Harvard Law Review Harvard Law Review Harvard Law Review

Fifth Amendment

Decryption Originalism: The Lessons of Burr

The full text of this Article may be found by clicking on the PDF link to the left.

The Supreme Court is likely to rule soon on how the Fifth Amendment privilege against self-incrimination applies to compelled decryption of a digital device. When the Court rules, the original understanding of the Fifth Amendment may control the outcome. This Article details an extraordinary case that illuminates the original understanding of the privilege and its application to compelled decryption. During the 1807 treason trial of Aaron Burr, with Chief Justice John Marshall presiding, the government asked Burr’s private secretary if he knew the cipher to an encrypted letter Burr had sent to a co-conspirator. Burr’s secretary invoked the privilege against self-incrimination, leading to an extensive debate on the meaning of the privilege and an opinion from the Chief Justice.

The Burr dispute presents a remarkable opportunity to unearth the original understanding of the Fifth Amendment and its application to surprisingly modern facts. The lawyers in Burr were celebrated and experienced advocates. The Chief Justice allowed them to argue the Fifth Amendment question in exhaustive detail. And an attorney recorded the entire argument in shorthand, including dozens of legal citations to the specific pages of the authorities the lawyers invoked. The rich materials allow us to reconstruct for the first time precisely how the privilege was understood by leading lawyers and Chief Justice John Marshall soon after the Fifth Amendment’s ratification. The Article presents that reconstruction, and it concludes by applying Burr’s lessons to the modern problem of compelled decryption of digital devices such as cell phones and computers.

Introduction

Lower courts recently have divided on how the privilege against self-incrimination applies to accessing encrypted digital evidence. The issue usually comes up when the police seize a computer or cell phone and have a warrant to search it for evidence. The police can’t execute the search, however, because the device is encrypted. Prosecutors get a court order compelling a suspect to unlock the device, either by disclosing the password or by entering it. The suspect then pleads the Fifth, forcing a court to determine whether entering or disclosing the password would force him to “be a witness against himself.”1×1. U.S. Const. amend. V (“No person . . . shall be compelled in any criminal case to be a witness against himself . . . .”).

The cases applying the Fifth Amendment to compelled decryption are all over the map.2×2. See generally Laurent Sacharoff, Unlocking the Fifth Amendment: Passwords and Encrypted Devices, 87 Fordham L. Rev. 203, 207 (2018) (describing the Fifth Amendment law of compelled access to encrypted data as a “fundamental question bedeviling courts and scholars” (footnote omitted)). The confusion is understandable. Lower courts must follow Supreme Court precedent. But the Supreme Court’s cases in this area are notoriously difficult, and none of the cases involve facts that resemble compelled decryption.3×3. The major cases have involved the production of documents by subpoena or its equivalent. See generally United States v. Hubbell, 530 U.S. 27, 30 (2000); Fisher v. United States, 425 U.S. 391, 393 (1976). The lower court disagreement makes Supreme Court review highly likely. And the prospect of Supreme Court review introduces an important new question to the debate: How should an originalist Justice decide such a case?

This is not just a hypothetical question. Three members of the Supreme Court, Justices Clarence Thomas, Neil Gorsuch, and Amy Coney Barrett, are committed originalists. Two of them have strongly suggested that they are prepared to reject current Fifth Amendment doctrine in favor of whatever an originalist approach might reveal.4×4. See Hubbell, 530 U.S. at 49 (Thomas, J., concurring) (stating that the Fifth Amendment act of production doctrine “may be inconsistent with the original meaning of the Fifth Amendment’s Self-Incrimination Clause” and that he “would be willing to reconsider the scope and meaning” of the clause in a future case); Carpenter v. United States, 138 S. Ct. 2206, 2271 (2018) (Gorsuch, J., dissenting) (stating that although existing Fifth Amendment precedent treats the privilege against self-incrimination as “applicable only to testimony, not the production of incriminating evidence[,] . . . there is substantial evidence that the privilege . . . was also originally understood to protect a person from being forced to turn over potentially incriminating evidence”). Several other Justices are influenced by originalism and might do the same.5×5. See, e.g., Confirmation Hearing on the Nomination of Elena Kagan to Be an Associate Justice of the Supreme Court of the United States: Hearing Before the S. Comm. on the Judiciary, 111th Cong. 62 (2010) (statement of Elena Kagan) (“Either way we apply what they say, what they meant to do. So in that sense, we are all originalists.”). When compelled decryption reaches the Supreme Court, the outcome may very well hinge on the original understanding of the Fifth Amendment and how it applies to entering or disclosing a password.

At first blush, this sounds impossible to know. The Framers did not have cell phones. They did not have computers. Can we ever know how the eighteenth-century Constitution would have been understood to apply to the latest twenty-first-century facts? The question calls to mind Justice Alito’s criticism of Justice Scalia’s originalist decision in United States v. Jones,6×6. 565 U.S. 400 (2012). which held that it was a Fourth Amendment search to attach a hidden GPS device to a car.7×7. Id. at 404. According to Justice Alito, it was “almost impossible to think of late-18th-century situations that are analogous” to those facts.8×8. Id. at 420 (Alito, J., concurring in the judgment). Justice Alito derided the majority’s originalist approach with a rhetorical question: “Is it possible to imagine a case in which a constable secreted himself somewhere in a coach and remained there for a period of time in order to monitor the movements of the coach’s owner?”9×9. Id.

But historical coincidences can happen. This Article is about one of them. It turns out that a remarkable historical case exists on how the Fifth Amendment applies to compelled decryption. It’s not just a case, but a whopper of a case: An opinion written by Chief Justice John Marshall himself.10×10. See United States v. Burr (In re Willie), 25 F. Cas. 38, 39 (C.C.D. Va. 1807) (No. 14,692e). In 1807, Chief Justice Marshall presided over the treason trial of former Vice President Aaron Burr.11×11. See infra section I.A, pp. 915–17. Burr had written his alleged co-conspirators in cipher, encrypting his letters to avoid their contents being understood by nosy interlopers.12×12. See infra section I.B, pp. 917–20. During its investigation, the grand jury wanted to learn the contents of one of Burr’s encrypted letters to better understand Burr’s plot. The prosecution called Burr’s private secretary to the stand and asked him whether he understood the letter’s cipher.13×13. See infra section I.D, pp. 922–24. The secretary refused to answer, citing the privilege against self-incrimination. In United States v. Burr,14×14. 25 F. Cas. 38. Chief Justice Marshall ruled that Burr’s secretary was required to answer the government’s questions.

Viewed in isolation, Chief Justice Marshall’s opinion offers a fascinating lens for an originalist approach to compelled decryption. Scholars have disagreed on the original understanding of the Fifth Amendment privilege.15×15. Much of the disagreement focuses on what role the Fifth Amendment privilege was understood to serve and when modern concepts of the privilege were established. See, e.g., Leonard W. Levy, Origins of the Fifth Amendment: The Right Against Self-Incrimination, at vii–viii (1968) (arguing that the privilege as known today was largely and firmly established at common law); Katharine B. Hazlett, The Nineteenth Century Origins of the Fifth Amendment Privilege Against Self-Incrimination, 42 Am. J. Legal Hist. 235, 240 (1998) (arguing that the modern privilege developed in the nineteenth century); John H. Langbein, The Historical Origins of the Privilege Against Self-Incrimination at Common Law, 92 Mich. L. Rev. 1047, 1048 (1994) (contending that the notion of the privilege as enabling a defendant not to testify at trial was a relatively modern development); Eben Moglen, Essay, Taking The Fifth: Reconsidering the Origins of the Constitutional Privilege Against Self-Incrimination, 92 Mich. L. Rev. 1086, 1087 (1994) (arguing that the privilege became more established and less contentious in the nineteenth century); John Fabian Witt, Making the Fifth: The Constitutionalization of American Self-Incrimination Doctrine, 1791–1903, 77 Tex. L. Rev. 825, 831 (1999) (similar); John H. Wigmore, The Privilege Against Self-Crimination; Its History, 15 Harv. L. Rev. 610, 610, 623 (1902) (arguing that the privilege originated in the ecclesiastical courts, notably the Star Chamber, and was then continued by early common law courts). Yet here we have an opinion on the privilege, written by the celebrated Chief Justice just sixteen years after the Fifth Amendment’s ratification, that has not been closely studied. The opinion specifically addresses how the privilege applies to efforts to learn about the contents of an encrypted letter. Given the date of the decision, the similarity of the facts to the present, and the prominence of John Marshall, a close reading of Burr can offer unique insights into the original public meaning of the privilege and its application to modern facts.

But a skeptical originalist may be unconvinced. How can we know that Marshall’s 1807 opinion reflects the understanding of 1791? The decision does not include a close textual reading of the Fifth Amendment. It mentions only one precedent in passing.16×16. Chief Justice Marshall refers to “Goosely, in this court.” Burr, 25 F. Cas. at 39. As detailed at infra notes 202 to 208 and accompanying text, the case is United States v. Goosely, 25 F. Cas. 1363 (C.C.D. Va.) (No. 15,230). And John Marshall’s originalist bona fides are the subject of some controversy in the originalist literature.17×17. Compare Mike Rappaport, Chief Justice Marshall’s Textualist Originalism, Law & Liberty (Mar. 21, 2019), https://www.lawliberty.org/2019/03/21/chief-justice-marshalls-textualist-originalism [https://perma.cc/S3XT-53JN] (arguing that John Marshall “had a general approach, which was a form of originalist textualism”), with D. A. Jeremy Telman, John Marshall’s Constitution: Methodological Pluralism and Second-Order Ipse Dixit in Constitutional Adjudication, Lewis & Clark L. Rev. (forthcoming) (manuscript at 54), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3461874 [https://perma.cc/E94X-8B47] (arguing that Marshall was a constitutional eclectic). In an ideal world, we could learn not only how Marshall ruled, but also how the finest legal minds of the era might have framed the question. In a perfect world, we would want to know: What did lawyers of the Founding era think the privilege meant and how might it apply to a compelled decryption case?

Remarkably, it turns out that we can know this, too. Burr’s trial drew intense public scrutiny.18×18. See 1 David Robertson, Reports of the Trials of Colonel Aaron Burr, for Treason, and for a Misdemeanor, at preface (Philadelphia, Hopkins & Earle 1808). During the nineteenth century, Robertson’s report was regularly cited in federal courts, including by the U.S. Supreme Court, as “Burr’s Tr.” See, e.g., Luther v. Borden, 48 U.S. (7 How.) 1, 81 (1849). I adopt that citation practice in this Article. An attorney, David Robertson, attended most of the proceedings and wrote down in shorthand what was said.19×19. See 1 Burr’s Tr. at preface. The next year, Robertson published a two-volume report of the trial that included a “full and correct statement of all the testimony and documents.”20×20. Id. Critically for us, Robertson reported the legal arguments in glorious detail. He reprinted the “arguments of the counsel on all points of importance” in the proceeding “verbatim as uttered,”21×21. Id. One note of caution is that Robertson states that he did not record the early parts of the proceeding at the time and had to reconstruct them later: The proceedings previous to the trials, before and while the grand jury were in deliberation, are also detailed, but the first part of them not so fully as the rest of the report; because it was the middle of June, before the reporter was prevailed on to undertake the publication. He has however consulted the best sources of information, in order to enable him to present to the public a correct statement of those preliminary proceedings . . . . Id. The Burr proceedings on the Fifth Amendment occurred on June 16–18, 1807, see id. at 209–12, 227, 242, about the time that Robertson began his full report. The extraordinary detail of the material presented in Parts II and III suggests to me that Robertson’s verbatim reporting likely had begun by this time. including the citations to “the authorities referred to” by counsel.22×22. Id. at preface.

Robertson’s report is a remarkable resource for those seeking the original understanding of the privilege against self-incrimination. Chief Justice Marshall allowed the lawyers to debate the Fifth Amendment question at what one historian has called “tedious” length.23×23. R. Kent Newmyer, The Treason Trial of Aaron Burr 104 (2012). The result is an oral argument transcript that likely reflects hours of in-court advocacy solely about the privilege and its application to compelled decryption. It thoroughly details the arguments and counterarguments for how the relevant precedents and principles might apply. It also includes dozens of legal citations, all of which are to sources that can be found and read today to reconstruct the materials the lawyers and the court had at their disposal. And all of this is waiting to be explored for the first time. Although Robertson’s report is known to historians, its exhaustive arguments about the privilege against self-incrimination have not been closely examined either by historians or by lawyers.24×24. I suspect this is true for two reasons. First, the dispute has no importance for historians because Willie’s testimony ultimately went nowhere. Willie eventually testified that he did not know the cipher, and the issue never arose in the Burr trial again. See infra p. 945. Second, the Federal Cases reporter excludes the discussion, making the arguments hard for lawyers to access. The abbreviated summary of the argument begins with the argument that introduced the question, leading up to when Chief Justice Marshall indicates that he would like to hear additional argument. But then the reporter simply notes: “The point was argued at some length on the two following days by Mr. Botts, Mr. Williams, Mr. Martin, and Mr. Wickham on one side, and by Mr. MacRae and Mr. Hay on the other.” United States v. Burr (In re Willie), 25 F. Cas. 38, 39 (C.C.D. Va. 1807) (No. 14,692e).
This does not mean that the Chief Justice’s opinion in Burr has been ignored. To the contrary, the Supreme Court has frequently relied on it as an important Fifth Amendment precedent. See infra notes 344–48 and accompanying text. Further, some scholars (including myself) have noted in passing that Burr may be relevant for constitutional debates on decryption today. See, e.g., Orin S. Kerr, The Fourth Amendment in Cyberspace: Can Encryption Create a “Reasonable Expectation of Privacy?,” 33 Conn. L. Rev. 503, 528–29 (2001) (using Burr to argue that modern encryption does not present a “latent ambiguity” that could justify translation under the Fourth Amendment); Jeffrey Kiok, Missing the Metaphor: Compulsory Decryption and the Fifth Amendment, 24 B.U. Pub. Int. L.J. 53, 65 (2015) (noting Burr). But Marshall’s opinion in Burr is brief, and, viewed in isolation, cryptic. It is the discovery and exploration of Robertson’s report that now opens the door.

The usefulness of Robertson’s report is bolstered by the outstanding lawyering in the Burr case. The three prosecutors and four defense attorneys were among the finest advocates of their era.25×25. See Peter Charles Hoffer, The Treason Trials of Aaron Burr 147 (2008) (describing the defense team as arguably “the finest legal talent assembled in any trial in the history of the young nation”). Prosecutor William Wirt is still considered “one of the greatest Supreme Court advocates of all time.”26×26. Proceedings in Commemoration of the 200th Anniversary of the First Session of the Supreme Court of the United States, 493 U.S., at x (1990) [hereinafter Proceedings in Commemoration] (speech of Solicitor General Rex Lee). Wirt’s 170 Supreme Court arguments included “virtually all of the landmark cases of the first third of the nineteenth century.”27×27. Id. at xii (quoting Joseph C. Robert, The Hon. William Wirt: The Many-Sided Attorney General, Sup. Ct. Hist. Soc’y Y.B. 51, 52 (1976)); see, e.g., Daniel A. Farber, Endnote, The Story of McCulloch: Banking on National Power, 20 Const. Comment. 679, 692 (2003). Defense lawyer Edmund Randolph had served as the first Attorney General of the United States.28×28. Proceedings in Commemoration, supra note 27, at x. Randolph was also “the most active of [the Supreme] Court’s early practitioners.” Id. Defense lawyer Luther Martin had recently served twenty-eight years as Maryland’s Attorney General,29×29. See William Reynolds II, Essay, Luther Martin, Maryland and the Constitution, 47 Md. L. Rev. 291, 292 (1987). and he was considered “the undisputed head of the profession” in his home state.30×30. Samuel Tyler, Memoir of Roger Brooke Taney, LL.D. 65 (Baltimore, John Murphy & Co. 1872). Aaron Burr participated actively in his defense, and he was not just a brilliant lawyer but also the former Attorney General of New York.31×31. Burr had served as New York’s Attorney General from 1789 to 1791. See Nancy Isenberg, Fallen Founder: The Life of Aaron Burr 104–06 (2007).

Several advocates in Burr did not just know the new Constitution but had participated in its creation. Two had been delegates to the Constitutional Convention in Philadelphia.32×32. The two were Edmund Randolph and Luther Martin. See generally John J. Reardon, Edmund Randolph: A Biography 96–120 (1975); Reynolds, supra note 30, at 294–305. Edmund Randolph had helped secure Virginia’s ratification of the Constitution while serving as Governor of Virginia.33×33. See Reardon, supra note 33, at 137–50 (detailing Randolph’s role). He also had served with James Madison and John Marshall on the Virginia committee that proposed a Bill of Rights for the U.S. Constitution that included a privilege against self-incrimination.34×34. Jean Edward Smith, John Marshall: Definer of a Nation 142 (1996). The privilege against self-incrimination included in their proposal stated that “in all criminal and capital prosecutions, a man . . . [cannot] be compelled to give evidence against himself.” 3 The Debates in the Several State Conventions on the Adoption of the Federal Constitution 658 (Jonathan Elliot ed., 2d ed., Philadelphia, J.B. Lippincott 1888). In addition to Madison, Marshall, and Randolph, the committee members included Patrick Henry, George Mason, and James Monroe, among others. Id. at 656. Randolph had been in close contact with Madison as he introduced in Congress the constitutional amendments that became the Bill of Rights.35×35. For example, in Randolph’s letter to Madison upon Madison’s introducing the Bill of Rights, Randolph writes: “The amendments, proposed by you, are much approved by the strong fĹ“deralists here and at the Metropolis; being considered as an anodyne to the discontented.” “I am still in hopes to see reported from your mouth some review of the various amendments proposed, and reasons against the fitness of such, as appeared improper for adoption.” Letter from Edmund Randolph to James Madison (June 30, 1789), Nat’l Archives: Founders Online, https://founders.archives.gov/documents/Madison/01-12-02-0169 [https://perma.cc/9U4H-9AU8]. Defense lawyer Luther Martin had served as a delegate to the Constitutional Convention and was “one of the leaders of the Anti-Federalist revolt that led to the ratification of the Bill of Rights.”36×36. Roger Roots, The Rise and Fall of the American Jury, 8 Seton Hall Cir. Rev. 1, 14 (2011); see also Bill Kauffman, Forgotten Founder, Drunken Prophet: The Life of Luther Martin 36 (2008). See generally Reynolds, supra note 30. These were stellar advocates intimately familiar with the new Constitution that they asked Chief Justice Marshall to interpret.

This Article reconstructs the Burr proceedings to shed light on the original public meaning of the Fifth Amendment privilege and its application to compelled decryption. It explores the legal materials in Burr in depth to understand how the privilege against self-incrimination was conceived and interpreted shortly after the Fifth Amendment’s ratification in a case with striking similarity to modern-day facts. The rich materials on how the participants in a leading case understood the privilege provide unique insights that so far have gone unmined. Studying Burr answers open questions such as: What did lawyers with experience with the enactment of the Bill of Rights think the privilege meant? What sources did they examine to shed light on its meaning? How did the Chief Justice’s opinion fit with that understanding? And specifically, how did they think the privilege applied to efforts to decipher encrypted letters?

Understanding Burr illuminates the original understanding of the privilege in two distinct ways. First, the lawyers’ arguments can explain how the privilege was understood. The dispute over the Fifth Amendment was hotly contested and vigorously argued. The Founding-era experience of the lawyers, plus their celebrated legal talents, inspire confidence that they presented the best arguments for their sides based on the Framing-era understanding of the privilege. Understanding their common ground and the interpretive sources they relied on — all in a case specifically about access to encrypted documents — reveals a shared understanding of what the privilege meant. It’s not the same as having a time machine, going back to 1791, and asking how the privilege should apply to unlocking an iPhone. But it’s a lot closer than we would normally expect to get.

Second, Chief Justice Marshall’s opinion can also yield originalist insights. The advocates in Burr could frame but not resolve the debate. In contrast, Chief Justice Marshall’s opinion in Burr grapples thoughtfully and carefully with the issues. Granted, Marshall’s specific reasoning in 1807 did not necessarily reflect settled understandings from 1791. The 1791 materials may not establish the precise answers that Marshall needed to provide. Nonetheless, Marshall’s well-reasoned decision working with the arguments of counsel may be the closest we can come to understanding how the privilege would have been understood to apply to compelled decryption.37×37. See William Baude, Constitutional Liquidation, 71 Stan. L. Rev. 1, 61–62 (2019) (discussing the originalist value of post-ratification understandings).

The Article offers three conclusions about how the original understanding of the Fifth Amendment applies to modern compelled decryption. First, the privilege ordinarily should not bar requiring a person to disclose a password to enable a search for digital evidence. Doing so will not normally be incriminating in the sense recognized in Burr.38×38. See infra section V.B, pp. 952–57. Second, the opposite is typically true when the government searches for contraband. Password disclosure often will be incriminating in a contraband search case because it admits control of the encrypted file.39×39. See infra section V.C, pp. 957–60. Finally, whether the privilege bars compelled entry of the password (as opposed to its disclosure) depends on a choice of analogy. If compelled entry is analogized to admitting knowledge of the password, the rules on compelled entry should mirror those for compelled disclosure. On the other hand, if compelled entry is analogized to compelled production, the Fifth Amendment may bar the act entirely.40×40. See infra section V.D, pp. 960–62.

The Article has five parts. Part I presents the facts and procedural history of the Fifth Amendment question in Burr. It starts by detailing the criminal charges considered against Aaron Burr. It then explores what we know of the encrypted letter and focuses on the government’s strategy of authenticating and decrypting the letter using the testimony of Burr’s secretary Charles Willie and his co-conspirator Erick Bollman. It concludes by discussing Willie’s appearance in court, including the specific questions the government asked and how Willie asserted the privilege.

Part II details the legal authorities relied on by the lawyers in Burr to interpret the privilege against self-incrimination. Willie’s assertion of privilege led to extensive argument. This section shows that all of the lawyers in Burr understood the Fifth Amendment privilege as the common law privilege from England. The lawyers all relied on the same basic sources to understand the privilege: an English evidence treatise, a handful of English cases primarily from the State Trials report, and American cases from 1799 and 1806 that had already interpreted the privilege.

Part III presents the arguments of counsel based on the authorities of Part II. It explains that the lawyers in Burr diverged on three questions. First, who determines when a valid privilege exists, the witness or the court? Second, how much evidence of a crime must an answer reveal for it to count as incriminating? And third, how did the test apply to the questions posed to Burr’s secretary? This section details the arguments of both sides on all three questions.

Part IV details Chief Justice Marshall’s opinion. It explains how he picked among the arguments of counsel, ultimately siding with the defense on the second question but with the government on the first and third questions. The section also uncovers a postscript to Marshall’s ruling, when he briefly considered how the privilege might apply to compelling a different letter directly from Burr.

Part V draws originalist lessons from Burr for modern facts of compelled disclosure or entry of a password. Extrapolating from the Burr arguments and ruling, it concludes that compelled disclosure of a password should ordinarily be permitted when the government seeks evidence but not contraband. The rules for compelled entry of a password should depend on whether the facts are analogized to compelled disclosure of the password or compelled production of the files.


* Professor, University of California, Berkeley School of Law. Thanks to Will Baude, Kellen Funk, Sarah Seo, John Fabian Witt, Andrew Guthrie Ferguson, Kevin Cole, Sara Mayeux, Gerard Magliocca, Kevin Arlyck, Steven Bellovin, Larry Solum, and Doug Linder for helpful discussions and comments. I am indebted to Edna L. Lewis of the UC Berkeley School of Law Library for outstanding help finding original sources.