronNet.Show More states and localities, and foreign allies all possess the capacities and incentives to check problematic uses of government secrecy. For example, technology companies exchange threat information and operational details of cyber attacks with government officials, comparing intelligence and sometimes litigating to contest government decisions to keep programs secret.12 12.Similar interactions conceivably occur with utility companies and private banks because these companies have been targets of thousands of hostile cyber operations. The Intelligence Community’s 2019 Threat Assessment noted, “Russia has the ability to execute cyber attacks in the United States that generate localized, temporary disruptive effects on critical infrastructure—such as disrupting an electrical distribution network for at least a few hours—similar to those demonstrated in Ukraine in 2015 and 2016.” Daniel R. Coats, Dir. of Nat’l Intel., Statement for the Record: Worldwide Threat Assessment of the U.S. Intelli-gence Community 6 (2019), https://www.dni.gov/files/ODNI/documents/2019-ATA-SFR—SSCI.pdf [https://perma.cc/M7DS-WPY4]. Further, the Intelligence Community is authorized to disseminate “classified reports to critical infrastructure entities authorized to receive them.” Off. of the Dir. of Nat’l Intel., Dep’t of Homeland Sec., Dep’t of Def. & Dep’t of Just., Sharing of Cyber Threat Indicators and Defensive Measures by the Federal Government Under the Cybersecurity Information Sharing Act of 2015, at 13 (2016), https://www.us-cert. gov/sites/default/files/ais_files/Federal_Government_Sharing_Guidance_(103).pdf. [https://p erma.cc/HYH7-XUCD]. This Article focuses on technology companies because they are most likely to be sophisticated consumers (and providers) of intelligence on cyber operations. Some utility companies are reportedly insufficiently focused on cybersecurity and thus are poorly positioned—at least right now—to play a significant role in checking a sophisticated actor like the Executive. For a critique of utility company cybersecurity practices, see Joseph Marks & Tonya Riley, The Cybersecurity 202: Activist Wants Court To Name and Shame Electric Utilities for Violating Cybersecurity Rules, Wash. Post (Dec. 3, 2019, 4:35 AM), https://www.washingtonpost.com/news/powerpost/paloma/the-cybersecurity-202/2019/12/0­3/the-cybersecurity-202-activist-wants-court-to-name-and-shame-electric-utilities-for-viol­ating-cybersecurity-rules/5de550bf88e0fa652bbbdb18/ [https://perma.cc/N43C-Z4ZL]; Catherine Stupp, Utilities Are Prime Targets for Cyberattacks, WSJ Pro Cybersecurity (Aug. 7, 2019, 11:19 AM), https://www.wsj.com/articles/utilities-are-prime-target-for-cyberattacks-11565170204 [https://perma.cc/2S5J-EJU6] (“Utilities often don’t have enough employees with security clearance that lets them quickly get classified information about cyber threats from the federal government.”). Utilities, banks, and other operators of critical infrastructure may play a greater checking function in the future as they gain more experience with these threats. For a report recommending that the government enhance its sharing of classified information with utilities grid operators, see Nat’l Comm’n on Grid Resilience, Grid Resilience: Priorities for the Next Administration 28 (2020), https://gridresilience.org/wp-content/uploads/2020/08/NCGR-Report-2020-Full.pdf [https://perma.cc/5QMV-XUAN].Show More Foreign allies sometimes disagree with the U.S. intelligence community’s substantive intelligence judgments, challenging the United States to produce more or better intelligence.13 13.For example, the Trump Administration has tried for more than a year to persuade allies not to employ Huawei equipment in their 5G networks and has received pushback from a range of foreign governments. Robbie Gramer & Lara Seligman, Can the U.S.-U.K. Special Relationship Weather the Huawei Storm?, Foreign Pol’y (Jan. 30, 2020, 5:10 PM), https://foreignpolicy.com/2020/01/30/huawei-intelligence-united-states-britain-trump-5g-in­frastructure-concerns-digital-espionage-special-relationship-five-eyes/ [https://perma.cc/7Z­48-TKWX].Show More Although there are limited public examples of states and localities challenging secret executive activities directly in the election and cyber arenas, these sub-federal officials have the potential to do so because they possess fine-grained information about the election systems and critical infrastructure that are the targets of hostile cyber operations.14 14.For instance, in 2013, Los Angeles created a Cyber Intrusion Command Center, City of Los Angeles, Exec. Directive No. 2 (Oct. 30, 2013), https://www.lamayor.org/sites/g/files/wph446/f/page/file/ED2_with_signature_and_letterhead.pdf?1426620047 [https://perma.cc/6E9C-JC7M] (anticipating collaboration with the FBI and other federal agencies), and in 2017, New York City created its own Cyber Command, City of New York, Exec. Order No. 28 (July 11, 2017), https://www1.nyc.gov/a­ssets/home/downloads/pdf/executive-orders/2017/eo_28.pdf [https://perma.cc/7CNG-HA­MH] (anticipating collaboration with federal and state government agencies and the private sector); see also Brennan Weiss, Inside New York City Cyber Command, Bus. Insider (May 5, 2018, 8:00 AM), https://www.businessinsider.com/nyc-cyber-command-protecting-new-yorkers-2018-4 [https://perma.cc/46VA-MWC7].Show More Further, they historically have challenged certain federal counter-terrorism programs, which suggests that they may start to push back in the election and cyber settings as they gain expertise about the threat landscape.15 15.See Matthew C. Waxman, National Security Federalism in the Age of Terror, 64 Stan. L. Rev. 289, 333 (2012).Show More

These three groups have several important advantages over our traditional secrecy surrogates. First, they possess specific expertise about the new threats and new targets that Congress, the courts, and leakers might not. Certain technology companies and many allies are highly specialized in intelligence-gathering and analysis, and so are particularly well-suited to detect problematic executive performance in the secrecy space.16 16.A possible analogy is to the role of auditors, who verify the accuracy of the government’s records (here, its intelligence and analysis) and point out deficiencies in its operations. For the use of the concept of substantive audits in the intelligence setting, see Eric Posner, It’s Time to Audit America’s Secrets, Foreign Pol’y (Feb. 2, 2018, 5:13 PM), https://foreignpolicy.com/2018/02/02/its-time-to-audit-all-of-americas-secrets/ [https://per­ma.cc/LK65-W4NA]; Elizabeth Goitein & J. William Leonard, Opinion, America’s Unnecessary Secrets, N.Y. Times (Nov. 7, 2011), https://www.nytimes.com/­2011/11/07/opinion/national-security-and-americas-unnecessary-secrets.html [https://perma.cc/2UAL-GFKL] (arguing that one way to combat government over-classification is to allow agencies’ inspectors general to “audit officials’ classification decisions”).Show More Second, each of the three groups brings to the table irreplaceable access to information and infrastructure that the Executive needs to perform its job.17 17.See Michael Wines, State Officials Say They Are Told Too Little About Election Threats, N.Y. Times (Feb. 19, 2018), https://www.nytimes.com/2018/02/19/us/elections-states-hacking.html [https://perma.cc/77LU-KFCV] (describing the relationship between the Department of Homeland Security and local election officials as an “arranged marriage”).Show More For technology companies, it is the ability to observe and defend the front lines of critical infrastructure systems, attribute the sources of cyber attacks, and operate the very systems that are subject to foreign manipulation. For states and localities, it is control over and knowledge about election operations and machinery (and other critical infrastructure at the sub-federal level), as well as ground-level intelligence about terrorist activities inside the United States. For foreign allies, it is intelligence and expertise that the United States may not possess about shared threats. The Executive has persistent incentives to share intelligence with these actors to allow them to take necessary steps to enhance U.S. national security. This, in turn, renders them an audience that the Executive must persuade of the soundness of its intelligence and proposed operations.18 18.Deeks, Reason-Giving, supra note 9.Show More Third, unlike leakers, these three groups are positioned to challenge secret government operations without revealing those operations.19 19.Those who frame the government secrecy debate as a choice between secrecy and disclosure thus are misframing the issue. See Philip H. Melanson, Secrecy Wars 8, 183 (2001) (describing the “ongoing battle between secret keepers and those seeking access”).Show More

These, then, are our unsung “secrecy surrogates”: actors who are given access to secret information that average U.S. citizens are not and who can improve secret executive operations and help mitigate abuses. By “surrogates,” I do not mean that these groups have a direct constitutional, contractual, or agency relationship with the national populace—they generally do not.20 20.There is an irony here: in certain areas of classified government operations, these actors, which lack a constitutional relationship to our national polity, may be better positioned to alter the non-public behavior of the Executive than the courts and congressional committees, which are our direct surrogates.Show More Instead, I mean that these groups serve as our surrogates in a more general sense: as actors who take our place or are given a particular role in government operations because we are not able to serve in that role ourselves. In particular, these surrogates are positioned to enhance the Executive’s adherence to public law values by (1) stimulating the Executive to improve the accuracy of its intelligence; (2) diminishing the Executive’s opportunity to undertake illegal actions; and (3) increasing the Executive’s accountability for its classified choices.21 21.See Jody Freeman, Private Parties, Public Functions and the New Administrative Law, 52 Admin. L. Rev. 813, 818–19 (2000) (listing “openness, fairness, participation, consistency, rationality, impartiality, and accessibility of judicial review” as well as accountability and legality as public law values); id. at 819 (“Private actors are not just rent-seekers that exacerbate the traditional democracy problem in administrative law; they are also regulatory resources capable of contributing to the efficacy and legitimacy of administration.”); Michael Taggart, The Province of Administrative Law Determined?, in The Province of Administrative Law 1, 3 (Michael Taggart ed., 1997) (defining public law values to include openness, participation, accountability, honesty, and rationality); Mark Aronson, A Public Lawyer’s Response to Privatisation and Outsourcing, in The Province of Administrative Law 40, 43 (Michael Taggart ed., 1997).Show More It is difficult to obtain empirical, unclassified information about the full range of effects of these secrecy surrogates, and so this Article’s conclusions are necessarily tentative. However, based on available analyses of the ways that technology companies, foreign allies, and states and localities have behaved to date in the surveillance,22 22.Alan Z. Rozenshtein, Surveillance Intermediaries, 70 Stan. L. Rev. 99, 106 (2018).Show More cybersecurity,23 23.Kristen E. Eichensehr, Public-Private Cybersecurity, 95 Tex. L. Rev. 467, 471 (2017) [hereinafter Eichensehr, Public-Private Cybersecurity].Show More and counter-terrorism24 24.Waxman, supra note 15.Show More settings, it is clear that these actors can help ensure that U.S. intelligence operations are attentive to legal, procedural, and accuracy concerns and have begun to play this role.

David Pozen has argued that it is preferable in a democratic system like ours, in which the government must keep certain information secret, for those secrets to be shallow rather than deep. (By deep secrets, he means government secrets that only a small number of similarly situated officials know.)25 25.Pozen, supra note 5, at 274 (“[A] government secret is deep if a small group of similarly situated officials conceals its existence from the public and from other officials, such that the outsiders’ ignorance precludes them from learning about, checking, or influencing the keepers’ use of the information.”).Show More In his view, which I share, our system should favor shallow secrecy whenever possible, because doing so “will systematically lead to . . . outcomes that are deemed acceptable from a greater variety of perspectives, that have been more thoroughly reasoned and refined through a dialogic vetting process, that are better documented, that take longer to be finalized, and that are more likely to be publicized.”26 26.Id. at 275.Show More He argues that we can shift a secret from being deep to shallow by expanding the number and types of people who know the secret, even if the underlying information remains classified. Pozen, however, contemplates this as occurring primarily by expanding the number and type of secret keepers within the executive branch itself, as well as within Congress.27 27.Id. at 329–30, 333.Show More This Article argues that the Executive has, by necessity, begun to expand and diversify the number and type of secret keepers in areas that reach far beyond the executive branch or Congress. In so doing, the government is both decreasing the depth of its secrets and positioning these actors to check some of the persistent problems of government secrecy: the concealment of incompetent execution or illegality and the ability of the Executive to avoid justifying its decisions to outsiders.

This new system of surrogates, like the existing one it supplements, is imperfect. For many of the same reasons that our traditional secrecy surrogates do not act as fully faithful agents for the public, these secrecy surrogates offer only partial fixes to our secrecy challenges, even if they are independently powerful actors.28 28.Several scholars have considered government secrecy problems through a principal-agent lens. Daniel Epps identifies three mechanisms that could help reduce agency costs in the secrecy setting, one of which is the use of proxies. Daniel Epps, Note, Mechanisms of Secrecy, 121 Harv. L. Rev. 1556, 1558 (2008). Epps explores the use of proxies only briefly, however, and focuses on government actors who have a direct duty to the public (the FISC and the Executive). His note does not consider the operations of other proxies, such as those treated here as surrogates. Sidney Shapiro and Reina Steinzor use agency theory to evaluate how to hold Congress and the Executive accountable to the public in the face “burgeoning secrecy.” Sidney A. Shapiro & Rena I. Steinzor, The People’s Agent: Executive Branch Secrecy and Accountability in an Age of Terrorism, 69 Law & Contemp. Probs. 99, 101 (2006). They focus on who should have the power to declassify information, however, not on how other actors can check secret U.S. military and intelligence activities.Show More They have their own pathologies and policy preferences, have incomplete access to classified information, and could serve as a new source of leaks. These actors will not supplant the existing messiness of today’s interplay among the Executive, Congress, the courts, and leakers. Rather, they will supplement the reach of existing surrogates, expanding what Jack Goldsmith has framed, in the wider national security setting, as a “synopticon”—a distributed network of actors that surveils the Executive.29 29.Jack Goldsmith, Power and Constraint: The Accountable Presidency After 9/11, at 205–07 (2012).Show More This Article argues that adding knowledgeable players to the “secrecy synopticon” who can provide increased checks and monitoring without sacrificing much secrecy is a desirable development worth sustaining.

This Article makes three contributions. First, it shows descriptively that there are several unsung actors in the government secrecy ecosystem that help guard against the Executive’s misuse of secrecy, and that any evaluation of government secrecy that ignores these actors is importantly incomplete. It is well-understood that these types of actors play an important checking function in the public parts of government operations.30 30.See, e.g., Freeman, supra note 21, at 816–17 (describing how private actors participate publicly in governance through the regulatory process).Show More This Article shows that these actors also serve a checking function behind the veil of secrecy. Second, it analyzes the features of these groups that allow them to provide these checks. It draws attention to their role as necessary actors in the conduct of national security today; their on-the-ground expertise; and their ability to challenge the Executive without disclosing classified information. Third, the Article offers a normative defense of this development, identifying the surrogates’ incentives to improve the quality of intelligence, challenge legally questionable executive activities, and demand reasons for secret decisions. Thinking about the existence, role, and possibilities of these unsung secrecy surrogates can sharpen how we approach the challenges of government secrecy; identify where the coincidences of interest lie between these surrogates and the national public; suggest ways to preserve salutary overlaps in interest; and allow us to see where the most pressing gaps in oversight remain.

Part I identifies how the Executive can abuse secrecy, as well as the strengths and weaknesses of our traditional secrecy surrogates. Part II argues that several recent developments in national security threats have positioned technology companies, states and localities, and foreign allies to serve as unsung secrecy surrogates. Part III explores the incentives that these actors have to serve as checks on executive abuses of secrecy and considers how those incentives are aligned with the public law values of legality, accountability, rationality, participation, and, to some extent, transparency. Part IV places these unsung surrogates in the context of a broader “secrecy synopticon.” It addresses challenges to the argument that the unsung surrogates can perform robust checking functions and proposes modest ways to enhance their role in the synopticon.

  1. * Professor, University of Virginia Law School. Thanks to George Cohen, Jen Daskal, Kristen Eichensehr, Mike Flowers, John Harrison, Debbie Hellman, Rebecca Ingber, Nate Jones, Aaron Karczmer, Matt Olsen, Daphna Renan, Rich Schragger, Paul Stephan, and participants in the 2019 Duke-Virginia Foreign Relations Roundtable and in workshops at Harvard Law School and the University of Houston Law Center for very helpful conversations and comments.

  2. Julian E. Barnes, Michael Crowley, Matthew Rosenberg & Mark Mazzetti, White House Classified Computer System Is Used To Hold Transcripts of Sensitive Calls, N.Y. Times (Sept. 29, 2019), https://www.nytimes.com/2019/09/27/us/politics/nsc-ukraine-call.html [https://perma.cc/F9ZC-68Z3]; see Dustin Volz, Andrew Duehren & Natalie Andrews, White House Official Feared Trump Transcript Leak Could Be Politically Damaging, Wall St. J. (Nov. 17, 2019, 5:52 AM), https://www.wsj.com/articles/white-house-official-feared-trump-transcript-leak-could-be-politically-damaging-11573942481 [https://perma.cc/LV53-PL42].

  3. See, e.g., Ayesha Rascoe & Franco Ordoñez, Former Officials Say White House’s Use of Secret System Is Unusual, ‘Disturbing,’ NPR (Sept. 27, 2019, 5:40 AM), https://www.npr.org/2019/09/27/764759182/former-officials-say-white-houses-use-of-secret-system-is-unusual-disturbing [https://perma.cc/7PBJ-5MF8] (quoting former White House official as stating that “[t]his seems to be nothing more than an abuse of the classification and the information security system to safeguard not the information, but to effect a cover-up” (internal quotation marks omitted)).

  4. See Loch K. Johnson, Governing in the Absence of Angels: On the Practice of Intelligence Accountability in the United States, in Who’s Watching the Spies? Establishing Intelligence Service Accountability 57, 61 (Hans Born, Loch K. Johnson & Ian Leigh eds., 2005) (quoting intelligence scholar as stating that the major problem facing U.S. intelligence in 2005 was that the “CIA [had] not been gathering enough quality data”).

  5. See Frederick A.O. Schwarz Jr., Democracy in the Dark: The Seduction of Government Secrecy 2 (2015) (“[T]oo much is kept secret not to protect America but to keep embarrassing or illegal conduct from Americans.”).

  6. Robert M. Pallitto & William G. Weaver, Presidential Secrecy and the Law 3 (2007) (noting executive interest in “maintain[ing] presidential prerogative against congressional inquiries and judicial orders”); id. at 6 (“Where a president may do what is desired in secret, there is no reason to withstand the ordeal of a political battle to achieve the same ends.”); H.R. Select Comm. To Investigate Covert Arms Transactions with Iran & S. Select Comm. on Secret Military Assistance to Iran and the Nicaraguan Opposition, 100th Cong., Rep. of the Congressional Committees Investigating the Iran-Contra Affair, Minority Report 450, 515 (Comm. Print 1987) (stating that the President should undertake “democratic persuasion” to develop support for his policies and that he will not succeed “unless the public is exposed to and persuaded by a clear, sustained, and principled debate on the merits”); David E. Pozen, Deep Secrecy, 62 Stan. L. Rev. 257, 279 (2010) (discussing how secrecy provides “insulation from scrutiny”).

  7. See Pozen, supra note 5, at 280; S. Select Comm. To Study Governmental Operations with Respect to Intel. Activities, Final Report, S. Rep. No. 94-755, at 4 (1976) (“[T]here are many necessary and proper governmental activities that must be conducted in secrecy. . . . [However,] intelligence activities conducted outside the framework of the Constitution and statutes can undermine the treasured values guaranteed in the Bill of Rights. Further, if the intelligence agencies act in ways inimical to declared national purposes, they damage the reputation, power, and influence of the United States abroad.”).

  8. See, e.g., Rahul Sagar, Secrets and Leaks: The Dilemma of State Secrecy (2013) (considering role of Congress, the courts, whistleblowers, and leakers in managing government secrecy); Pallitto & Weaver, supra note 5 (considering Congress, the courts, and leaks); Michael P. Colaresi, Democracy Declassified: The Secrecy Dilemma in National Security 181 (2014) (discussing role of legislatures, freedom of information laws, and press freedom laws in checking executive secrecy); Mark Fenster, The Transparency Fix: Secrets, Leaks, and Uncontrollable Government Information (2017) (discussing the role of the press, freedom of information laws, anti-corruption non-governmental organizations, and leaks); Pozen, supra note 5, at 269, 274 (focusing on deep secret keeping by executive officials, treating Congress and the courts as the primary recipients of shallow secrets, and assuming that very few private actors generally will be aware of government secrets); Johnson, supra note 3, at 58 (noting that the “most consistent and serious manifestation of intelligence oversight has come not from presidential commissions, but from the media and the Congress”); Josh Chafetz, Whose Secrets?, 127 Harv. L. Rev. F. 86, 87 (2013) (emphasizing the role of Congress in the secrecy regime).

  9. Members of the committees that oversee military and foreign relations issues also have access to classified information relevant to their legislative and oversight tasks.

  10. I have argued elsewhere that other actors and mechanisms constrain the Executive in its classified operations, including executive branch lawyers and norms of reason giving. See Ashley Deeks, Checks and Balances from Abroad, 83 U. Chi. L. Rev. 65 (2016) [hereinafter Deeks, Checks and Balances]; Ashley S. Deeks, Secret Reason-Giving, 129 Yale L.J. 612 (2020) [hereinafter Deeks, Reason-Giving]; Ashley S. Deeks, The Substance of Secret Agreements and the Role of Government Lawyers, 111 AJIL Unbound 474 (2018).

  11. Additional actors play a role in what this Article terms the “secrecy ecosystem.” These actors include the Privacy and Civil Liberties Oversight Board, the Presidential Intelligence Advisory Board (“PIAB”), and the Defense Advisory Board. See, e.g., Kenneth Michael Absher, Michael C. Desch & Roman Popadiuk, Privileged and Confidential: The Secret History of the President’s Intelligence Advisory Board (2012) (discussing the PIAB); Johnson, supra note 3, at 73 (noting that when Brent Scowcroft was Chair of the PIAB, he produced a “hard-hitting review” of the intelligence organization). For reasons of space, however, this Article does not address the role of these other secrecy surrogates.

  12. By “technology companies,” I mean large social media, software, and computer technology companies such as Google, Microsoft, and Facebook, as well as companies that provide cybersecurity services, such as FireEye/Mandiant, Crowdstrike, and IronNet.

  13. Similar interactions conceivably occur with utility companies and private banks because these companies have been targets of thousands of hostile cyber operations. The Intelligence Community’s 2019 Threat Assessment noted, “Russia has the ability to execute cyber attacks in the United States that generate localized, temporary disruptive effects on critical infrastructure—such as disrupting an electrical distribution network for at least a few hours—similar to those demonstrated in Ukraine in 2015 and 2016.” Daniel R. Coats, Dir. of Nat’l Intel., Statement for the Record: Worldwide Threat Assessment of the U.S. Intelli-
    gence Community 6 (2019), https://www.dni.gov/files/ODNI/documents/2019-ATA-SFR—SSCI.pdf [https://perma.cc/M7DS-WPY4]. Further, the Intelligence Community is authorized to disseminate “classified reports to critical infrastructure entities authorized to receive them.” Off. of the Dir. of Nat’l Intel., Dep’t of Homeland Sec., Dep’t of Def. & Dep’t of Just., Sharing of Cyber Threat Indicators and Defensive Measures by the Federal Government Under the Cybersecurity Information Sharing Act of 2015, at 13 (2016), https://www.us-cert. gov/sites/default/files/ais_files/Federal_Government_Sharing_Guidance_(103).pdf. [https://p erma.cc/HYH7-XUCD]. This Article focuses on technology companies because they are most likely to be sophisticated consumers (and providers) of intelligence on cyber operations. Some utility companies are reportedly insufficiently focused on cybersecurity and thus are poorly positioned—at least right now—to play a significant role in checking a sophisticated actor like the Executive. For a critique of utility company cybersecurity practices, see Joseph Marks & Tonya Riley, The Cybersecurity 202: Activist Wants Court To Name and Shame Electric Utilities for Violating Cybersecurity Rules, Wash. Post (Dec. 3, 2019, 4:35 AM), https://www.washingtonpost.com/news/powerpost/paloma/the-cybersecurity-202/2019/12/0­3/the-cybersecurity-202-activist-wants-court-to-name-and-shame-electric-utilities-for-viol­ating-cybersecurity-rules/5de550bf88e0fa652bbbdb18/ [https://perma.cc/N43C-Z4ZL]; Catherine Stupp, Utilities Are Prime Targets for Cyberattacks, WSJ Pro Cybersecurity (Aug. 7, 2019, 11:19 AM), https://www.wsj.com/articles/utilities-are-prime-target-for-cyberattacks-11565170204 [https://perma.cc/2S5J-EJU6] (“Utilities often don’t have enough employees with security clearance that lets them quickly get classified information about cyber threats from the federal government.”). Utilities, banks, and other operators of critical infrastructure may play a greater checking function in the future as they gain more experience with these threats. For a report recommending that the government enhance its sharing of classified information with utilities grid operators, see Nat’l Comm’n on Grid Resilience, Grid Resilience: Priorities for the Next Administration 28 (2020), https://gridresilience.org/wp-content/uploads/2020/08/NCGR-Report-2020-Full.pdf [https://perma.cc/5QMV-XUAN].

  14. For example, the Trump Administration has tried for more than a year to persuade allies not to employ Huawei equipment in their 5G networks and has received pushback from a range of foreign governments. Robbie Gramer & Lara Seligman, Can the U.S.-U.K. Special Relationship Weather the Huawei Storm?, Foreign Pol’y (Jan. 30, 2020, 5:10 PM), https://foreignpolicy.com/2020/01/30/huawei-intelligence-united-states-britain-trump-5g-in­frastructure-concerns-digital-espionage-special-relationship-five-eyes/ [https://perma.cc/7Z­48-TKWX].

  15. For instance, in 2013, Los Angeles created a Cyber Intrusion Command
    Center, City of Los Angeles, Exec. Directive No. 2 (Oct. 30, 2013), https://www.lamayor.org/sites/g/files/wph446/f/page/file/ED2_with_signature_and_letterhead.pdf?1426620047 [https://perma.cc/6E9C-JC7M] (anticipating collaboration with the FBI and other federal agencies), and in 2017, New York City created its own Cyber Command, City of New York, Exec. Order No. 28 (July 11, 2017), https://www1.nyc.gov/a­ssets/home/downloads/pdf/executive-orders/2017/eo_28.pdf [https://perma.cc/7CNG-HA­MH] (anticipating collaboration with federal and state government agencies and the private sector); see also Brennan Weiss, Inside New York City Cyber Command, Bus. Insider (May 5, 2018, 8:00 AM), https://www.businessinsider.com/nyc-cyber-command-protecting-new-yorkers-2018-4 [https://perma.cc/46VA-MWC7].

  16. See Matthew C. Waxman, National Security Federalism in the Age of Terror, 64 Stan. L. Rev. 289, 333 (2012).

  17.  A possible analogy is to the role of auditors, who verify the accuracy of the
    government’s records (here, its intelligence and analysis) and point out deficiencies
    in its operations. For the use of the concept of substantive audits in the intelligence
    setting, see Eric Posner, It’s Time to Audit America’s Secrets, Foreign Pol’y (Feb. 2, 2018, 5:13 PM), https://foreignpolicy.com/2018/02/02/its-time-to-audit-all-of-americas-secrets/ [https://per­ma.cc/LK65-W4NA]; Elizabeth Goitein & J. William Leonard,
    Opinion, America’s Unnecessary Secrets, N.Y. Times (Nov. 7, 2011), https://www.nytimes.com/­2011/11/07/opinion/national-security-and-americas-unnecessary-secrets.html [https://perma.cc/2UAL-GFKL] (arguing that one way to combat government over-classification is to allow agencies’ inspectors general to “audit officials’ classification decisions”).

  18. See Michael Wines, State Officials Say They Are Told Too Little About Election Threats, N.Y. Times (Feb. 19, 2018), https://www.nytimes.com/2018/02/19/us/elections-states-hacking.html [https://perma.cc/77LU-KFCV] (describing the relationship between the Department of Homeland Security and local election officials as an “arranged marriage”).

  19. Deeks, Reason-Giving, supra note 9.

  20.  Those who frame the government secrecy debate as a choice between secrecy and disclosure thus are misframing the issue. See Philip H. Melanson, Secrecy Wars 8, 183 (2001) (describing the “ongoing battle between secret keepers and those seeking access”).

  21. There is an irony here: in certain areas of classified government operations, these actors, which lack a constitutional relationship to our national polity, may be better positioned to alter the non-public behavior of the Executive than the courts and congressional committees, which are our direct surrogates.

  22. See Jody Freeman, Private Parties, Public Functions and the New Administrative Law, 52 Admin. L. Rev. 813, 818–19 (2000) (listing “openness, fairness, participation, consistency, rationality, impartiality, and accessibility of judicial review” as well as accountability and legality as public law values); id. at 819 (“Private actors are not just rent-seekers that exacerbate the traditional democracy problem in administrative law; they are also regulatory resources capable of contributing to the efficacy and legitimacy of administration.”); Michael Taggart, The Province of Administrative Law Determined?, in The Province of Administrative Law 1, 3 (Michael Taggart ed., 1997) (defining public law values to include openness, participation, accountability, honesty, and rationality); Mark Aronson, A Public Lawyer’s Response to Privatisation and Outsourcing, in The Province of Administrative Law 40, 43 (Michael Taggart ed., 1997).

  23. Alan Z. Rozenshtein, Surveillance Intermediaries, 70 Stan. L. Rev. 99, 106 (2018).

  24. Kristen E. Eichensehr, Public-Private Cybersecurity, 95 Tex. L. Rev. 467, 471 (2017) [hereinafter Eichensehr, Public-Private Cybersecurity].

  25. Waxman, supra note 15.

  26. Pozen, supra note 5, at 274 (“[A] government secret is deep if a small group of similarly situated officials conceals its existence from the public and from other officials, such that the outsiders’ ignorance precludes them from learning about, checking, or influencing the keepers’ use of the information.”).

  27. Id. at 275.

  28. Id. at 329–30, 333.

  29. Several scholars have considered government secrecy problems through a principal-agent lens. Daniel Epps identifies three mechanisms that could help reduce agency costs in the secrecy setting, one of which is the use of proxies. Daniel Epps, Note, Mechanisms of Secrecy, 121 Harv. L. Rev. 1556, 1558 (2008). Epps explores the use of proxies only briefly, however, and focuses on government actors who have a direct duty to the public (the FISC and the Executive). His note does not consider the operations of other proxies, such as those treated here as surrogates. Sidney Shapiro and Reina Steinzor use agency theory to evaluate how to hold Congress and the Executive accountable to the public in the face “burgeoning secrecy.” Sidney A. Shapiro & Rena I. Steinzor, The People’s Agent: Executive Branch Secrecy and Accountability in an Age of Terrorism, 69 Law & Contemp. Probs. 99, 101 (2006). They focus on who should have the power to declassify information, however, not on how other actors can check secret U.S. military and intelligence activities.

  30. Jack Goldsmith, Power and Constraint: The Accountable Presidency After 9/11, at 205–07 (2012).

  31. See, e.g., Freeman, supra note 21, at 816–17 (describing how private actors participate publicly in governance through the regulatory process).