Introduction

Can a president issue a pardon without telling anyone but the recipient that she has issued it? Yes, the president can grant a valid pardon without telling anyone but the recipient of her grace that she has done so. While a defendant must plead a pardon for a court to take notice of it and quash an indictment, the document may otherwise lay buried in a sock drawer in case it is ever needed without losing any of its force or effect.

Lawyers working in real property will sense that cached pardons savor of an obscure but nonetheless viable estate planning device: the “dresser drawer deed.”¹Brody Swanson, Allowing Farmers to “Take Back” What’s Theirs: Adoption of the Revocable Life Estate Deed, 21 Drake J. Agric. L. 409, 417 (2016). Despite the advent of modern recording statutes, deeds of real estate are valid upon receipt and do not need to be recorded or otherwise publicized to be effective.²See 4 Herbert Thorndike Tiffany, Real Property § 1033 (3d ed. 1975) (“Conveyances of land, including leases, contracts under seal, mortgages of land and of chattels, deeds of gift, insurance policies, and promissory notes, take effect by delivery.”); 5 id. § 1262 (“The requirement of record has almost invariably been regarded as intended for the protection of subsequent purchasers only, so that a failure to record the instrument in no way affects the passing of title as between the parties thereto.”). For example, suppose Brigitte wishes to placate her second husband Emmanuel while securing some of her separate property for her son Jean. Brigitte can please Emmanuel by letting him hold a will giving all of her property to him. At the same time, she can give Jean a deed to Blackacre and instruct him not to record it but keep it in a safe deposit box until she dies. When Brigitte gives up the ghost, Jean can walk to the courthouse and record the deed. The property will have passed long before Brigitte and will not move through probate. Brigitte has managed to save a piece of her wealth for Jean, all while maintaining family peace until such time as Emmanuel can no longer divorce or vex her.

Upon comparison with the dresser drawer deed, cunning minds should have little difficulty identifying strategic uses for a dresser drawer pardon. For example, the choice to pardon can make a president unpopular with the public.³See, e.g., Gabriel Sherman, And the Band Played On, Vanity Fair, Holiday 2020/2021, at 69. (describing the fall in President Clinton’s approval rating following his pardon of fugitive financier Marc Rich). If a friend, family member, ally, or donor is not now under indictment but at risk of future prosecution for a past crime, any of Warren Harding’s esteemed successors can write the favored one an insurance policy against prosecution for all crimes antedating the pardon. If the pardon never has to be pleaded and made known to the public, then so much the better for the reputations of the pardoner and pardoned.

This article’s argument for the validity of secret pardons focuses on Supreme Court precedents, especially United States v. Wilson,⁴32 U.S. (7 Pet.) 150 (1833). in which Chief Justice John Marshall expressly compares a pardon to other private acts, including conveyances of real estate.⁵Id. at 160–61. This analogy and the Supreme Court opinions endorsing it are the driving wheel of this argument. By and large, this article does not argue that it is desirable for the president to be able to write a “sugar bowl,” “back pocket” pardon. Indeed, a president should almost always opt to publicize the pardons and commutations she grants in the interest of transparency and accountability. However, secrecy and discretion can be important tools of statecraft, and the occasional need for secrecy in times of crisis was on the Founders’ minds when they wrote the pardons clause.

Part I reviews the history, caselaw, and regulations relevant to the federal clemency power. Part II next shows that, because pardons are private acts akin to deeds, a pardon need not be made public to be effective: it only needs to be delivered to and accepted by the recipient. Given that pardons are private acts, we can also preliminarily address two further issues. First, a president most likely cannot issue an oral pardon—proved when needed in court by witness testimony or audiovisual recording—and second, a president certainly cannot pardon herself because one cannot deed property to oneself. Part III then addresses counterarguments—including the Holmesian view of pardons and the historical practice of presidential pardons in this country.

I.      Background

A.      Origins of the Federal Pardon Power

The power of the English king to pardon crimes was nearly absolute at the time of the American Founding because the whole system of justice was supposed to be an effluence of the monarchy.⁶See William F. Duker, The President’s Power to Pardon: A Constitutional History, 18 Wm. & Mary L. Rev. 475, 487 (1977). As late as 1686, the Lord Chief Justice of the King’s Bench could explain that because “the laws [are] the King’s laws . . . the King [has] a power to dispense with any of the laws of Government as he [sees] necessary.”⁷Id. at 495 (alterations in original) (quoting Godden v. Hales, 89 Eng. Rep. 1050, 1051 (KB); 2 Show. 475, 478). His Lordship perhaps called too much attention to the king’s authority: in 1688, Parliament put through a bill that clarified that the king had no power to suspend the laws (as distinct from remitting punishment for their violation).⁸Id. at 496 (citing The Bill of Rights, 1 W. & M. c. 36 (1688)). And in 1700, Parliament blocked the pardoned from pleading their charters of clemency in cases of impeachment before the House of Commons.⁹Id. at 496 (citing The Act of Settlement, 12 & 13 Will. 3 c. 2 (1700)).

The Constitution explicitly incorporates the latter limitation on the president’s pardon power: “[H]e shall have Power to grant Reprieves and Pardons for Offences against the United States, except in Cases of Impeachment.”¹⁰U.S. Const. art. II, § 2, cl. 1. At the Philadelphia Convention where the Constitution was drafted, neither the Virginia nor the New Jersey plans called for a pardon power, but delegates like Alexander Hamilton successfully urged its adoption.¹¹Duker, supra note 6, at 501.

Roger Sherman sought to make pardons contingent until the next meeting of the Senate, which would have to ratify the president’s dispensation or else the reprieve would lapse.¹²Id. When called to defend the president’s unilateral power of grace, advocates for the text’s final form stressed the need for flexibility to temper the stringency of the law and the utility of the pardon power for a president seeking to quell insurrection.¹³See id. at 502–05. Speaking in favor of ratification, Hamilton said, “[I]n seasons of insurrection and rebellion there are often critical moments when a well-timed offer of pardon to the insurgents or rebels may restore the tranquility of the commonwealth . . . .”¹⁴Id. at 505 (quoting The Federalist No. 74 (Alexander Hamilton)). Addressing the same point, James Iredell laid special emphasis on the need for “secrecy”.¹⁵Id. at 503 (quoting James Iredell, Observations on George Mason’s Objections to the Federal Constitution, reprinted in Pamphlets on the Constitution of the United States 351–52 (Paul Leicester Ford ed., 1968)). In the ratification debates, he said, “One of the great advantages attending a single Executive power is the degree of secrecy and dispatch with which on critical occasions such a power can act.”¹⁶Id. (quoting Iredell, supra note 15, at 351–52).

Whereas the former defense of the pardon power speaks to concerns internal to criminal justice—the need to balance mercy with justice¹⁷John Tasioulas, Mercy, 103 Proc. Aristotelian Soc’y 101, 101–02 (2003) (“[W]e may characterize mercy as the putative ethical value that justifies leniency in the infliction of punishment that is due in accordance with justice.”). or achieve a fuller justice through mercy¹⁸E.g., Claudia Card, On Mercy, 81 Phil. Rev. 182, 191 (1972) (“When we temper (institutional) justice with mercy in deciding how to treat the offender, we consider not only facts about his offense but also facts about his character and suffering which may not be revealed simply by looking at his offense.”).—the latter apology proves that the pardon power was provided for with its political utility in mind. Consequently, respect for the founding era entails that the pardon power must be viewed through at least two lenses—one of criminal justice ethics and the other of statecraft.

B.      Cases

Chief Justice Marshall authored a broad review of the pardon power in the curious case of George Wilson, a man sentenced to death for robbing the mail who would not take advantage of a pardon written for him by President Jackson.¹⁹United States v. Wilson, 32 U.S. (7 Pet.) 150, 153–54 (1833). Marshall encapsulated the legal issue as whether the pardon “could avail, without being pleaded, or in any manner relied on by the prisoner.”²⁰Id. at 160. The opinion is modern in its self-aware, expository style; Marshall begins by declaring that the rules and principles controlling the constitutional pardon power should be taken from English sources.²¹See id. at 161–63. He writes, “[W]e adopt their principles respecting the operation and effect of a pardon, and look into their books for the rules prescribing the manner in which it is to be used by the person who would avail himself of it.”²²Id. at 160. The Court made the same commitment to following English tradition in this area of constitutional law a few decades later in Ex parte Wells.²³59 U.S. (18 How.) 307 (1855). The Wells Court wrote, “At that time both Englishmen and Americans attached the same meaning to the word pardon. In the convention which framed the constitution, no effort was made to define or change its meaning, although it was limited in cases of impeachment.” Id. at 311.

The English sources taught Marshall that a “pardon is an act of grace . . . . [T]he private, though official act of the executive magistrate, delivered to the individual for whose benefit it is intended, and not communicated officially to the court.”²⁴Wilson, 32 U.S. (7 Pet.) at 160–61. He stresses that a pardon is a “private deed” and that therefore a court knows nothing of it unless it is brought to the court’s attention by a party.²⁵Id. at 161.

There is nothing in the nature of a pardon to distinguish it from other private legal acts, explains Marshall.²⁶Id. Just like a deed of real estate, the validity of a pardon depends upon acceptance by the grantee.²⁷Id. Indeed, this understates the likeness, for Marshall places property deeds and pardons in the same genus:

A pardon is a deed, to the validity of which delivery is essential, and delivery is not complete without acceptance. It may then be rejected by the person to whom it is tendered, and if it be rejected, we have discovered no power in a court to force it on him.²⁸Id.

To support this characterization of pardons, Marshall calls up William Hawkins, Sir William Blackstone, and sundry other worthies.²⁹See id. at 161–63. Quoting Hawkins’ influential Treatise of Pleas of the Crown, Marshall copied out, “[H]e who pleads such a pardon must produce it . . . though it be a plea in bar, because it is presumed to be in his custody, and the property of it belongs to him.”³⁰Wilson, 32 U.S. (7 Pet.) at 162 (emphasis added) (quoting 2 William Hawkins, A Treatise of the Pleas of the Crown: Or A System of the Principal Matters Relating to That Subject, Digested Under Their Proper Heads 397 (1721)). Note that Hawkins’ quote not only directly supports Marshall’s holding but also that Hawkins’ choice to describe pardons as property bolsters the case for grouping pardons with deeds.

Marshall likewise relies on Blackstone, who describes a pardon as something held or kept in hand by a defendant that he can choose whether to bring to the court’s attention: “The king’s charter of pardon must be specially pleaded; and that at a proper time, for if a man is indicted and has a pardon in his pocket, and afterwards puts himself upon his trial by pleading the general issue, he has waived the benefit of such pardon.”³¹Id. at 163 (emphasis added) (quoting 4 William Blackstone, Commentaries *402).

During Reconstruction, the Court had to rule on the pardon power in the weightier and far more inflammatory conflict between President Johnson and Congress over how to treat ex-Confederate leaders. In Ex parte Garland,³²71 U.S. (4 Wall.) 333 (1866). the Court decided that President Johnson’s pardon relieved Augustus Hill Garland of all the disabilities Congress had imposed upon him for betraying his oath to the Constitution.³³Id. at 381. Proclaiming the president’s unilateral constitutional authority on this turf, Justice Stephen Field wrote, “This power of the President is not subject to legislative control.”³⁴Id. at 380; cf. Schick v. Reed, 419 U.S. 256, 266 (1974) (“[T]he power flows from the Constitution alone, not from any legislative enactments, and that it cannot be modified, abridged, or diminished by the Congress.”).

Dissenting in Wells, Justice John McLean sought unsuccessfully to derail this train of reasoning from the power of the English monarch. He eloquently objected:

Instead of looking into the nature of our government, for the true meaning of terms vesting powers in the executive, are we to be instructed by studying the regalia of the crown of England; not to ascertain the definition of the word pardon, but to be assured what powers are exercised under it by the monarch of England. This is a new rule of construction of the constitutional powers of the President. I had thought he was the mere instrument of the law, and that the flowers of the crown of England did not ornament his brow.³⁵Ex parte Wells, 59 U.S. (18 How.) 307, 321 (1855) (McLean, J., dissenting).

The fundamental difference between republics and monarchies is a chief theme of McLean’s dissent. In a line quoted by McLean,³⁶Id. Justice Joseph Story echoes his skepticism about the compatibility of the prerogatives of monarchy with the proper function of the chief executive office of a republic: “The whole structure of our government is so entirely different, and the elements, of which it is composed, are so dissimilar from that of England, that no argument can be drawn from the practice of the latter, to assist us in a just arrangement of the executive authority.”³⁷2 Joseph Story, Commentaries on the Constitution of the United States 346 (1833). Yet, as rhetorically powerful as these appeals to the difference between a republic and monarchy must have been to patriotic American ears, they did not alter the antiquarian, Anglophilic track laid by the Court majorities in Wilson and Wells.

In fact, three twentieth-century cases roundly affirmed the reasoning and holdings of Chief Justice Marshall’s opinion in Wilson: Burdick v. United States,³⁸236 U.S. 79 (1915). Ex parte Grossman,³⁹267 U.S. 87 (1925). and Schick v. Reed.⁴⁰419 U.S. 256 (1974). In 1915, Justice Joseph McKenna wrote for the Court in Burdick: “The principles declared in Wilson v. United States [sic] have endured for years; no case has reversed or modified them.”⁴¹Burdick, 236 U.S. at 91. Ten years later, Chief Justice William Taft (curiously, the only justice of the court to have also been a national chief executive)⁴²Robert C. Post, Mr. Taft Becomes Chief Justice, 76 U. Cin. L. Rev. 761, 761 (2008). reiterated in Grossman that common law and “British institutions” were the best guide to interpreting the constitutional pardon power, quoting Wilson and Wells both to prove it.⁴³Grossman, 267 U.S. at 108–10. Skipping ahead to the 1970s, Chief Justice Warren Burger similarly wrote, “The history of our executive pardoning power reveals a consistent pattern of adherence to the English common-law practice.”⁴⁴Schick, 419 U.S. at 262. He too treated Wilson and Wells as keystone authorities and underlined how the Court had consistently recurred to them in pardons cases.⁴⁵See id. at 266.

Of the three twentieth-century cases which paid homage to Wilson and Wells, Burdick is the first and most relevant to our purposes. In Burdick, a grand jury investigating customs fraud sought information from a New York Tribune editor, George Burdick, whose paper had published articles on the customs frauds under inquiry.⁴⁶Burdick, 236 U.S at 85. Burdick refused to answer some questions about the articles, invoking his Fifth Amendment right not to incriminate himself.⁴⁷Id. President Wilson issued Burdick a pardon for any crimes he may have committed in relation to writing and publishing the articles, adding for good measure a pardon “in connection with any other article, matter or thing, concerning which he may be interrogated in the said grand jury proceeding, thereby absolving him from the consequences of every such criminal act.”⁴⁸Id. at 85–86. Burdick declined to accept the pardon and continued to refuse to answer the grand jury’s questions.⁴⁹Id. at 86. The Court quickly recognized that Wilson controlled the issue (thwarting its namesake’s scheme with federal prosecutors) and proceeded to teach its principles.⁵⁰See id. at 88. The Burdick opinion carries forward wholesale the conceit from Wilson that pardons are private acts, akin to deeds, and must be both delivered and accepted to be effective.⁵¹Id. at 90–91.

One case, Biddle v. Perovich,⁵²274 U.S. 480 (1927). stands aloof from the chorus of praise for Chief Justice Marshall and his work in Wilson.⁵³See G. Sidney Buchanan, The Nature of a Pardon Under the United States Constitution, 39 Ohio St. L.J. 36, 36 (1978) (contrasting Holmes’s view of the pardon power with Marshall’s). Unsurprisingly, the case flowed from the pen of Justice Oliver Wendell Holmes Jr., whose caustic intellect was to common-law lore what aqua regia is to golden idols.

Perovich, the petitioner, was convicted of murder in Alaska and sentenced to hang.⁵⁴Biddle, 274 U.S. at 485. Next, President Taft commuted his sentence to life imprisonment in a federal penitentiary.⁵⁵Id. In accordance with President Taft’s will, the authorities moved Perovich from an Alaskan jail to a penitentiary in Washington State, and later, from there to the federal penitentiary in Leavenworth, Kansas.⁵⁶Id. From his Kansas cell, he applied for a writ of habeas corpus, arguing that his transfer to prison from jail in Alaska was unlawful because he had not consented to Taft’s commutation.⁵⁷Id.

As we have seen, Burdick and Wilson established that a pardon must be accepted to be effective. Perovich thus relied on these authorities to argue that the commutation was invalid as he had never accepted it.⁵⁸See id. at 486; see also id. at 481–82 (Argument for Perovich). The Solicitor General argued that Burdick was wrongly decided and that in any event, the English authorities established that a pardoned man will not be hanged even if he waives the pardon.⁵⁹See id. at 486.

Although Holmes could have relied on the historical authorities excepting execution of capital sentences from the rule that a defendant may refuse a pardon,⁶⁰The Solicitor General relied on a very old English case that Marshall also cited in Wilson, which states, “[I]f the King pardons a felon, and it is shewn to the court; and yet the felon pleads not guilty, and waives the pardon, he shall not be hanged; for it is the King’s will that he shall not; and the King has an interest in the life of his subject.” See Biddle v. Perovich, 274 U.S. 485, 486 (quoting Case LXII 145 Eng. Rep. 90; Jenk. 129); see also United States v. Wilson, 32 U.S. 150, 162 (1833). he did not.⁶¹See Biddle, 274 U.S. at 486. The man who called it “revolting to have no better reason for a rule of law than that so it was laid down in the time of Henry IV,”⁶²Oliver Wendell Holmes, The Path of the Law, 10 Harv. L. Rev. 457, 469 (1897). was true to his principles. For the Court, he writes, “We will not go into history, but we will say a word about the principles of pardons in the law of the United States.”⁶³Biddle, 274 U.S. at 486. Rather than simply cite the Medieval reports that announced an exception in capital cases to the principle that a pardon must be accepted to be effective, Holmes opted to present what he saw as the true rhyme and reason of the pardon power in a modern nation. The pardon power, he explains, “is a part of the Constitutional scheme,” and when invoked “it is the determination of the ultimate authority that the public welfare will be better served by inflicting less than what the judgment fixed.”⁶⁴Id. (emphasis added).

To Holmes, a pardon was far from a matter of private right: it was a matter of what the head of state thought was in the public interest. “Just as the original punishment would be imposed without regard to the prisoner’s consent and in the teeth of his will, whether he liked it or not, the public welfare, not his consent, determines what shall be done.”⁶⁵Id. (emphasis added). Besides stressing that a pardon is a matter of public policy, Holmes also points up the incongruity of a person forcing the hand of the authorities to punish him: “Supposing that Perovich did not accept the change, he could not have got himself hanged against the Executive order.”⁶⁶Id. at 487. After all, the same officers who would hang the prisoner are officers of the federal government, the head of which expressed his wish that the prisoner not be hanged.

While breaking with the earlier pardons cases’ focus on history, Holmes does not purport to overrule them. Rather, he concludes the opinion by holding that Burdick’s reasoning is “not to be extended to the present case.”⁶⁷Id. at 488.

1.      Making Sense of Refusing Pardons

When it discussed the pardons clause, the Supreme Court repeatedly asked whether a defendant had to accept a pardon in order for it to be valid. Thus, this article frequently relies on cases about defendants who did not invoke or accept a pardon. The defendant who refuses to accept a pardon seems like a strange, counterintuitive example—like a contrived, lawyerly case barely less implausible at first blush than the fertile octogenarian. By this logic, lavishing attention on these cases risks missing the main functions of pardons under the Constitution. But a few reasons might explain why a person would refuse a pardon and why a plausible lawmaker might allow him or her to do so.

One readily imagines a person who refuses to accept a pardon because she does not wish to avow, expressly or implicitly, that she committed a crime. However, this reason for the acceptance rule depends upon the existence of the rule, for an implied admission of guilt can only attach to a pardon so long as the defendant can accept or reject it.⁶⁸See Buchanan, supra note 53, at 44 (“A pardon, however, carries an imputation of guilt only if the validity of the pardon depends upon its acceptance by the pardonee. If the requirement of acceptance is removed, the imputation of guilt vanishes. It becomes circular, therefore, to use the imputation of guilt argument as a justification for the requirement of acceptance.”).

On the other hand, there is an allied but distinct reason why a person might want the ability to refuse a pardon and resent that the option is denied to her. To wit, it is possible she does not wish to acknowledge the moral or political legitimacy of the president who pardons her. Even if she has already been convicted of a crime and is serving a sentence of confinement, she might prefer to reject an offer of clemency in order to refuse to recognize the authority of the pardoner. Relatedly, the president could be targeting the pardonee for clemency as a way of alienating her from her supporters or keeping her from drawing sympathy to her cause as a martyr.

Additionally, there is a perfectly sensible motive, prior to conviction, for a defendant to want the ability to refuse a pardon: she wishes to retain the opportunity to plead her case in court, whether to show her factual innocence or make legal and moral arguments against the law she is charged with violating. In truth, the president controls the executive branch and could in theory stop the prosecution by either directing the prosecuting attorneys to drop charges or firing them if they refused.⁶⁹See U.S. Const. art. II §§ 1–2; Steven G. Calabresi & Christopher S. Yoo, The Unitary Executive: Presidential Power from Washington to Bush 4 (2008) (supporting the president’s full power to order about the members of the executive branch). But see Bruce A. Green & Rebecca Roiphe, Can the President Control the Department of Justice?, 70 Ala. L. Rev. 1, 2 (2018) (arguing that “the Department of Justice is independent of the President, and its decisions in individual cases and investigations are largely immune from his interference or direction”). That said, in practice and by custom, the president can more easily pardon someone than he can steer the bureaucracy at the Department of Justice.⁷⁰See Green & Roiphe, supra note 69, at 32–33 (“[T]he tradition of prosecutorial independence has grown into a permanent feature of American government.”); Harold H. Bruff, Presidential Power Meets Bureaucratic Expertise, 12 U. Pa. J. Const. L. 461, 480 (2010) (“In modern times, every President still struggles mightily to control his own immediate subordinates, to say nothing about the vast and remote bureaucracy.”). Moreover, a pardon is permanent, while a decision to drop charges is potentially reversible by a future administration.

There are also less egoistic motives to refuse a pardon. A defendant could choose to remain in prison out of repentance, remorse, or sympathy for the victim. More selfishly, there is a conceivable defendant who wishes to be seen as repentant, remorseful, or sympathetic. This hypothetical defendant—for example, a politician or popular entertainer—wants to be forgiven by the public and return to their favor by completing his sentence in full.

Perhaps the clearest and least psychologically exotic reason to refuse a pardon or commutation appears when we consider conditional pardons and commutations. The punishment substituted as a condition of the pardon could be more hurtful than the original punishment.⁷¹See generally Harold J. Krent, Conditioning the President’s Conditional Pardon Power, 89 Cal. L. Rev. 1665, 1682 (2001) (“Society should permit individuals subjected to punishment to determine whether to accept ‘a new deal’ predicated on different conditions . . . . [E]ven if a proposed conditional commutation objectively appears to be more favorable, the offender’s subjective evaluation of the options should prevail.”). A defendant could receive a shorter sentence but be forced to serve that sentence in a high security prison—spending most of her days in solitary confinement. A defendant might be forced to pay a fine that would financially ruin her, when she would prefer to serve out her sentence and return to the free world with her property intact. Perhaps most insidiously, the president may abuse conditional pardons to silence her foes by replacing confinement with civil disabilities like disqualification from holding private or public office.⁷²Cf. Hoffa v. Saxbe, 378 F. Supp. 1221, 1224 (D.D.C. 1974) (quoting President Nixon’s pardon of legendary Teamsters leader Jimmy Hoffa “upon the condition that the said James R. Hoffa not engage in direct or indirect management of any labor organization” for almost ten years). The courts have held that the president possesses broad discretion to attach conditions to grants of clemency—even allowing him or her to substitute punishments not otherwise authorized by law.⁷³See Schick v. Reed, 419 U.S. 256, 266 (1974) (“[C]onsiderations of public policy and humanitarian impulses support an interpretation of that power so as to permit the attachment of any condition which does not otherwise offend the Constitution.”). But see Hoffa, 378 F. Supp. at 1236 (“[W]e find that if conditions are to be attached, they must relate to the reason for the initial judgment of conviction, because it is the crime and the circumstances surrounding it that give rise to the public’s interest in regulating and circumscribing the future behavior of the offender.”). Though only an opinion of the District Court for the District of Columbia, the Hoffa opinion is learned and well-reasoned, and the Court or the appellate courts could well adopt it in a future pardons case. As such, the ability to refuse the grant seems almost a necessary corollary to the ability to attach conditions to pardons.⁷⁴Patrick R. Cowlishaw, writing as a student at Stanford Law School, recognized this point early on when he referred to the right to reject a pardon as the “Traditional and Necessary Protection” against abusive conditions on pardons. Patrick R. Cowlishaw, Note, The Conditional Presidential Pardon, 28 Stan. L. Rev. 149, 169–70 (1975).

These reasons could motivate a defendant to refuse a pardon. Yet, with the interest of the state and society in mind instead, why would a rational lawgiver allow a defendant to refuse a pardon? First, constitutions do not just empower state actors to do what they think best but rather place limitations to safeguard individuals. A wise lawmaker designing a pardons clause could respect individual consciences and interests by considering the motivations for refusing a pardon. Furthermore, society is sometimes well served by the work of dissenters, whose cause may actually represent the best interests of the nation and constitutional order.

Second, the acceptance rule matches the power to grant conditional pardons and commutations with the defendant’s power to refuse the pardon or commutation. Keeping both powers in play is significant for partially protecting⁷⁵The protection is not perfect because the president can still shift the punishment as to kind or quality of punishment so long as the defendant does not object. Cowlishaw illustrates the kind of wholesale usurpation of the legislative role that is possible even with the acceptance requirement in a telling hypothetical. Id. at 157 (“Suppose Congress rejected an administration bill providing public service as the punishment for most federal offenses, reaffirming its belief that imprisonment was more appropriate. Subsequently the President pardoned all offenders who would have been reached by the bill on the condition of serving a term of public service. The usurpation of the congressional power to define and fix punishments is obvious.”). the prerogative of the legislature and judiciary to define the range of acceptable punishments and set punishments within that range.⁷⁶See Leo M. Romero, Punitive Damages, Criminal Punishment, and Proportionality: The Importance of Legislative Limits, 41 Conn. L. Rev. 109, 115 (2008) (“Legal punishment is the prerogative of the state, and punishment, to be legitimate in a democratic society, must be authorized and limited by the state in the form of legislative enactment. The legislature, representing society’s judgments, must both define the conduct that deserves punishment and determine the limits of that punishment.”); Ex parte United States, 242 U.S. 27, 41–42 (1916) (“[U]nder our constitutional system the right to try offences against the criminal laws and upon conviction to impose the punishment provided by law is judicial . . . .”). Justice McLean perfectly articulated the threat that conditional pardons pose to the separation of powers and the rule of law in his dissenting opinion in Ex parte Wells:

The power of commutation overrides the law and the judgments of courts. It substitutes a new, and, it may be, an undefined punishment for that which the law prescribes a specific penalty. It is, in fact, a suspension of the law, and substituting some other punishment which, to the executive, may seem to be more reasonable and proper.⁷⁷Ex parte Wells, 59 U.S. (18 How.) 307, 319 (1855) (McLean, J., dissenting).

Nonetheless, the acceptance rule should quell these concerns. If a defendant can refuse a pardon, the alternative punishment or conditions devised by the president and accepted by a prudent defendant are likely to be no worse than those judicially imposed.⁷⁸But see id. (“It is true the substituted punishment must be assented to by the convict; but the exercise of his judgment, under the circumstances, may be a very inadequate protection for his rights.”). The acceptance rule imperfectly guards the role of the legislature and the rule of law in punishments because no defendant can be compelled to forego her legally preauthorized punishment for an alternative one.⁷⁹Commutations of death sentences are an exception to the acceptance requirement, Biddle v. Perovich, 274 U.S. 480, 487 (1927), but execution is a sui generis harm that ranks near the top of earthly harms one can suffer. See Cowlishaw, supra note 74, at 168 (explaining that the comparative severity of punishments is easy enough to judge when death is at issue because “it is generally, if not universally, conceded that a pardon on any conditions is less severe than capital punishment”). By the same token, the acceptance requirement partly shields the right to have one’s sentence decided by a court against a scenario in which judicial sentences are routinely commuted for some other punishment of the executive’s devising.⁸⁰Compare the English practice of routinely commuting the sentences of convicted defenders “on condition that the felon be transported to another place.” Schick v. Reed, 419 U.S. 256, 261 (1974).

If the right to refuse a grant of clemency is cast aside, conditional pardons and commutations would become another bramble bush for the courts.⁸¹See Cowlishaw, supra note 74, at 168 (explaining how appeals to intuition and commonsense to judge the relative severity of punishments become “dangerously, perhaps unfathomably, murky when applied to a noncapital crime”). Specifically, the courts would need to decide which alternative punishments are acceptable and when one punishment is worse than another.⁸²This is a consequence of the fact that the law does not permit the executive to unilaterally increase the severity of punishment. See Schick, 419 U.S. at 267 (“Of course, the President may not aggravate punishment; the sentence imposed by statute is therefore relevant to a limited extent.”). The answers to these questions are fraught with imprecision and subjectivity. Ultimately, the courts could feel compelled to abandon the rule that a substitute punishment need not be part of the statutory range for the president to impose it as a condition of clemency. Under current law, these concerns are largely obviated by the defendant’s ability (outside of death sentences) to decide for himself whether the conditions of clemency are worse than his judicial punishment.

C.      Regulations and Statutes

The Department of Justice has issued regulations setting procedures for receiving and considering pardon applications.⁸³28 C.F.R. §§ 1.1–.11 (2022). The regulations are expressly advisory, “create no enforceable rights in persons applying for executive clemency,” and do not “restrict the authority granted to the President under Article II, section 2 of the Constitution.”⁸⁴Id. § 1.11. Even if they lacked this disclaimer, the courts would almost certainly reject any attempt to use them to limit the president’s exercise of constitutional authority.⁸⁵See Schick, 419 U.S. at 267 (“We therefore hold that the pardoning power is an enumerated power of the Constitution and that its limitations, if any, must be found in the Constitution itself.”).

Those who want a pardon are directed to submit their forms to the Department’s Pardon Attorney.⁸⁶28 C.F.R. § 1.1. The regulations next call for the attorney general to implement an investigation of the request’s merits and to notify any victims of the request.⁸⁷Id. § 1.6. At the end of the process, the Attorney General produces a report with a recommendation and forwards it to the president.⁸⁸Id. § 1.6(c). If the president decides to grant clemency, “the petitioner or his or her attorney shall be notified of such action and the warrant of pardon shall be mailed to the petitioner.”⁸⁹Id. § 1.7. Notably, the pardon regulations do not call for any special recording process or ceremony to be followed in issuing a pardon.⁹⁰See id. § 1.1–.11. For example, they do not call for the president to register a pardon with the National Archives, inform the Bureau of Prisons, affix the Great Seal of the United States, or ask the Secretary of State to do the same. Rather, in keeping with Wilson,⁹¹See United States v. Wilson, 32 U.S. (7 Pet.) 150, 161 (1833). they only require that the pardon be delivered to the pardonee.⁹²See 28 C.F.R. § 1.7. A president who chose to hew to these rules, however, would bring the pendency of the pardon application within the knowledge of employees at the Department of Justice and of the pardonee’s victims, if any.⁹³See id. § 1.6.

The Presidential Records Act⁹⁴44 U.S.C. §§ 2201–07. The president is notably exempt from the Freedom of Information Act, 5 U.S.C. § 552, because the Office of the President is not an “agency” to which that Act applies. Kissinger v. Reps. Comm. for Freedom of the Press, 445 U.S. 136, 156 (1980). I am grateful to Albert Alschuler for prompting me to consider the Presidential Records Act in this piece. requires the president to keep a record of the pardons she has issued, but the Act carries no penalties, and noncompliance would certainly not invalidate a pardon. The Act provides that the “United States shall reserve and retain complete ownership, possession, and control of Presidential records; and such records shall be administered in accordance with the provisions of this chapter.”⁹⁵44 U.S.C. § 2202.

The law defines Presidential records to include “documentary materials . . . created or received by the President,” and “documentary materials” includes all “correspondence, memoranda, documents, [and] papers.”⁹⁶Id. § 2201(1)–(2). The Act then compels the president to “take all such steps as may be necessary to assure that the activities, deliberations, decisions, and policies that reflect the performance of the President’s constitutional, statutory, or other official or ceremonial duties are adequately documented and that such records are preserved and maintained as Presidential records.”⁹⁷Id. § 2203(a).

But for all the reams of paper, reels of tape, and racks of hard drives the Act has surely compelled luckless White House staff to assemble, there are no actual penalties for violating it, and there is no suggestion that official business that goes undocumented becomes void or invalid.⁹⁸See Jessica L. Roberts, Note, #280 Characters of Legal Trouble: Trump, Twitter, and the Presidential Records Act, 2019 U. Ill. J.L. Tech. & Pol’y, 489, 499 (2019) (referring to the “abject lack of enforcement power in the [Presidential Records Act]”). Furthermore, so long as the president remains in office, the Act “accords the President virtually complete control over h[er] records.”⁹⁹Armstrong v. Bush, 924 F.2d 282, 290 (D.C. Cir. 1991). This includes the power to destroy records subject only to informing the head of the National Archives of her intentions (the Archivist may then inform Congress of her plans, though neither of them can block the president’s decision).¹⁰⁰Id.

Most importantly, as was true in the context of the Department of Justice’s regulations covering the pardon power, any argument that failure to comply with the Act renders a presidential pardon null and void would run headlong into Supreme Court precedent: “We therefore hold that the pardoning power is an enumerated power of the Constitution and that its limitations, if any, must be found in the Constitution itself.”¹⁰¹Schick v. Reed, 419 U.S. 256, 267 (1974). Ultimately, there is nothing in statute or regulation that would invalidate a dresser drawer pardon or penalize a president for issuing one.

D.      State Constitutions

The vast majority of state constitutions do not give governors the sweeping clemency powers the federal Constitution bestows upon the president.¹⁰²Kristin H. Fowler, Note, Limiting the Federal Pardon Power, 83 Ind. L.J. 1651, 1662 (2008). Only in North Dakota and South Dakota do governors possess a pardon power on a par with the American president.¹⁰³N.D. Const. art. V, § 7 (“The governor may grant reprieves, commutations, and pardons. The governor may delegate this power in a manner provided by law.”); S.D. Const. art. IV, § 3 (“The Governor may, except as to convictions on impeachment, grant pardons, commutations, and reprieves, and may suspend and remit fines and forfeitures.”). In other state constitutions, the pardons clauses are much more prolix than the federal one and regulate the clemency process in greater detail.¹⁰⁴The Texas Constitution provides a good example of the comparatively specific and lengthy texts found in state constitutions:

In all criminal cases, except treason and impeachment, the Governor shall have power, after conviction or successful completion of a term of deferred adjudication community supervision, on the written signed recommendation and advice of the Board of Pardons and Paroles, or a majority thereof, to grant reprieves and commutations of punishment and pardons; and under such rules as the Legislature may prescribe, and upon the written recommendation and advice of a majority of the Board of Pardons and Paroles, he shall have the power to remit fines and forfeitures. The Governor shall have the power to grant one reprieve in any capital case for a period not to exceed thirty (30) days; and he shall have power to revoke conditional pardons. With the advice and consent of the Legislature, he may grant reprieves, commutations of punishment and pardons in cases of treason. Tex. Const. art. IV, § 11(b).

To begin, many state constitutions confine the governor’s pardon power to seconding clemency decisions of a pardons board or recommending pardons for such a board’s approval.¹⁰⁵Fowler, supra note 102, at 1662 (“In twelve states, the executive has no unilateral pardon power; either the legislature or a separate (though often governor-appointed) board of pardons exercises the pardon power.”). Obviously, a dresser drawer pardon is impossible when the governor cannot act alone. In nearly half the states, the governor cannot unilaterally pardon someone prior to conviction.¹⁰⁶Id. at 1663. This change eliminates most of the utility of a secret pardon to those one wants to help since the defendant who is already imprisoned or wishes to regain his civil rights presumably wants to use (and thereby reveal) the pardon posthaste.

Roughly a third of state constitutions target transparency by requiring that the governor transmit a list of all pardons to the legislature.¹⁰⁷Id. In most of these states, the governor must also include his reasons for the pardons in his report.¹⁰⁸Id. at 1664. Two states focus on publicity by requiring announcements of pardons in newspapers: Mississippi mandates that a person applying for a pardon publish an announcement in a newspaper in the county where he committed the crime,¹⁰⁹Miss. Const. art. 5, § 124. and Maryland demands that the governor publish such an announcement before giving a pardon.¹¹⁰Md. Const. art. II, § 20.

Finally, some state constitutions, such as those of California, Colorado, and Illinois, empower the legislature to pass laws controlling the process of applying for a pardon.¹¹¹Cal. Const. art. V, § 8(a); Colo. Const. art. IV, § 7; Ill. Const. art. V, § 12. This power could be used to force pardoner and pardonee to give and receive clemency in the public eye.

E.      Dresser Drawer Deeds

“A deed is an instrument executed by a private citizen, and was formerly only known to be his act or deed because he made delivery of it as such.”¹¹²1 Robert T. Devlin, The Law of Real Property and Deeds § 5, at 8–10 (3d. ed. 1911) (quoting United States v. The Planter, 27 Fed. Cas. 544, 545 (D. Mo. 1852) (No. 16,054). As private actions, deeds need not be recorded to be effective,¹¹³5 Tiffany, supra note 2, § 1262. and official intervention, registration, or approval is not needed to make them valid.¹¹⁴Cf. id. § 1314 (explaining the “Torrens System” of land registration which requires recording land transactions in a register maintained by a government official). Deeds, however, must be signed by the donor and prepared in writing.¹¹⁵4 Tiffany, supra note 2, § 966. They must be delivered,¹¹⁶Id. § 1033. and they must be accepted¹¹⁷Id. § 1055 (“In many of the states, perhaps a majority, an acceptance of the conveyance by the grantee named therein has been stated to be essential to its validity . . . .”). (although acceptance is presumed subject to evidence of the transferee’s dissent¹¹⁸Id. § 1057.). Blackstone contrasts “deeds” with “charters”:

[A] deed is a writing sealed and delivered by the parties. It is sometimes called a charter, carta, from it’s [sic] materials; but mostly usually, when applied to the transactions of private subjects, it is called a deed, in Latin factum . . . because it is the most solemn and authentic act that a man can possibly perform, with relation to the disposal of his property . . . .¹¹⁹2 William Blackstone, Commentaries *295 (emphasis added).

The fact that transfers of real estate can occur in private corners, outside of public record and state oversight, is the archaic feature of modern law that makes dresser drawer deeds part of the arsenal of the clever attorney. Dresser drawer deeds can be used to discretely carry out estate planning, allowing clients to except certain real estate from the estate plan presented for public or family consumption in their will.

Clients who own real estate under a joint tenancy can discretely sever the four unities essential to this form of ownership by executing a dresser drawer deed.¹²⁰See generally 7 Richard R. Powell, On Real Property § 51.03 (Michael Allan Wolf ed., 2008) (explaining the four unities of time, title, interest, and possession necessary for formation and maintenance of a joint tenancy). I owe my familiarity with dresser drawer deeds (and this example of their use) to Professor Stanley Johanson’s masterful course in Wills and Estates at The University of Texas School of Law. By doing so, the client has secretly converted the joint tenancy to a tenancy-in-common and thereby ensured that his partial interest will pass on, rather than be eliminated at his death in favor of the other members of the erstwhile joint tenancy.¹²¹See generally id. § 51.04 (explaining that a “joint tenancy can be severed by a unilateral act of one of the tenants that is inconsistent with the continued existence of the joint tenancy or that operates to destroy or terminate any one or more of the essential unities, and such act effects conversion of the joint tenancy into a tenancy in common and destruction of the right of survivorship.” (footnote omitted)). An unscrupulous client could even instruct the beneficiary of the dresser drawer deed to keep it in the dresser forever unless the client predeceases the other members of the joint-tenancy and the deed is needed to prove prior conversion to a tenancy in common.

II.      Argument

Under the approach to the constitutional pardon power described by Chief Justice Marshall in Wilson, the president has the power to issue pardons, effective upon delivery and acceptance, without informing anyone but the recipient that she has done so.¹²²See supra Section I.B. As we will see, Wilson remains good law, and the Court has recurred to it in nearly every major pardons case of the nineteenth and twentieth centuries. Marshall rests his opinion on two planks: (1) the authority of the English treatise writers; and (2) formalistic reasoning working from the premise that pardons are private acts and so have to share the characteristics of other private legal actions such as real estate deeds.¹²³See discussion supra Section I.B. Both streams of reasoning lead us to the conclusion that secret pardons are no less valid for being concealed at their birth.

Against Marshall stand the Magnificent Yankee¹²⁴See The Magnificent Yankee (Metro-Goldwyn-Mayer 1950) (depicting the life of Supreme Court Justice Oliver Wendell Holmes, Jr.). and those scholars who have praised his opinion in Biddle for taking a modern, pragmatic view of the nature of presidential pardons.¹²⁵Cf. Godzilla vs. Kong (Legendary Pictures 2021) (depicting a clash of comparable titans). The clash between these two views of the pardon power—which Holmes notably declined to press to the point of overruling precedent—is of great interest in its own right for starkly pitting erudite common law formalism (that looks backwards) against shrewd reasoning from the observed needs and uses of public life (that looks forwards). As a matter of the state of the law today, however, Holmes’ failure to overrule Wilson or its progeny, the case’s persistent fecundity in the decades after Biddle, and the fact that the holding in Biddle can readily be subsumed by the elder case’s principles, all show that Wilson would govern the issue of a secret pardon’s validity today.¹²⁶But see Cowlishaw, supra note 74, at 160–64 (conceding that courts have followed the historical approach but critiquing their decision to do so as out of step with the modern criminal justice system and federal state).

Some scholars might respond that secret pardons are contrary to over two centuries of tradition in which presidents have made their pardons known to the public upon issuance. However, this argument is mistaken in fact and limited in persuasive power. Presidents have granted pardons without announcing them before, and the weight of centuries of de jure precedent, especially when it originates with a founding era titan like Marshall, should outweigh centuries of de facto precedent.

A.      Dresser Drawer Pardons Are Valid

The story of the constitutional pardon power in the United States Reports begins with Marshall announcing, in the Court’s first case about the pardons clause, that English principles will control how the Court reads the pardons clause. After reciting the brief text of the clause, Marshall tells us to “look into their books” to interpret it.¹²⁷United States v. Wilson, 32 U.S. (7 Pet.) 150, 160 (1833). The English books taught Marshall that a pardon is a “private” act of mercy or grace done by the “executive magistrate.”¹²⁸Id. at 160–61. As Coke wrote, to speak of pardons is to “speak of his [the king’s] mercy.” 3 Edward Coke, Institutes of the Laws of England 233 (5th ed. 1671). Marshall’s major premise then is that pardons are private acts of mercy. By his lights, it follows that, as private acts or “deeds,” pardons are to be analyzed like other private acts or deeds.¹²⁹See Wilson, 32 U.S. at 161; cf. 2 Blackstone, supra note 119 (“[W]hen applied to the transactions of private subjects, it is called a deed, in Latin factum . . . because it is the most solemn and authentic act that a man can possibly perform, with relation to the disposal of his property . . . .”). From here on out, the solution to Marshall’s problem in Wilson is treated as a matter of deduction from the categorical properties of the genus to the properties of the species: delivery and acceptance are essential to deeds and so are essential to pardons.¹³⁰See Wilson, 32 U.S. at 161. It followed that Wilson was free to reject the pardon and not plead it, leaving the court with no right to take notice of it.¹³¹Id. at 163.

Eighty-two years later, Justice McKenna confirmed this reading of Wilson for the Court through his opinion in Burdick.¹³²Burdick v. United States, 236 U.S. 79, 91 (1915). Speaking for the majority, he wrote, “A pardon was denominated as the ‘private’ act, the ‘private deed,’ of the executive magistrate, and the denomination was advisedly selected to mark the incompleteness of the act or deed without its acceptance.”¹³³Id. at 90.

What happens if we accept the same major premise and apply it to the question of whether a pardon kept secret from all but the recipient in its inception is a valid pardon? Put differently, what answer does the fact that pardons are a species of the genus deed compel to this question? To begin with, Blackstone says that “a deed is a writing sealed and delivered by the parties” and that it “is the most solemn and authentic act that a man can possibly perform, with relation to the disposal of his property.”¹³⁴2 Blackstone, supra note 119, at *295. Being “delivered by the parties,” a pardon needs only two people; the recognition of other individuals, be they public officials or private citizens, is not required to make this private act of grace effective.¹³⁵Id.; cf. 5 Tiffany, supra note 2, § 1262 (explaining that deeds need not be recorded to be effective). And lest we doubt this reasoning is tethered to the law here in America, we know that Marshall would not have objected to speaking of pardoning as disposing of property because he warmly quotes both Hawkins, who says that pardons are a species of “property,”¹³⁶Wilson, 32 U.S. at 162 (quoting 2 Hawkins, supra note 30, at 397). and Blackstone, who says that a pardon is something a man can get from the king and hold “in his pocket.”¹³⁷Id. at 163 (quoting 4 Blackstone, supra note 31, at *401). It follows that the president, having the power to give a pardon, can effectively grant it to anyone who will accept delivery without giving notice to any stranger to the transaction.

A scholar can work out this conclusion from reading Wilson and perusing the English sources Marshall endorses therein. She can then confirm that his approach is still good law by looking at the cases that followed it up to the end of the last century. As recently as 1993, the Supreme Court quoted extensively from Wilson,¹³⁸Herrera v. Collins, 506 U.S. 390, 412–13 (1993). and in 1974, it affirmed that “[t]he teachings of Wilson and Wells have been followed consistently by this Court.”¹³⁹Schick v. Reed, 419 U.S. 256, 266 (1974). The only ground for doubting the continued viability of Marshall’s approach to the pardons clause lies in Holmes’ opinion in Biddle.¹⁴⁰See supra Section I.B. But as discussed in the next section, the stream that springs from Wilson bends around this rock to resume its inaugural course in the jurisprudence of today.

B.      In Defense of Rare Birds, Pardons as Private Acts

Chief Justice Marshall’s august name notwithstanding, readers may have balked at the description of presidential pardons as “private acts.” The idea that the decision to pardon—a decision that binds prosecutors, courts, and the Bureau of Prisons—is only a private act, not a public matter, is bound to shock and disorient some readers. Initial shock notwithstanding, these are the words chosen by the Court in cases that the Court has never overruled but, on the contrary, has harkened back to with so many hosannahs and amens. In fact, the Court stressed in Burdick that Chief Justice Marshall had used the words “‘private’ act” and “private deed” to describe pardons “advisedly.”¹⁴¹Burdick v. United States, 236 U.S. 79, 90 (1915).

The law is what it is, but what can be said in apology for it? First, pardons, considered as private acts, may seem less anomalous when we realize that they are not the only example from the founding period of private rights to determine public affairs. Second, the emotion and subjectivity in mercy offers a reason for treating pardons as private acts even in the twenty-first century.

At the time of the founding, pardons by the English king were not the only species of private act that affected public matters. Readers of Regency fiction like Jane Austen’s novels will recall certain characters seeking or enjoying clerical benefices, or, in the novels’ words, “livings.”¹⁴²Jane Austen, Sense and Sensibility 237 (Cassell & Co. 1908). In Sense and Sensibility, the generous Colonel Brandon gifts the protagonist’s warm and pacific beau Edward “the living of Delaford,” making him rector to the local villagers.¹⁴³Id. In fiction as in life, a private man of means has in his gift preferment to an office in the established Church of England. This private act certainly has public consequences for the villagers who will have the man he names for their official spiritual shepherd.

Blackstone teaches that what Colonel Brandon had was an “advowson”—”the right of presentation to a church, or ecclesiastical benefice.”¹⁴⁴2 Blackstone, supra note 119, at *21. The patron’s choice was subject to approval by the bishop, and the patron could only successfully appoint someone who met certain qualifications, like being an educated member of the clergy.¹⁴⁵1 id. at *389. Still, it is quite remarkable that appointment to such a public-facing state office was a matter of private right as late as Blackstone’s and Austen’s days.

Blackstone says that a type of advowson was the advowson appendant, which was attached to the manor and conveyed whenever the manor was conveyed.¹⁴⁶2 id. at *22. Like the president’s or king’s right to pardon, the advowson was a right to convey certain rights to another: Blackstone writes, “It is not itself the bodily possession of the church and its appendages; but it is a right to give some other man a title to such bodily possession.”¹⁴⁷Id. at *21. Like pardons, the livings conveyed by holders of advowsons like Colonel Brandon were a form of property “conveyed” from one person to another: “The patronage can therefore be only conveyed by operation of law, by verbal grant, either oral or written, which is a kind of invisible mental transfer . . . .”¹⁴⁸Id. at *22. Moreover, the patronage conveyed could “lie[] dormant and unnoticed, till occasion calls it forth; when it produces a visible corporeal fruit, by intitling some clerk, whom the patron shall please to nominate, to enter and receive bodily possession of the lands and tenements of the church.”¹⁴⁹Id. In like manner, a pardon conveyed can lie dormant and unnoticed until the occasion of an indictment or the desire to walk out of prison calls it forth, when it produces the visible corporeal fruit of the pardoned person walking free.

Leaving Regency Britain for twenty-first-century America, there remains at least one freestanding policy reason for treating pardons as private acts. Namely, while pardons are a tool of statecraft, they are also part of the criminal justice system, and within that system, they serve both to correct miscarriages of justice (such as excessive sentences or cases of actual innocence) and to temper justice with mercy.¹⁵⁰See supra notes 17–18 and accompanying text. As mentioned previously, the word “mercy” sometimes refers to a fuller form of justice.¹⁵¹See discussion supra Section I.A. For example, mercy may mean doing perfect justice by taking into account facts about a person’s character and background that ordinary legal proceedings abstract away from.¹⁵²See supra note 18 and accompanying text. However, “mercy” also refers to clemency that runs counter to justice (i.e., when a person deserves a greater punishment but a lesser punishment is imposed out of compassion for the individual).¹⁵³See supra note 17 and accompanying text.

Mercy—in the sense in which it conflicts with justice rather than completes it—is a canonical reason for the existence of the pardon power. Talking of pardons and quoting the Book of Proverbs, Coke writes, “Mercy and truth preserve the King, and by clemency is his Throne strengthened.”¹⁵⁴3 Coke, supra note 128, at 233 (quoting Proverbs 20:28). Johnson’s dictionary in turn defines “clemency” as “[m]ercy; remission of severity; willingness to spare; tenderness in punishing.”¹⁵⁵Clemency, Samuel Johnson, A Dictionary of the English Language (1755). Picking up on the theme, Hamilton argued that without the pardon power, “justice would wear a countenance too sanguinary and cruel.”¹⁵⁶The Federalist No. 74 (Alexander Hamilton).

Tenderness and leniency can best be achieved if the pardon power is vested in an individual who can act without explaining herself rather than a group of people who must deliberate, announce reasons, and conform their decisions to objective standards. As Hamilton said, “[O]ne man appears to be a more eligible dispenser of the mercy of the government than a body of men.”¹⁵⁷Id.

After all, mercy proceeds at least as much from the heart as it does from the head. Johnson’s dictionary defined “tenderness” as “susceptibility of impression. . . . [k]ind attention . . . . [s]crupulousness; caution.”¹⁵⁸Tenderness, 1 Samuel Johnson, A Dictionary of the English Language (1755). Hence, mercy implicates an individual’s susceptibility to be pained or moved to kindness by the hard lots of others, including their deserved hard lots for the crimes they have committed. Whether or not a person is moved to compassion by the plight of others—how he is affected by hearing of their plight—depends on how hardhearted or softhearted he is. It is definitionally subjective.

By contrast, objective reasons are the essential stuff of public business in a liberal republic of laws.¹⁵⁹Cf. 4 Blackstone, supra note 31, at *390 (“In democracies . . . this power of pardoning can never subsist; for there nothing higher is acknowledged than the magistrate who administers the laws . . . .”). John Rawls explains, “Our exercise of political power is proper only when we sincerely believe that the reasons we would offer for our political actions—were we to state them as government officials—are sufficient, and we also reasonably think that other citizens might also reasonably accept those reasons.”¹⁶⁰John Rawls, The Idea of Public Reason Revisited, 64 U. Chi. L. Rev. 765, 771 (1997). On this view, a public official, be she a judge, bureaucrat, or legislator, should always be able to state objective reasons for her actions. By contrast, when she is choosing how to spend her money (i.e., use her private property) she can spend it on a green car rather than a blue car based only on how the green one affects her and makes her feel.

It makes sense to admit that pardoning is a private act once we understand that merciful, compassionate pardons are incommensurable with the principle that all public actions are rationally justifiable actions. Yet, pardoning’s private character is exactly what we can plausibly want from the pardon power in order to introduce pathos—leniency and tenderness—into the criminal justice system. By the same token, what the founding generation plausibly wanted is a private power in the chief executive to act as he or she is moved by compassion to remit punishments.¹⁶¹See The Federalist No. 74 (Alexander Hamilton). Sometimes, the Constitution is not perfectly principled but contains historical idiosyncrasies—like the limitation of the presidency to natural born citizens—that are irreconcilable with its general tenor of equality, liberty, and democracy.¹⁶²See Robert Post, What Is the Constitution’s Worst Provision?, 12 Const. Comment. 191, 193 (1995) (“Thus at the very heart of the constitutional order, in the Office of the President, the Constitution abandons its brave experiment of forging a new society based upon principles of voluntary commitment; it instead gropes for security among ties of blood and contingencies of birth.”).

To have pathos in our criminal justice system, we need to let some individual’s heart serve as the fountain of mercy. It will not do for the government to have a heart by committee or subject the pardon power to a requirement of public deliberation or judicial review. While clearly an aberration in the United States’ republican political order, the fact that our law treats the president’s pardons as private acts does find some justification in the case for compassionate pardons.

C.      Oral Pardons Are Very Likely Invalid

Marshall’s approach to presidential pardons treated them as a kind of deed—a private act of grace—and insisted that they be understood in light of the English king’s pardon power. On that view of the pardons clause, could a president issue a valid oral pardon?

Suppose that Leslie Knope, former chief of staff to the late President Minerva, is indicted for federal income tax fraud but moves to quash the indictment. She alleges that during the last week of her former boss’s term of office, President Minerva stated—in the presence of Knope, the White House butler, and a West Wing secretary—that Knope was pardoned for all crimes she had ever committed against the United States. At a hearing on Knope’s motion to quash, she produces the butler and secretary who both confirm her allegations. In the alternative, imagine that Minerva is still living and that Knope produces her at the hearing to testify to her own words of pardon. Finally, imagine that Knope used her iPhone to record a video of President Minerva’s words and that she then produces it at the hearing on her motion instead of calling witnesses. How should the court rule?

English authorities offer support for the position that a pardon must be in writing. Namely, Blackstone requires that pardons be issued under the Great Seal of the United Kingdom.¹⁶³See 4 Blackstone, supra note 31, at *396. Hawkins also habitually speaks of pardons from the king as pardons “under the Great Seal.”¹⁶⁴E.g., 2 Hawkins, supra note 30, at 396. Hawkins also notes an English case in which a witness, to prove that he was now competent to testify despite a prior criminal conviction, had to produce a pardon under the Great Seal, as letters under the king’s sign manual were only evidence of the king’s intention to pardon, and not a pardon themselves. See id. at 433. This implies that pardons issued by the monarch must be in writing, elsewise there would be nothing to which he or she could affix the Great Seal.

The analogy to deeds also supports a requirement that pardons be written down. Blackstone says that “a deed is a writing sealed and delivered by the parties.”¹⁶⁵2 Blackstone, supra note 119, at *295 (emphasis added). Deeds are “instrument[s]”¹⁶⁶1 Devlin, supra note 112, § 5. executed by private parties and must be prepared in writing and signed (or sealed) in order to be effective.¹⁶⁷4 Tiffany, supra note 2, § 966. Making a deed is supposed to be “the most solemn and authentic act that man can possible perform, with relation to the disposal of his property.”¹⁶⁸2 Blackstone, supra note 119, at *295 (emphasis added). It stands to reason that if the president issues a pardon, this deed must be executed with the requisite solemnity, and, at minimum, be made in writing.

On the other hand, the Court has unequivocally held that the “pardoning power is an enumerated power of the Constitution and that its limitations, if any, must be found in the Constitution itself.”¹⁶⁹Schick v. Reed, 419 U.S. 256, 267 (1974). Significantly, English requirements like the use of the Great Seal were not carried over into the language of the Constitution itself. The text of the Constitution’s pardons clause says nothing about the Great Seal of the United States,¹⁷⁰See U.S. Const. art. II, § 2, cl. 1. and no authority insists that pardon documents must have the Seal. Likewise, the nonbinding regulations devised by the Department of Justice to guide the pardon process make no mention of the Great Seal¹⁷¹See 28 C.F.R. §§ 1.1–.11 (containing the pardon guidelines). or of certification by the Secretary of State.¹⁷²See id.; see also 4 U.S.C. § 42 (giving charge and custody of the seal of the United States to the Secretary of State). These facts leave some room to argue that oral pardons are valid. Nonetheless, while not without plausibility, this position makes the weaker side of the argument.

Faced with an oral pardon, the courts would likely hold it invalid. Whether a defendant provided video evidence, audio recording, or witness testimony, the courts would likely decline to honor it. The courts would surely be aware of the risk of perjury when witness testimony is used to prove a pardon, especially after the death of a former president. They also would not want to tempt an unscrupulous, living ex-president to perjure himself or herself about oral pardons that never occurred. With these considerations in the back of their minds, the courts would likely seize on the English authorities that imply pardons must be put down in writing, as all other deeds must be set down on paper.

D.      Self-Pardons Are Invalid

Under the approach to the pardons clause developed by Marshall, a reflexive presidential pardon would be a nullity. In short, one cannot make a deed to oneself; therefore, as a pardon is a kind of deed, one cannot issue a pardon to oneself.¹⁷³See Allgood v. Allgood, 132 S.E. 48, 51 (S.C. 1926). While scholars have thus far neglected dresser drawer pardons, authors of law review pieces have flocked to discuss self-pardons in recent years. Their arguments range broadly from constitutional structure to basic principles of justice forbidding self-dealing and judging one’s own cause.¹⁷⁴E.g., Michael Conklin, Please Allow Myself to Pardon . . . Myself: The Constitutionality of a Presidential Self-Pardon, 97 U. Det. Mercy L. Rev. 291, 292–301 (2020); Alberto R. Gonzales, Presidential Powers, Immunities, and Pardons, 96 Wash. U. L. Rev. 905, 933–36 (2019); Robert Nida & Rebecca L. Spiro, The President as His Own Judge and Jury: A Legal Analysis of the Presidential Self-Pardon Power, 52 Okla. L. Rev. 197, 216–20 (1999); Brian C. Kalt, Note, Pardon Me?: The Constitutional Case Against Presidential Self-Pardons, 106 Yale L.J. 779, 781–82 (1996). In this section, the discussion of self-pardons confines itself to the view—reflected in the English authorities, Wilson, and its progeny—that pardons are a species of deed.

A reflexive deed is a nullity:

It is a rule asserted from early times that no grant can exist without a grantee. This is, of course, axiomatic, for the title cannot pass from the grantor unless it passes to some one; and it is also axiomatic that a deed must have both a grantor and a grantee, and one person cannot occupy, at law, at the same time, the position of both grantor and grantee in regard to the same property.¹⁷⁵Allgood, 132 S.E. at 51; accord City Bank of Portage v. Plank, 124 N.W. 1000, 1001 (Wis. 1910).

“So that, as clear as is the summer’s sun,”¹⁷⁶William Shakespeare, Henry V act 1, sc. 2, l. 91. “[a]t common law a man cannot make a conveyance to himself. Such action on his part is nugatory and if attempted he still holds under his original title.”¹⁷⁷Strout v. Burgess, 68 A.2d 241, 247 (Me. 1949).

Attorneys reading this will likely remember the use of straw-grantees in order to avoid this rule: a grantor deeds property to a third person who can then deed it back to her.¹⁷⁸See generally John V. Orth, The Perils of Joint Tenancies, 44 Real Prop. Tr. & Est. L.J. 427, 429 (2009) (“The common law solution to this common law problem was to use a so-called ‘straw conveyance,’ that is, for the grantor to convey the entire estate to a compliant third party (the ‘straw man’), who then immediately reconveyed to the original grantor and the other as joint tenants, thereby perfecting the unities.”). This device is needful, for instance, where the grantor wishes to create a joint tenancy in herself and another person. Otherwise, a joint tenancy cannot be established by deeding property to oneself and another—only a tenancy in common is created.¹⁷⁹See Strout, 68 A.2d at 254 (“The overwhelming weight of authority, in States which deny the power on the part of a grantor to create a joint tenancy between himself and another by a direct conveyance to himself and another, is that such conveyance creates a tenancy in common between the grantor and his intended joint tenant.”). As the Supreme Judicial Court of Maine explained in one such case, “It follows therefore that after an attempted transfer from one to himself and another the transferor still holds under his original title which accrued to him at the time of his original acquisition of the property, be the same real or personal.”¹⁸⁰Id. at 247.

It follows that as pardons are a species of deed, a pardon to oneself is void for want of a distinct grantor and grantee. Recall that Marshall quoted Blackstone’s description of a pardon as something granted by the king.¹⁸¹United States v. Wilson, 32 U.S. (7 Pet.) 150, 162–63 (1833) (quoting 4 Blackstone, supra note 31, at *401). Specifically, he wrote that a pardon is something privately “delivered” by “the executive magistrate” “to the individual for whose benefit it is intended.”¹⁸²Id. at 160–61. It is nigh too simple to bear saying that the “words ‘convey,’ ‘transfer,’ and similar words [like ‘delivered’] employed in conveyancing, signify the passing of title from one person to another.”¹⁸³Deslauriers v. Senesac, 163 N.E. 327, 328 (Ill. 1928). And as we have seen, the common law is not so subtle that it can make one into two and let grantor and grantee be the same person: “For every alienation there must be an alienor and an alienee, for every grant a grantor and a grantee, and for every gift a donor and a donee.”¹⁸⁴Id. In the case of a pardon, the gift is the pardon; in the case of a self-pardon, the fault is want of a distinct grantor and grantee.

Self-pardons are also void because the pardon is a kind of property that must change hands when the pardon is granted. Remember that Marshall quoted Hawkins, who described a pardon as a kind of property held by the pardonee from the monarch¹⁸⁵Wilson, 32 U.S. at 162 (quoting 2 Hawkins, supra note 30, at 397). Yet, a person “cannot by deed convey an estate to himself or take an estate from himself.”¹⁸⁶Deslauriers, 163 N.E. at 329. Where the “charter of pardon”¹⁸⁷4 Blackstone, supra note 31, at *401. is the estate in question, the president cannot pass it from his right hand to his left hand and call it a grant of clemency. The courts have reminded putative grantors of this by calling their attention to the old requirement of livery of seizin: “At common law, livery of seizin was necessary to pass the title to real property, and it was recognized that a person could not make livery of seizin to himself.”¹⁸⁸Deslauriers, 163 N.E. at 329. Nowadays, even though “livery of seizin has been rendered unnecessary . . . the muniment of title, namely, the deed, must still be delivered.”¹⁸⁹Id.

III.      Counterarguments

A.      Justice Holmes and the “Modern” View of Pardons

The reasoning reciprocating in Wilson diverges from the thinking humming in Biddle as much as a steam engine differs from a diesel motor. As Professor G. Sidney Buchanan, put it, “the Holmes definition of a pardon conflicts with that advanced by Marshall; there is here little room for peaceful coexistence.”¹⁹⁰Buchanan, supra note 53, at 36; see also Todd David Peterson, Congressional Power of Pardon & Amnesty: Legislative Authority in the Shadow of Presidential Prerogative, 38 Wake Forest L. Rev. 1225, 1234 (2003) (“Justice Holmes suggested a different rationale for the pardon power than Chief Justice Marshall had enunciated early in the 19th century. Rather than being a private act of grace that must be accepted and proffered to the court by the one pardoned, Justice Holmes saw the President’s pardon as serving public policy ends . . . .”). Marshall’s approach in Wilson is historical and deductive, beginning by establishing from English authority that pardons are a kind of deed, and proceeding to the conclusion that like other deeds, pardons must be accepted to be effective.¹⁹¹See Wilson, 32 U.S. at 161–63. Holmes, on the other hand, “perceived the pardon power as an instrument of the public welfare, as serving a policy function much broader than satisfying the wishes of the pardonee.”¹⁹²Buchanan, supra note 53, at 47; see also S. Elizabeth Gibson, Constitutional Law—Presidential Pardons and the Common Law, 53 N.C. L. Rev. 785, 789 (1975) (“[Justice Holmes] rested [his] conclusion concerning the nature of a pardon on logic rather than on common-law principles or on concepts existing at the time the Constitution was drafted.”). In what Buchanan correctly called a “total repudiation of the Marshall conception of a pardon as a private act of grace,” “Holmes stressed that the pardon power is a part of the constitutional scheme, a power possessed by the executive, not for the purpose merely of bestowing executive grace, but as a check against judicial and legislative excesses.”¹⁹³Buchanan, supra note 53, at 47. As Buchanan highlights,¹⁹⁴Id. the gruff Justice underscores the public character of a pardon by declaring that the pardonee should not get “any voice in what the law should do for the welfare of the whole.”¹⁹⁵Biddle v. Perovich, 274 U.S. 480, 487 (1927).

Buchanan may be right that Holmes meant to make a revolution in the Court’s approach to pardons. But the result has more in common with historical dead ends like the Jacobite “45” than the epoch-making turns of 1776 and 1789. Whether it was for lack of will, votes, or some other reason, Holmes declined to overrule any of the cases in Wilson’s line, declaring only that a recent decision reaffirming Wilson—Burdick—was “not to be extended to the present case.”¹⁹⁶Id. at 488. Holmes settled thereby that effective commutation of a death sentence does not depend upon the recipient’s acceptance.¹⁹⁷See id. at 487. But jurisprudentially, this was only a special case, a tangential spur. On the mainline, Marshall’s approach to pardons chugged along unhampered. When the Supreme Court returned to the pardons clause with Schick in 1974, Chief Justice Burger omitted citing to Biddle entirely, instead letting Wilson and its progeny shine in the limelight.¹⁹⁸Schick v. Reed, 419 U.S. 256, 264–66 (1974). What’s more, when Chief Justice William Rehnquist wanted to print a hornbook exposition of the pardon power in the 1990s, he reprinted two paragraphs from Wilson while omitting any mention of Biddle in his opinion for the court.¹⁹⁹Herrera v. Collins, 506 U.S. 390, 412–13 (1993).

For his part, Buchanan argues that Schick reflects the spirit of Biddle, not Wilson.²⁰⁰See Buchanan, supra note 53, at 47. Scholars who have come after Buchanan, Kathleen Ridolfi for example, claim that the Court changed course for good with Biddle: “Thus, the Supreme Court, without expressly overruling Burdick, altered the theory of the pardoning power. The pardon was a private act of grace no longer, but an act for the public welfare.” Kathleen M. Ridolfi, Not Just an Act of Mercy: The Demise of Post-Conviction Relief and a Rightful Claim to Clemency, 24 N.Y.U. Rev. L. & Soc. Change 43, 59 (1998) (footnote omitted); see also Haugen v. Kitzhaber, 306 P.3d 592, 609 (Or. 2013) (en banc) (adopting Holmes’s view of pardons as a matter of the public welfare); Fletcher v. Graham, 192 S.W.3d 350, 412 (Ky. 2006) (Green, S.J., concurring in part and dissenting in part) (“Under the modern view, the pardon power is a public policy tool rather than a mechanism for the granting of mercy.”). He asserts that the Court reaffirmed Holmes’ position “obliquely” at that time.²⁰¹Buchanan, supra note 53, at 47. Of course, when assessing Schick, the reader must assess whether the Court could have been obliquely approving Biddle despite neglecting to cite it even once—instead putting Wilson and Wells in the seats of honor.²⁰²See Schick, 419 U.S. at 264–66. That said, Buchanan sees in Burger’s comments on conditional pardons an implicit repudiation of the acceptance requirement described in Wilson and Burdick.²⁰³See Buchanan, supra note 53, at 48.

Schick was an American soldier, stationed in Japan, convicted by court-martial of the brutal murder of an eight-year-old girl.²⁰⁴Schick, 419 U.S. at 257. President Eisenhower commuted his death sentence to life imprisonment without the possibility of parole in 1960.²⁰⁵Id. at 258. Eisenhower explicitly wrote that the commutation was made “on the condition” that Schick would never benefit from parole.²⁰⁶Id. When the Supreme Court held the death penalty unconstitutional twelve years later in Furman v. Georgia,²⁰⁷408 U.S. 238, 239–40 (1972). Schick argued that the condition on his pardon either had never been valid or had become invalid after Furman.²⁰⁸Schick, 419 U.S. at 260.

In relevant part, Schick took the position that Eisenhower exceeded his pardon powers because the Uniform Code of Military Justice had not authorized a sentence of life imprisonment without parole for Schick’s offense.²⁰⁹Id. The Court responded by considering whether the English king had ever made pardons or commutations conditional on a punishment not authorized by law.²¹⁰Id. at 261. The Court noted the king had made pardons conditional on transportation to the colonies, even though “British subjects generally could not be forced to leave the realm without an Act of Parliament.”²¹¹Id. The Court remarked that the king relied on “a legal fiction”—the defendant consented to transportation—to make her banishment copacetic.²¹²Id.

Buchanan seizes on the Court’s choice to describe the transportees’ consent to commutation as a “legal fiction.”²¹³Buchanan, supra note 53, at 48. He argues that by belittling consent to commutation as a “legal fiction,” the Court “sides more with Holmes than with Marshall.”²¹⁴Id. The trouble with this reasoning is that the words “legal fiction” come in the context of deciding a pardons case by reference to history and English legal tradition—precisely the method recommended by Marshall in Wilson and foresworn by Holmes in Biddle!²¹⁵See United States v. Wilson, 32 U.S. (7 Pet.) 150, 162–63 (1833); Biddle v. Perovich, 274 U.S. 480, 486 (1927). The Court’s recourse to history—and its explicit citations to Wilson and Wells to prove that history is the correct lens through which to read the pardons clause—belie Buchanan’s claim that Schick “obliquely” reaffirmed Biddle.

Ultimately, awarding pride of place to Biddle and its policy concerns makes for a poor reading of the caselaw in a field where it is the outlier. We can, however, harmonize the Supreme Court’s decisions on the pardons clause by reading Biddle narrowly to stand for the proposition that commutation of a death sentence does not require the recipient’s acceptance to be effective. This was the holding in Biddle,²¹⁶Biddle, 274 U.S. at 487–88. and it can be readily subsumed to the historical approach followed in Wilson and its progeny. English authority, quoted by Marshall in Wilson, already contained an exception for death sentences: “[I]f the king pardons a felon, and it is shown to the court, and yet the felon pleads guilty, and waives the pardon, he shall not be hanged . . . .”²¹⁷Wilson, 32 U.S. at 162 (quoting Case LXII 145 Eng. Rep. 90; Jenk. 129). Holmes acknowledges this text in his opinion when he narrates the position of the Solicitor General, but then confessedly opts not to “go into history.”²¹⁸Biddle, 274 U.S. at 486. There was thus no real conflict about the right outcome between the historical–formalist approach of Wilson and Holmes’ new path. Since Holmes then declined to overturn precedent, and since the Court ignored his approach in later years, it now makes for a better reading of the caselaw to understand Biddle as the Court’s acknowledgement of the exception for death cases that the English authorities pointed to all along.²¹⁹Cf. Krent, supra note 71, at 1682 (“The legal community should understand Biddle narrowly, reflecting a widespread, but by no means irrefutable, belief that rational offenders would accept any conditional commutation that saves them from death.” (footnote omitted)).

 

* * *

 

In a curious irony, Holmes’ own approach to deciding the content of the law would lead us to adopt Marshall’s historical–formalist approach at this time. Holmes famously wrote, “If you want to know the law and nothing else, you must look at it as a bad man, who cares only for the material consequences which such knowledge enables him to predict . . . .”²²⁰Holmes, supra note 62, at 459. A bad man with a pardon in his sock drawer would be keen to know that the current Court would likely hold that the pardon is valid.

The Roberts Court is “very formalistic” and often relies on Founding-Era history.²²¹Ofer Raban, Between Formalism and Conservatism: The Resurgent Legal Formalism of the Roberts Court, 8 N.Y.U. J.L. & Liberty 343, 344–45 (2014). It “makes many decisions that rely exclusively on narrow fact-like considerations, like a dictionary definition or some historical circumstance.”²²²Id. at 345. Likewise, the contemporary Court looks askance at “any deliberations over the merits of cases—including time-honored considerations like expected consequences, the legislative purpose or history of a statutory or constitutional provision, changed conditions, coherence with other laws, or, of course, considerations of justice.”²²³Id. Moreover, in the constitutional domain, originalism is ascendant at the Court.²²⁴Id. For a ready example, consider the Court’s treatment of the Second Amendment in District of Columbia v. Heller.²²⁵554 U.S. 570, 605–10 (2008). In that opinion, the Court was proudly antiquarian, looking to Founding-Era conditions and English legal history between the Glorious Revolution and the American Revolution to decide the meaning of the Amendment’s text when it was adopted.²²⁶See Raban, supra note 221, at 347 (citing Heller, 554 U.S. at 581, 591–92) (observing that the Supreme Court cited eighteenth century dictionaries and debates in the House of Lords). We should not expect such an archaist Court to agree with Holmes to “not go into history.”²²⁷Biddle v. Perovich, 274 U.S. 480, 486 (1927). Rather, we should tell the bad man to bet that the Court will agree with Marshall to travel to English archives and “look into their books for the rules.”²²⁸United States v. Wilson, 32 U.S. (7 Pet.) 150, 160 (1833).

B.      Historical Practice and Tradition

A critic could argue that presidents historically have announced their pardons publicly and that the courts should refuse to break with that de facto precedent by honoring dresser drawer pardons. For example, even when President Clinton incurred opprobrium by issuing midnight pardons to campaign contributors at the end of his term, the fact that he issued them was immediately known to the public.²²⁹Marc Lacey, Clinton Issues Pardons, Clearing Deutch and McDougal, but Not Milken or Hubbell, N.Y. Times (Jan. 21, 2001), https://perma.cc/V8H6-4WUN. On this argument, what Justice Felix Frankfurter called the “gloss” put on the Constitution by decades of practice by officials working under it carries weight when interpreting the Constitution.²³⁰Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 610 (1952) (Frankfurter, J., concurring). According to the critic, the choices of past presidents to make their pardon decisions public weighs decisively against the validity of dresser drawer pardons now. This argument draws support from the fact that the Court has given weight to historical practice in a prior pardons case. In Schick, the Court defended attaching conditions, not authorized by statute, to pardons by noting that presidents had done so throughout American history.²³¹Schick v. Reed, 419 U.S. 256, 266 (1974). Quoting Holmes, Chief Justice Burger wrote, “‘If a thing has been practiced for two hundred years by common consent, it will need a strong case’ to overturn it.”²³²Id. (quoting Jackman v. Rosenbaum Co., 260 U.S. 22, 31 (1922)).

However, the argument from historical practice is weak here for three reasons. First, the argument relies on the absence of a practice rather than its unchallenged maintenance. Second, courts—besides regarding it as best invoked only when judicial precedent is sparse—rightly see de facto precedent as weaker authority than judicial precedent on issues of criminal law. And third, presidents have in fact issued pardons without publicity in the past, even if they have not done so (to our knowledge) in recent decades.

In his concurrence in Youngstown Sheet & Tube Co. v. Sawyer,²³³343 U.S. 579 (1952). Justice Frankfurter explained how historical practice puts a gloss on the Constitution:

The Constitution is a framework for government. Therefore the way the framework has consistently operated fairly establishes that it has operated according to its true nature. Deeply embedded traditional ways of conducting government cannot supplant the Constitution or legislation, but they give meaning to the words of a text or supply them. It is an inadmissibly narrow conception of American constitutional law to confine it to the words of the Constitution and to disregard the gloss which life has written upon them. In short, a systematic, unbroken, executive practice, long pursued to the knowledge of the Congress and never before questioned, engaged in by Presidents who have also sworn to uphold the Constitution, making as it were such exercise of power part of the structure of our government, may be treated as a gloss on “executive Power” vested in the President by § 1 of Art. II.²³⁴Id. at 610–11 (Frankfurter, J., concurring).

The gist of Frankfurter’s claim is that there are times when official actions are not mere mute events but are articulate happenings. The challenge is to distinguish the dumb acts from the speaking ones and settle on an objective way of deciding the meaning of the latter. In a landmark article on the role custom should play in constitutional disputes, Professor Michael Glennon set out three criteria: “First, the custom in question must consist of acts; mere assertions of authority to act are insufficient. Second, if a coordinate branch has performed the act, the other branch must have been on notice of its occurrence. Third, the branch placed on notice must have acquiesced in the custom.”²³⁵Michael J. Glennon, The Use of Custom in Resolving Separation of Powers Disputes, 64 B.U. L. Rev. 109, 134 (1984). More recently, Professors Curtis Bradley and Neil Siegel produced a similar list of three criteria based on National Labor Relations Board v. Noel Canning,²³⁶573 U.S. 513 (2014). the Court’s most recent deep discussion of the role of historical gloss in constitutional interpretation.²³⁷Curtis A. Bradley & Neil S. Siegel, Historical Gloss, Madisonian Liquidation, and the Originalism Debate, 106 Va. L. Rev. 1, 18–20 (2020). They aver that a historical practice must be a governmental practice, must be of longstanding duration, and must have been acquiesced in by the affected other branches of government.²³⁸Id. (describing governmental practice as “the actions and inactions of government institutions, whether executive, legislative, or judicial”).

While written legal precedent does not speak unequivocally on the issue of secret pardons, it cannot be said that historical facts speak plainly or unambiguously either. If a man comes to a diner everyday but never orders eggs, can we assert confidently that he does not enjoy eating eggs, or is it more sensible to infer that he chooses oatmeal instead because he thinks it better for his cholesterol to eat porridge? Can we assume that his wife approves of him avoiding eggs if she never complains about his choices? When it comes to secret pardons, the supposed fact that presidents have (as far as we know) refrained from issuing them in the past means that Congress has not had the occasion to take notice of and question them.²³⁹There is, however, one instance of pardons only becoming publicly known after a President left office. This minor scandal at the end of the Truman administration and beginning of the Eisenhower administration is discussed at length below. While public and Congressional criticism prompted the attorney general to announce a policy of making all pardons public in the future, this one instance is not enough to make a strong argument from historical practice. See infra notes 252–54 and accompanying text. Secret pardons cannot have been either hallowed or profaned by the nodding or balking of other branches because there has been nothing at which to nod or balk. In short, none of Glennon’s three criteria can get any bite on a record of omissions and not actions.

In fairness, Bradley and Siegel do allow that a history of not doing something could favor a finding that the inactive branch does not have the constitutional authority to do that thing.²⁴⁰Id. at 21–22. For instance, when the Court established the modern anticommandeering rule in Printz v. United States,²⁴¹521 U.S. 898 (1997). the justices reasoned that “if . . . earlier Congresses avoided use of this highly attractive power, we would have reason to believe that the power was thought not to exist.”²⁴²Id. at 905; see also Zivotofsky v. Kerry, 576 U.S. 1, 24 (2015); Free Enter. Fund v. Pub. Co. Acct. Oversight Bd., 561 U.S. 477, 505 (2010); Medellin v. Texas, 552 U.S. 491, 532 (2008). I owe all these citations to Bradley and Siegel. The two authors though do not see a lack of precedent as strong interpretive evidence; rather, they suggest that a stricter test is appropriate when inaction, not action, is marshaled as evidence.²⁴³Bradley & Siegel, supra note 237, at 22. For example, they contemplate “requir[ing] evidence that the inaction has been the result of perceived unconstitutionality.”²⁴⁴Id. at 28 n.114. But there is no evidence that presidents have refrained from issuing dresser drawer pardons because they feared their unconstitutionality.

Let us turn our attention now to the comparative weight given to historical precedent on the one hand and judicial precedent on the other. In the above quote by Frankfurter, he writes as if the only pieces on the board are the text of the Constitution itself, statutes enacted by Congress, and the de facto precedents of the past.²⁴⁵Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 610–11 (1952) (Frankfurter, J., concurring). Of course, in this case, we know that the board also displays Marshall’s opinion in Wilson and the cases that came after it; all of these sources dictate that the pardons clause be understood in light of English law at the time of the Revolution.²⁴⁶See supra Section I.B. If historical and judicial precedent are in conflict, how should we resolve it?

In their recent article, Bradley and Siegel explain that courts are more justified in following historical precedent when they lack judicial precedent, especially in cases involving matters of state where courts usually play a limited role.²⁴⁷Bradley & Siegel, supra note 237, at 24–25. For example, in matters of war and diplomacy, courts are inclined to defer to what has been done consistently and without objection from Congress in the past.²⁴⁸Id. at 25. As further evidence, Bradley and Siegel point to the Court’s approach in Noel Canning, where it fell back on practice only after it found the text on recess appointments ambiguous and judicial precedent vanishingly sparse.²⁴⁹Bradley & Siegel, supra note 237, at 24–25 (quoting NLRB v. Noel Canning, 573 U.S. 513, 526 (2014)).

As we have seen, there is a fair amount of caselaw and treatise writing on the pardons clause: this is not an area where the Justices would be starving for guidance from legal authorities. Moreover, the criminal law is not a land far removed from the competence of the judiciary.²⁵⁰See Craig Hemmens, We [Should] Take Care of Our Own: The Role of Law and Lawyers in Criminal Justice and Criminology Programs, 32 Just. Q. 749, 750 (2015) (“Law is a key component of the criminal justice system. It is a truism that without laws, there would be no crime . . . .”). On the contrary, the criminal law is the domain of the judiciary par excellence,²⁵¹See Craig Hemmens, Teaching Law and Courts in Criminal Justice: Outside Looking In, 27 J. Crim. Just. Educ. 497, 497 (2016) (describing the “significance of law, and by implication the courts wherein the law is enforced” to the criminal justice system). rivaled only by core private law subjects like contracts and torts where the common law courts have been less subject to legislative interference.²⁵²See Sanford H. Kadish, Fifty Years of Criminal Law: An Opinionated Review, 87 Cal. L. Rev. 943, 948–49 (1999) (describing the codification of American criminal law by state legislatures); Jeffrey A. Pojanowski, Private Law in the Gaps, 82 Fordham L. Rev. 1689, 1691 (2014) (describing “tort, contract, and property” as part of the “judiciary’s repository of common law”). We must remember that the ultimate use of a pardon granted in advance of conviction is to be pleaded in court to quash an indictment. As such, courts have been drawn—and will inevitably be drawn—into questions of how to interpret the scope, meaning, and validity of the alleged pardons brought before them. Since they are in their workshop with the tools of their trade ready at hand, courts are not compelled by inexperience to look beyond legal texts to historical practice to decide on the validity of a dresser drawer pardon.

Even if courts looked to historical practice when interpreting the pardons clause, presidents have not been consistent about publicly announcing their grants of clemency. In other words, the historical record is not uniform, and there are at least some cases of presidents issuing pardons in secret. Specifically, President Truman issued pardons that only became known to the public after his successor had taken office and revealed their existence.²⁵³Eisenhower Adopts a New Policy of Complete Publicity on Pardons, N.Y. Times, Feb. 1, 1953, at 1. I am indebted to Albert Alschuler for pointing me to this article. The Eisenhower administration made the pardons known in response to the request of a senator for information on acts of clemency by Truman between election day and Ike’s inauguration.²⁵⁴Id. At the same time he revealed Truman’s pardons, Herbert Brownell, Jr., the new Attorney General, announced that henceforth “the names of persons recommending pardons or commutations granted will be made a matter of public record and the record will be open to inspection by Congress, the press and others who have a legitimate interest.”²⁵⁵Id.

Although these pardons were not publicized, they were not necessarily kept secret from everyone but the president and the recipient either. Many of them pertained to individuals who had already been convicted and would have had little use for pardons kept in dresser drawers that they could not invoke to relieve them of the lingering stigma and disabilities caused by their convictions.²⁵⁶See id. at 47. Furthermore, there presumably was some record of the pardons kept within the executive branch in order for Eisenhower’s new officials to discover the pardons and report them to Congress. Lastly, the fact that Truman’s successor repudiated secret pardons arguably reinforces the norm of publicizing pardons. Nonetheless, it remains the case that Truman’s example shows that there is no unbroken American history of announcing pardons.²⁵⁷But see NLRB v. Noel Canning, 573 U.S. 513, 533 (2014) (explaining that “three-quarters of a century of settled practice is long enough to entitle a practice to ‘great weight in a proper interpretation’ of the constitutional provision.” (quoting The Pocket Veto Case, 279 U.S. 655, 689 (1929))). In 2028, it will have been seventy-five years since Brownell announced the policy of publicizing all pardons.

Conclusion

A.      Practical Implications

Dresser drawer pardons are full of strategic potential for both a president and the recipients of the pardons. Insofar as a president trusts that the courts will honor them, she may be sorely tempted to issue them to furnish peace of mind for herself, her allies, and her loved ones. Should presidents start to give into temptation, and should the pardons ultimately come out of the dressers to forfend indictments, abuses could lead to public outcry and a call for reform by constitutional amendment.

From the perspective of the chief executive, keeping pardons quiet offers a way to avoid or delay embarrassment while still protecting those the president wishes to protect from criminal prosecution and punishment. A pardon may never have to be invoked—prosecutors may decline to act or the crimes of the recipient may never be discovered. Even if prosecutors do ultimately choose to indict a recipient, the prosecution may come sometime after the president leaves office. This is especially true when the president issues pardons during the lame duck. By keeping these secret, the president can avoid exiting on a low note in the media, as happened to Presidents Clinton, Truman, and Trump, who all faced unflattering media coverage of the pardons they issued during the lame duck.²⁵⁸Eisenhower Adopts New Policy of Complete Publicity on Pardons, supra note 253, at 1; Eric Lichtblau & Davan Maharaj, Clinton Pardon of Rich a Saga of Power, Money and Influence, L.A. Times (Feb. 18, 2001, 12:00 AM), https://perma.cc/H9HF-693W; Neal Kumar Katyal, Trump’s Final Pardons Warped Presidential Powers for His Own Benefit, Wash. Post (Jan. 20, 2021, 8:26 AM), https://perma.cc/A6YP-V2TR. Finally, the president could enter into a “gentleman’s agreement” with the recipient of a pardon not to invoke it until after the next election, perhaps even attaching a delay to the pardon so that it only becomes valid after a politically critical date.²⁵⁹There is ample support for the president’s authority to attach conditions to pardons. See Schick v. Reed, 419 U.S. 256, 265 (1974) (“[T]his Court has long read the Constitution as authorizing the President to deal with individual cases by granting conditional pardons.”).

The president is human, and nothing human is alien to him or her. As such, we might expect a president, when feeling faint of heart, to issue some dresser drawer pardons (just in case) to loved ones whom he has no reason to think have committed a federal crime. He could issue pardons to his spouse, children, and best friend to immunize them from liability for all crimes committed prior to the last day of his presidency. He could even try issuing one to himself, though as shown above, the very reasons adduced in this article to support the validity of dresser drawer pardons militate against the validity of a self-pardon.

There are also more statesmanlike reasons that a president might have to issue a pardon in secret. In a time of insurrection—during a revolt by rogue elements in the armed forces for example—a president might secure peace by offering amnesty for all or some rebels and rebel leaders. At the same time, clemency may be deeply unpopular with the citizenry and politically hazardous for the president to deliver. Publicity surrounding pardons might even tempt others to revolt by gentling the prospect of being shot down with hopes for a soft landing. Thus, the ability to issue dresser drawer pardons to induce a mutinous or insurrectionist faction to lay down their arms could be a critical tool of peace and security. Indeed, this could be the use of the pardon power that Hamilton had in mind when he spoke of “critical moments when a well-timed offer of pardon to the insurgents or rebels may restore the tranquility of the commonwealth”²⁶⁰The Federalist No. 74 (Alexander Hamilton). and that Iredell envisioned when he defended vesting the power in the president alone because of “the degree of secrecy and dispatch with which on critical occasions such a power can act.”²⁶¹Iredell, supra note 15, at 351–52. And of course, who would know better than two veteran insurrectionists what might induce rebels to throw down their arms?

The recipient of a dresser drawer pardon may be able to use it as a negotiating tool with prosecutors. By revealing that he possesses a pardon, or implying that he holds one, the target of a federal investigation might dissuade attorneys at the Department of Justice from bringing an indictment that will be immediately quashed. On the other hand, attorneys at the Justice Department may be undaunted, issuing indictments anyway to force the recipient of the pardon to both prove he has it and bring public attention to the choices of the defendant and president to give and accept the pardon.

There are also some major limitations on the utility of dresser drawer pardons. For example, if a person is currently serving a custodial sentence or on probation, the pardon can only help her get out of confinement or supervision if she makes known to the authorities that she has received it. The same is true of those who are currently wanted fugitives, presently in jail awaiting trial, or going through the strain and expense of trial.

B.      Final Apologies

This article may strike some readers as excessively formalistic. The analogy to deeds may seem too light and glib to justify a decision for or against extending the pardon power to secret-, oral-, or self-pardons. Should such a case ever come to the courts, it will likely attract immense public concern for its role in determining the powers of presidents and the fate of prominent people. One might agree with Justice Holmes that pardons ought to be treated as high matters of policy, with the first considerations being the constitutional order and the due allocation of authority within it.

Nevertheless, the pardons clause is one of the places where constitutional law and the criminal law intersect. The validity of a pardon would likely be judicially tested in a criminal prosecution or else a habeas corpus proceeding. The judge hearing the case would then likely have before her an individual who wants to avoid going to prison or leave early. The high and solemn responsibility of deciding whether a person should be made to suffer or go on suffering in prison is then inextricable from the question of constitutional law and presidential authority presented by a novel pardons case.

As Professor Louis Seidman has shown, the criminal law ought to be and has been an area of the law where formalism reigns.²⁶²Louis Michael Seidman, Points of Intersection: Discontinuities at the Junction of Criminal Law and the Regulatory State, 7 J. Contemp. Legal Issues 97, 97–99 (1996). “Criminal law is preoccupied by discourse about rights, fault, consent, and separate private and public spheres.”²⁶³Id. at 97. As the Supreme Court said in apology for the rule of lenity, judges interpreting penal laws are meant to be mindful of “the tenderness of the law for the rights of individuals.”²⁶⁴United States v. Wiltberger, 18 U.S. (5 Wheat.) 76, 95 (1820). The criminal sanction simply “cannot be imposed when doing so violates the criminal’s rights.”²⁶⁵Seidman, supra note 262, at 103.

Moreover, criminal law is designed to narrow the vision of the judge and center her attention on the defendant before her, his actions, and his entitlements under the extant rules. Seidman tells us, “The formal model is individualistic and adjudicative. It focuses attention on the particular defendant before the court and on the legal consequences that ought to flow from a particular set of facts.”²⁶⁶Id. at 106 (emphasis omitted). It follows that a criminal’s rights are invaded “unless the state complies with rules set out in advance that limit its power.”²⁶⁷Id. at 103–04. These precepts include the rules for the Get Out of Jail Free Cards that the Constitution says the president can hand out.

We must remember that the recipient of a controversial pardon is the one coming before the court and insisting that the president’s actions give him or her the right to go free or quash the indictment. The right in question might derive from moth-eaten English books, be described in archaic prose, or depend upon a chain of stylized but nonetheless sound deductive reasoning, and yet the courts should honor it.

To see that this must be so, consider how the courts would deal with a situation that did not implicate the Constitution or matters of state. Suppose a defendant has been charged with a misdemeanor prostitution offense, but the last charged act took place eight months before the prosecution filed charges. Suppose too that the defendant’s attorney finds an 1805 statute, never repealed but little recognized, that establishes a six-month statute of limitations for misdemeanors of “moral turpitude.”

Deciding the case requires more than adhering to the plain text of a law enacted by the legislature long ago. On the contrary, the court will have to consider arguments from both sides about the meaning of the words “moral turpitude”; it will need to study old dictionaries, antique caselaw, and treatises written in nineteenth century legalese. The court may not even agree with the nineteenth century concept of “moral turpitude” and disagree with attaching special, stigmatizing labels to sexual offenses. This does not change the fact that the conscientious judge will do her best to respect and understand the parties’ formalistic arguments about the meaning of the words “moral turpitude” as used in the statute and to rule accordingly. The conscientious judge should do the same when called upon to interpret the pardons clause.