Joshua S. Ha*
“The rule that penal laws are to be construed strictly, is perhaps not much less old than construction itself.” That is how Chief Justice Marshall described the rule of lenity in United States v. Wiltberger. The doctrine is rooted in seventeenth-century England, where it arose to counteract the increasingly widespread imposition of the death penalty for felonies. The rule traveled to America, and today, courts typically justify the rule on three grounds: (1) fair notice to the defendant, (2) separation of powers, and (3) a presumption in favor of liberty.
Today’s rule of lenity is far removed from its English origin. Though it remains a tool of statutory construction, it is now employed “at the end of the process of construing what Congress has expressed,” making it difficult to conceptualize as a principle of strict construction. And even if its utility—as a tiebreaker reserved for instances of “grievous ambiguity”—is extremely limited, it is nonetheless difficult to apply.
This Article argues that we can avoid that difficulty by excluding certain criminal statutes from the rule’s grasp altogether. In particular, this Article claims that courts erred by applying the rule of lenity to the First Step Act—a statute governing whether already-sentenced criminals are eligible for resentencing. Part I discusses the history of the rule of lenity in England and argues that the rule strictly construed criminal statutes to frustrate—not discern—legislative intent. Part II considers the rule of lenity as it is applied today and argues that its placement at the end of the interpretive process is incompatible with the rule as conceived in England. Part III concludes that there are certain criminal statutes to which no rationale for the rule of lenity applies and that the rule of lenity should therefore not be used when interpreting such statutes.
I. A History of Lenity
Courts and scholars generally accept that the rule of lenity arose as a response to the severity of English penal law—and specifically, laws carrying the death penalty. Such laws were commonplace. Sir William Blackstone noted that it was “difficult to justify the frequency of capital punishment to be found therein; inflicted (perhaps inattentively) by a multitude of successive independent statutes, upon crimes very different in their natures.” Although occasionally the relative mercy of “transportation”—i.e., an arrangement by which the criminal would voluntarily leave the country—might excuse a robber from execution, the death penalty remained pervasive.
The rule of lenity took form against this backdrop. It is received wisdom that the rule of lenity can be traced to legal developments surrounding the benefit of clergy. That privilege provided for the “[e]xemption of the persons of clergymen from criminal process before the secular judge” and could be claimed either at the time of arraignment or after conviction. The benefit of the clergy was once limited to those who had the “habitum et tonsuram clericalem”—that is, the “clerical habit and tonsure.” But eventually, the benefit was extended to “every one that could read,” though he be “neither initiated in holy orders, nor trimmed with the clerical tonsure.” Over time, the benefit became widespread, as “learning, by means of the invention of printing, and other concurrent causes, began to be more generally disseminated than formerly.”
Perhaps because it was never meant to extend so far, or perhaps as a favor to clergy dismayed at the loss of this once-exclusive privilege, the benefit of clergy was gradually limited by statute. In the late fifteenth century, a statute was enacted that permitted a layman to use the benefit only once, whereupon he would be branded. And it appears that one of the earlier instances of withdrawing the benefit of clergy from an offense altogether was in 1496, when a statute was passed providing that “if any layperson hereafter . . . murder their lord, master, or sovereign immediate, that they hereafter be not admitted to their clergy.” Instead, that person would “be put in execution as though he were no clerk.” That practice of removing the benefit of clergy from specific offenses continued, until, at the time of Blackstone, 160 felonies were statutorily exempted from the benefit of clergy. For those crimes from which the benefit of clergy was withdrawn, the death penalty was no longer easily avoidable.
The classic narrative is that the rule of strictly construing penal statutes was the courts’ response to the widespread elimination of the benefit of clergy. Whatever the exact mechanics of the rule’s genesis, it was “firmly established” by the mid-seventeenth century. Thus, while the benefit of clergy itself was completely abolished in 1827, the rule of lenity had already taken on a life of its own. But while the rule of lenity, and even its impetus, may be straightforward, its exact justification remained unclear. Was it a good-faith attempt of courts to apply the intent of the legislature, coupled with a disbelief that the legislature could truly desire its strict laws to be liberally applied? Or was it an instance of judicial obstruction, of courts hampering—by technicality—a legislature that meant what it said?
History tends to support the latter. The debate is nicely framed by comparing the two most-cited sources as to lenity’s origins: Sir Peter Benson Maxwell and Professor Livingston Hall. Maxwell was among the first to explicitly link the rule of lenity to the benefit of clergy, and he described the rule of lenity as one faithful to legislative intent. According to Maxwell, the rule was based on the “reasonable expectation that, when the Legislature intends so grave a matter as the infliction of suffering, or an encroachment on natural liberty or rights, or the grant of exceptional exemptions, powers, and privileges,” it “will express [its intention] in terms reasonably plain and explicit.” Thus, in Maxwell’s view, the rule of lenity is also constrained by the legislature’s intent. Maxwell’s lenity did not allow “the imposition of a restricted meaning on the words, for the purpose of withdrawing from the operation of the statute a case which falls both within its scope and the fair sense of its language,” because that “would be to defeat, not to promote, the object of the statute.” In short, “no construction is admissible which would sanction an evasion of an Act.”
Hall thought the rule of lenity served a different purpose. He described the rule as an offspring of a “conflict . . . between the legislature on the one hand and courts, juries, and even prosecutors on the other.” The legislature, either from “inertia” or “pressure from property owners,” pursued “a policy of deterrence through severity,” while the courts “tempered this severity with strict construction carried to its most absurd limits.” In other words, the courts were not using the rule of lenity to determine the legislature’s intent, but rather in direct opposition to whatever policy the legislature was pursuing. Simply, it was a “veritable conspiracy for administrative nullification.”
Hall’s account of the rule of lenity originating as a tool to counteract the legislature’s purpose appears to be the prevailing one. It also seems more historically grounded than Maxwell’s theory of the rule of lenity as a means of effectuating the legislature’s purpose. To start, there is little evidence that the legislature ever intended any of its statutes to be narrowed to oblivion. English legal reformer Samuel Romilly observed that “[t]here probably never was a law made in this country which the legislature that passed it did not intend should be strictly enforced.” For support, Romilly noted that even a strange law “which made it a capital offence for any person above the age of fourteen to be found associating for a month with persons calling themselves Egyptians” was vigilantly enforced “down to the reign of King Charles the first.” According to Romilly, who was writing in 1810, it was “only in modern times that this relaxation of the law has taken place.”
The handful of vignettes from that era also seem to support that the rule of lenity produced results contrary to the legislature’s intent—and, more importantly, that courts were not concerned by that possibility. One example is the courts’ construction of a 1740 statute on cattle-stealing. That statute provided that the stealing of “sheep, or other cattle” was a “felony without benefit of clergy.” As Blackstone tells it, the courts considered the words “or other cattle” to be “much too loose to create a capital offense,” and so “the act was held to extend to nothing but mere sheep.” It is hard to find in such an interpretation an attempt to give effect to the legislature’s intent. Perhaps—depending on what else “cattle” might have entailed back then—the phrase “other cattle” might have been narrowed by the explicit reference to “sheep.” But to render “other cattle” surplusage runs afoul of the spirit of another rule of construction: the command to read a statute in such a way that the “whole may (if possible) stand.” And sure enough, the legislature passed a law the next year “extending the former to bulls, cows, oxen, steers, bullocks, heifers, calves, and lambs, by name.” It appears that the phrase “other cattle” was written, and meant, to be expansive.
A more ambivalent example that was closer in time to the rule of lenity’s inception is the judicial treatment of the horse-stealing felony. During the reign of Henry VIII, a 1545 statute withdrew the benefit of clergy from “the stealing of any Horse Geldinge Mare Foole or Filley.” After Edward VI succeeded to the throne, a new statute was enacted in 1547 that withdrew benefit of clergy from a long list of felonies, including the “felonious stealing of horses geldings or mares.” It also purported to abrogate the 1545 statute, providing that “in all other cases of felony, other than such as be before mentioned,” all persons “shall have and enjoy the privilege and benefit of his or their clergy . . . in like manner and form as he or they might or should have done before the Reign of the said late King Henry the Eighth.” Because the 1545 statute referred to “any Horse” while the 1547 statute referred to horses in plural, Lord Matthew Hale noted that the 1547 statute “made some doubt, whether it were not intended to enlarge clergy, where only one horse was stolen.”
The courts responded by interpreting the 1547 statute to permit the benefit of clergy for stealing a single horse, and the story goes that this is yet another instance of the courts thwarting the legislature’s intent. But that narrative is significantly weaker here. For one, the courts had a textual hook in relying on the plural form of “horses,” “geldings,” and “mares.” And unlike the interpretation of the phrase “other cattle,” here the courts did give effect to the words of the statute (indeed, every letter). Ultimately, though, it appears that the courts once again failed to effectuate the legislature’s intent. The next year, the legislature clarified that “all and singuler p[er]son and p[er]sons felonyously takinge or stealinge any horse geldynge or mare shall not be admytted to have or enjoye the p[ri]viledge or benefyte of his or their Clergy.”
Overall, then, in the absence of a systematic study, it seems that the early uses of the rule of lenity largely led to results counter to the legislature’s intent. Thus, we may consider Maxwell’s conception of lenity as a tool for discerning the legislature’s “true” (as opposed to expressed) intent as a post hoc legitimization of the rule. Such a lens explains why the rule of lenity has survived the abandonment of previous draconian laws and why the rule now applies to laws that do not impose capital punishment. An actual wresting of authority from the legislature to enact criminal punishments as it sees fit can only be justified—if at all—by a penalty as drastic as death. But once the rule was conceived as fitting within the usual judicial system, where legislatures enact and judges faithfully interpret those enactments, it could apply, as Maxwell suggested, regardless of “whether the proceeding prescribed for the enforcement of the penal law be criminal or civil.”
II. Our Rule of Lenity
The rule of lenity thus originated in England. When Chief Justice Marshall described the rule in Wiltberger as being “not much less old than construction itself,” he could only have been incorporating a rule older than the Republic. Wiltberger was the first case in which the rule of lenity was explicitly applied by the United States Supreme Court. It concerned the Crimes Act of 1790. That Act provided for a crime punishable in federal court “[i]f any person or persons, shall commit upon the high seas, or in any river, haven, basin or bay, out of the jurisdiction of any particular state, murder.” In another section, the Act provided for punishment “[i]f any seaman, or other person, shall commit manslaughter on the high seas.” Wiltberger involved manslaughter on a river.
Because the manslaughter provision only referred to the “high seas,” the Court held that “the offence charged in this indictment is not cognizable in the Courts of the United States.” That holding was reached because “Congress has not . . . inserted the limitation of place inadvertently; and the distinction which the legislature has taken, must of course be respected by the Court.” Thus, while the language of the rule of lenity appears throughout the opinion, this case seems much more akin to the “horses” example than the “other cattle” one: a strange and arguably wrong conclusion, but a conclusion that at least has a textual basis. It is fitting that it is unclear how much work this most nebulous doctrine did in the Supreme Court’s first case applying it by name.
Today, our rule of lenity is oft-summarized as “the rule that ambiguity concerning the ambit of criminal statutes should be resolved in favor of lenity.” And whether a statute is ambiguous is determined by using the other methods of statutory interpretation. According to the Supreme Court, “the rule applies if at the end of the process of construing what Congress has expressed, there is a grievous ambiguity or uncertainty in the statute.” It is “reserved for cases where, after seizing every thing from which aid can be derived, the Court is left with an ambiguous statute.” The rule is not one of general strict construction of penal statutes—where “[t]he statute is clear enough,” we do not “rely on the rule of lenity” at all. The rule is, in effect, a tiebreaker.
At the same time, the rule of lenity remains a tool of statutory construction—it is just the last one applied. That is why, where a statute is given a certain meaning on account of lenity, it retains that meaning even in a noncriminal context. In Leocal v. Ashcroft, the Supreme Court held in the immigration context that a DUI was not a “crime of violence” under 18 U.S.C. § 16. Under the Immigration and Naturalization Act, if a DUI were a “crime of violence,” then petitioner Josue Leocal would be deportable. In reaching its conclusion, the Court noted that “[e]ven if § 16 lacked clarity on this point, we would be constrained to interpret any ambiguity in the statute in petitioner’s favor.” Even though this case arose in the immigration context, “[b]ecause we must interpret [§ 16] consistently, whether we encounter its application in a criminal or noncriminal context, the rule of lenity applies.”
That leads to an interesting conundrum where the statute is ambiguous as to a particular criminal defendant but resolving the ambiguity in the defendant’s favor may not benefit criminal defendants generally. The rule would not fit its name if, to avoid a particular result for one criminal defendant, the rule ended up extending to impose penalties on more defendants overall. Some courts of appeals have recognized that the rule of lenity must be applied with an eye toward defendants in general. For instance, in United States v. Olvera-Cervantes, the Ninth Circuit considered the application of U.S.S.G. § 2L1.2, which penalized illegal reentry differently if the previous deportation followed a felony conviction. The question before the court was “whether the district court should look to the maximum penalty authorized by the state statute under which the defendant was convicted or whether it should look to the maximum penalty authorized by the analogous federal statute.” The more favorable outcome to Olvera-Cervantes in particular would have been to look to the federal statute, but the court found that “the rule of lenity . . . is of little use here because we do not know whether the defendant’s interpretation of section 2L1.2 would end up benefitting defendants in general.” Indeed, it is hard to conceive how a court would be able to measure either interpretation’s benefit to defendants, given how federal and state statutes are mutable.
The most straightforward scenario, then, for applying the rule of lenity is when the criminal statute defines conduct and one of the dueling interpretations is narrower than the other. Such a narrow interpretation, completely included within the broader one, will always be beneficial to criminal defendants overall and would not require any hypothesizing by the court. This conception of the rule of lenity—as choosing the narrow over the broad interpretation—resembles strict construction.
But it seems well settled that the rule of lenity comes at the end of the analysis. Only at the end, if two interpretations are in “equipoise,” does the court apply the rule of lenity and therefore choose the narrower construction. Given how few times two interpretations will ever be in true “equipoise,” only rarely will the narrower construction be preferred over the broader one merely by reason of narrowness. That is a sharp contrast to the rule of lenity described in the early English treatises, which treated the rule of strictly construing penal statutes as a general one. In that sense, according to our rule of lenity, penal statutes will rarely be strictly construed.
In sum, the rule of strict construction of penal statutes has survived in American law insofar as the rule of lenity embodies a built-in bias (however slight) for the narrower interpretation. But by giving that bias effect only at the very end of the interpretive process—and only in the limited scenario of “equipoise”—our rule of lenity is significantly less applicable than the historic rule.
III. Lenity’s Limits: A Case Study on the First Step Act
Given the foregoing, applying the rule of lenity is difficult. As a prerequisite to even consider its use in a given case, one must first apply all other tools of statutory construction. Even then, it remains a mystery what counts as “equipoise.” It is therefore unsurprising that in United States v. Hansen, then-Judge Scalia, referring to the rule of lenity, remarked, “It is, quite frankly, difficult to assess the scope of this accepted principle.” Pessimistic as it may sound, these application problems may prove to be intractable. Concepts such as interpretive “equipoise” and narrowing constructions have little capacity to be clarified for easy application.
Looking to the rationales given over time for the rule of lenity might be a decent way to boil down those concepts into something more concrete. Below, this Article considers the three usual rationales provided for the rule of lenity. This Article concludes that most of them are in tension with the origin of the rule and that none tells us when a statute is ambiguous enough that we must make an assumption in favor of the defendant. That leads to the final conclusion of this Article: perhaps the rationales for the rule of lenity do not provide clear rules—but when none of the rationales are applicable, the rule of lenity itself should not apply to the criminal statute.
A. Lenity’s Rationales
Courts and commentators generally give three rationales for the rule of lenity. The first is fair notice; as Justice Holmes observed,
[a]lthough it is not likely that a criminal will carefully consider the text of the law before he murders or steals, it is reasonable that a fair warning should be given to the world in language that the common world will understand, of what the law intends to do if a certain line is passed.
The second is the separation of powers: “[B]ecause of the seriousness of criminal penalties, and because criminal punishment usually represents the moral condemnation of the community, legislatures and not courts should define criminal activity.” And third, “the rule of lenity serves our nation’s strong preference for liberty.”
The first two rationales are inconsistent with the rule of lenity’s origins. To start, the fair notice and separation of powers rationales seem to be post hoc rationalizations. After all, neither logically should have been triggered by the withdrawal of the benefit of clergy. If the scope of conduct criminalized by a seventeenth-century felony was so uncertain as to raise fair notice concerns, that uncertainty would have existed before the legislature decided to make the felony unavoidably punishable by death. Indeed, the vagueness doctrine—another doctrine animated by fair notice—applies beyond the penal context. A similar critique applies to the separation of powers rationale. The legislature is the only branch empowered to enact any statute, not just criminal ones, and not just criminal ones that may impose the death penalty. Furthermore, the separation of powers rationale is a particularly odd fit given that the rule of lenity originated to defeat the legislature’s intent.
The preference for liberty, on the other hand, does match the origin of the rule of lenity, at least superficially. If the death penalty is the most severe deprivation of liberty a state can effect, it makes sense that the rule of lenity only appeared once the benefit of clergy began to be taken away. Underlying this last rationale is a normative assumption—that the criminal penalty is a severe sanction, and that the state must therefore speak clearly if it wants to deprive a citizen of liberty. We might question whether those assumptions hold in the context of many cases where the rule of lenity is applicable; if a defendant had committed some sort of morally culpable act (even if that act is not clearly cognized by the criminal statute), a reasonable person might consider it unfair to add another thumb to the scale in the defendant’s favor. In any event, this rationale is also a relatively good fit with the practice of applying lenity. As Maxwell described the rule in England, “[t]he degree of strictness applied to the construction of a penal statute depends in great measure on the severity of the statute.” A few states––and federal courts now and then––seem to do the same when they differentiate between felony statutes and misdemeanor statutes.
But the liberty rationale too is imperfect. If the liberty rationale is meant to require legislatures to speak clearly, why does lenity only apply when there is grievous ambiguity, as opposed to just ambiguity? More importantly, a sliding scale of lenity adds yet another variable to an already-unclear equation. How much stricter should a felony statute be interpreted? What about a statute that provides a ten-year maximum sentence versus a fifteen-year maximum sentence? The liberty rationale’s capaciousness makes it a decent justification for the rule. But it is also a poor guiding principle for the rule’s application.
One might ask, why should we care whether the rationales for the rule of lenity—all of which are widely cited—are consistent with the rule’s origin? After all, that a rule sprung up in response to a particular confluence of events does not necessarily mean that we should keep the rule bound to that scenario. Be that as it may, Wiltberger itself described the rule of lenity as not just merely old, but “perhaps not much less old than construction itself,” and courts have consistently picked up on and repeated that phrase. It would be odd to abandon any attempt to keep lenity moored to its past, considering the courts’ constant reminders of its ancient roots.
In any event, at the very least, it seems that no single rationale can claim supremacy over the other. And even if one could, each rationale alone does not shed much light on how to apply the rule to a particular statute. At what point is a statute ambiguous enough to raise fair notice concerns? Our other doctrine that responds to the need for fair notice relies on such concepts as what a “person of ordinary intelligence” could “reasonably understand”—hardly a hopeful ground upon which to find a clear rule. As for the separation of powers rationale, scholars have noted that courts are accepted to have, in certain criminal matters, vast discretion. Thus, if violating the separation-of-powers principle by giving too much discretion to courts in criminal matters is merely a matter of degree, that rationale is also unlikely to lead to any clear rule.
None of this is to suggest that a lack of a clear rule means that lenity should be discarded altogether. Law does not always give clear rules, and it might be especially odd to require clarity from a doctrine designed to enter the legal analysis in response to ambiguity. Just because a doctrine cannot be reduced into bright-line rules does not make it invalid—far from it. Rather, this Article concludes only that, insofar as this Article attempts to find a clear way to apply lenity in at least some cases, there probably will be none based on balancing various of the three rationales for the rule.
B. A Proposed Limit on Lenity
This Article argues that there may be some categories of penal laws to which none of the rationales of the rule of lenity apply, and that lenity should therefore be inapplicable to those statutes. The rule of lenity has generally been described in terms referring to “penal laws,” without any suggestion that some penal laws might not be proper subjects of the rule. But considering the rule of lenity’s rationales in order to define a class of statutes to which the rule does not apply is not unheard of. Emlin McClain, former Chief Justice of the Iowa Supreme Court, in a late-nineteenth century treatise of American criminal law, noted the view that, because the rule “was adopted at the common law in favor of life, or the liberty of the citizen,” it “has never been observed in the construction of statutes enacted for the punishment of mere misdemeanors.” For that reason, McClain described several categories of criminal law to which the rule was not applied, including “statutes for the prevention of fraud and suppression of public wrong” and “statute[s] relating to procedure.”
With that in mind, one potential limit on the scope of lenity starts with the observation that every rationale for lenity is, in effect, a prohibition on what the legislature can do. The legislature may not criminalize conduct without providing fair notice to potential defendants. The legislature may not pass such an open-ended statute that it effectively delegates lawmaking to the courts. The legislature may not infringe upon a person’s liberty without clearly stating its intent to do so. Put another way, there are limits to how a legislature may enact a statute used to the detriment of the defendant. If those are fair characterizations of the rationales underlying the rule of lenity, then the rule should not be applied in instances of legislative grace.
One recent example of so-called legislative grace is the First Step Act of 2018. Before 2010, distributing 5 and 50 grams of crack triggered mandatory minimum sentences of five and ten years, respectively. The Fair Sentencing Act of 2010 raised those threshold amounts to 28 and 280 grams. Eight years later, the First Step Act made already-sentenced criminals potentially eligible for the lower penalties of the Fair Sentencing Act, even though they had committed their crimes before the Fair Sentencing Act was passed. Eligibility for relief under the First Step Act depended on whether the criminal had committed a “covered offense.” And a “covered offense” was defined in the First Step Act as “a violation of a Federal criminal statute, the statutory penalties for which were modified by section 2 or 3 of the Fair Sentencing Act of 2010, that was committed before August 3, 2010.”
Predictably, the definition of “covered offense” led to questions, often raised by criminals seeking resentencing. One of the most significant questions was whether the term referred to the actual conduct that the criminal committed or merely the statutory elements of the offense. That distinction matters for someone who distributed, say, a kilogram of crack. If actual conduct mattered, then he would not be eligible for resentencing, because someone who sold a kilogram of crack today would be subject to the same penalties as someone who did so before 2010. On the other hand, if only the statutory elements mattered, then the criminal would be eligible for resentencing because the penalty for selling fifty grams of crack has been modified.
From the beginning, most district courts adopted the categorical approach. For many of these courts, whether actual conduct mattered depended on what the phrase “statutory penalties for which were modified” was in reference to and what the term “violation” meant. And time and time again, those courts would invoke the rule of lenity to say that whether a covered offense was committed did not depend on the actual underlying conduct. Even district courts that reached the conclusion that eligibility for resentencing depended on actual offense conduct dismissed the use of lenity because the statute was unambiguous, not for some other reason.
This Article argues that the rule of lenity should never have applied in construing eligibility under the First Step Act—even if there were a “grievous” ambiguity—because none of the rationales for the rule apply here. To begin with, the fair notice concern is irrelevant in this context. Whatever applicability that doctrine has in the sentencing context in general, it has no relevance when discussing a criminal statute that alters the penalties for a crime after the sentence has already been handed down. To say that fair notice concerns are implicated here would be to say that the criminal should be given the benefit of the doubt in case he was misled by the ambiguous wording of a statute that had not yet been passed.
Neither does it make sense to apply the rule in the name of separation of powers. In fact, that rationale would suggest that the court should resolve ambiguities against the criminal. Sentence modifications “are not constitutionally compelled,” and courts themselves “lack the inherent authority to modify a term of imprisonment.” If it is in Congress’s domain, then, to permit resentencing, the separation of powers principle would have courts decline—not expand—the invitation to find the ability to do so unless Congress spoke clearly.
Even the vague preference for liberty rationale is especially weak here because the legislature did speak clearly under the old sentencing regime. As the Eighth Circuit noted in analyzing a different sentence modification statute, “[n]o new deprivation of liberty can be visited upon [a prisoner] by a proceeding that, at worst, leaves his term of imprisonment unchanged,” when “[h]is liberty is already deprived by virtue of a sentencing which gave him all the process the Constitution required.”
Looking forward, the same analysis may apply—but with less weight—when the statute or rule providing a potential resentencing was in place before the criminal’s initial sentence. For instance, in United States v. Puentes, the Eleventh Circuit considered whether a district court could reduce a defendant’s obligation to pay restitution under the Mandatory Victim’s Restitution Act (“MVRA”) through Federal Rule of Criminal Procedure 35(b), which provides that “the court may reduce a sentence if the defendant, after sentencing, provided substantial assistance in investigating or prosecuting another person.” The MVRA was enacted after Rule 35(b), and both were in place well before Puentes committed his crimes. Puentes argued that “[the court is] bound to apply the rule of lenity if [it] find[s] any ambiguity in the [MVRA], Rule 35(b), or the interplay between the two.” The court assumed for the sake of argument that the rule of lenity could apply to Rule 35(b) but decided that there was no grievous ambiguity that would support applying lenity there anyway.
Putting aside whether lenity should apply to procedural provisions at all, it is a harder question whether lenity should apply here compared to the First Step Act example analyzed above. Unlike a defendant under the First Step Act, Puentes in theory could have depended on the possibility of a Rule 35(b) sentence reduction at the time of his conduct. It is true that the fair notice concern is already weak in the sentencing context and the resentencing context is even a step further removed, but at least it would be temporally possible for a defendant to be misled by the wording of Rule 35(b). It also might make some sense under the framework mentioned above that views lenity as a constraint upon Congress: if Congress wants to deny an avenue for relief available to potential criminals, it must speak clearly. On the other hand, Rule 35(b) remains an exception to the usual inability to change an already-imposed sentence. In recognition of that general rule, courts often refer to 18 U.S.C. § 3582(c)(2)—which authorizes sentence-modification proceedings—as an “act of lenity.” If we consider Rule 35(b) itself to be a similar “act of lenity,” to apply the rule of lenity to it almost seems like double-counting.
So, it may not be as easy as saying all resentencing statutes should be excluded from the rule of lenity. But at the very least, an approach of narrowing which criminal laws are subject to our rule would be theoretically grounded and simple to administer. Unlike the usual approach, where courts always must consider “how much ambiguousness constitutes an ambiguity,” once a particular statute is found to be outside lenity’s ambit, a court will no longer need to engage in that last step of the interpretive process at all.
The rule of lenity has largely been unmoored from its English origins. And though it is an old doctrine that has rarely been questioned, its inconsistent application has prompted even Justice Scalia to suggest that “[i]f [the rule of lenity] is no longer the presupposition of our law, the Court should say so, and reduce the rule of lenity to an historical curiosity.” Even if a court seeking to apply the rule were to look to what rationales have been used to justify the modern American version of lenity, it would be confronted with three—all distinct, and none perfect.
It seems, then, that the easiest way to clarify the rule of lenity’s application is to start with considering its scope. Although the rule is taken to mean that “ambiguity concerning the ambit of criminal statutes should be resolved in favor of lenity,” this Article argues that the rule’s reference to “criminal statutes” should not be taken hyperliterally. Especially now that the term “criminal statutes” may fairly be read to mean any statute that touches upon criminal law, the rule of lenity—even in its expanded form—does not reach every such statute. Where none of the rationales for the rule of lenity apply, that old doctrine should have no role to play in interpreting the statute, atmospheric or otherwise.
*. Law clerk to Judge Steven J. Menashi, U.S. Court of Appeals for the Second Circuit, 2021–22, and to Judge Britt C. Grant, U.S. Court of Appeals for the Eleventh Circuit, 2020–21. J.D., Harvard Law School, 2020; M.A., State University of New York at Stony Brook, 2016; B.S., State University of New York at Stony Brook, 2015. Thanks to Aaron Gyde, Joshua Hoyt, Aaron Hsu, and DJ Sandoval for helpful comments and advice. This Article represents the views of the author alone.
. Frederick Howard Wines, Punishment and Reformation: A Study of the Penitentiary System 106 (1910) (“[M]ultitudes of prisoners under sentence of death were given the alternative, of which they hastened to take advantage, of voluntarily leaving the realm, if pardoned. Herein was the germ of English transportation.”).
. 4 Blackstone, supra note 9, at 18 (“[I]n England, besides the additional terrors of a speedy execution, and a subsequent exposure or dissection, robbers have a hope of transportation, which seldom is extended to murderers.”). In fact, this differing punishment for crimes earned commendation from Blackstone, who remarked that “[w]here men see no distinction made in the nature and gradations of punishment, the generality will be led to conclude there is no distinction in the guilt.” Id.
. 4 Blackstone, supra note 9, at 358. As a historical matter, then, the benefit of clergy was usually claimed after conviction. As Blackstone observed, “it is more to the satisfaction of the court to have the crime previously ascertained by confession or the verdict of a jury; and also it is more advantageous to the prisoner himself, who may possibly be acquitted, and so need not the benefit of his clergy at all.” Id. at 359–60; see also McRaney v. N. Am. Mission Bd. of S. Baptist Convention, Inc., 980 F.3d 1066, 1076 (5th Cir. 2020) (Oldham, J., dissenting from the denial of rehearing en banc).
. 4 J. W. Jones, A Translation of All the Greek, Latin, Italian, and French Quotations Which Occur in Blackstone’s Commentaries on the Laws of England 245 (1823). The clerical tonsure was a hairstyle. See 4 Blackstone, supra note 9, at 360.
. 4 Blackstone, supra note 9, at 360. It appears that this literacy test was a judicial misinterpretation of the legislature’s extension of the benefit to “secular” clerks. See Phillip M. Spector, The Sentencing Rule of Lenity, 33 U. Tol. L. Rev. 511, 515 n.22 (2002) (“[That] statute was intended to clarify that benefit of clergy would be afforded to ‘inferior Orders’ of the clergy, as well as bishops, priests, and deacons, but . . . the intent was not to extend clergy to lay persons.” (quoting 2 Sir William Hawkins, A Treatise of the Pleas of the Crown 338 (2d ed. 1724))). That misinterpretation was arguably deliberate. See Romantz, supra note 3, at 526 (“As Parliament and the king continued to proliferate capital felonies in the coming centuries, the courts responded by expanding the benefit of the clergy rule to include any citizen who could read.”).
. William L. Rushton, Shakespeare: A Lawyer 16 (1858) (“[A]s many laymen as clergymen enjoyed this privilege, which excited the jealousy of the clergy, in whose favour, therefore, a further distinction was made . . . .”).
. That statute was titled “Concerning the allowance of benefit of clergy,” and it provided that “every person, not being within orders, which once hath been admitted to the benefit of his clergy, eftsoons arraigned of any such offence, be not admitted to have the benefit or privilege of his Clergy.” 4 Hen. 7 c. 13.
. 12 Hen. 7 c. 7; see Spector, supra note 15, at 515–16 (“At first, benefit of clergy was stripped from murder and certain particularly nasty cases of robbery, but by the middle of the sixteenth century benefit of clergy had been withdrawn from the most trivial of felonies, including ‘stealing horses,’ pickpocketing, and ‘burning a dwelling or barn having grain therein.’” (footnotes omitted) (citing 12 Hen. 7 c. 7)).
. 4 Blackstone, supra note 9, at 18 (“It is a melancholy truth, that among the variety of actions which men are daily liable to commit, no less than a hundred and sixty have been declared by act of parliament to be felonies without benefit of clergy; or, in other words, to be worthy of instant death.”). To be sure, “a large number of capital offenses on the statute book is no test for severity,” but it remains the case that “by the nineteenth century, the government had so limited the [benefit of clergy] that it had fallen into disuse.” Newman F. Baker, Benefit of Clergy—A Legal Anomaly, 15 Ky. L.J. 85, 111 (1927).
. Livingston Hall, Strict or Liberal Construction of Penal Statutes, 48 Harv. L. Rev. 748, 750 n.13 (1935) (“By the time Hale wrote (he died in 1676), the doctrine of strict construction was firmly established.”); see also 1 William Blackstone, Commentaries on the Laws of England 88 (4th ed. 1765) (“Penal statutes must be construed strictly.”); 2 Sir Matthew Hale, Historia Placitorum Coronae: The History of the Pleas of the Crown 335 (1736) (“That where any statute . . . hath ousted clergy in any of those felonies, it is only so far ousted, and only in such cases and as to such persons, as are expressly comprised within such statutes, for in favorem vitae & privilegii clericalis such statutes are construed literally and strictly.”).
. Spector, supra note 15, at 514 n.16 (“Sir Peter Benson Maxwell and [Livingston] Hall were the first to trace the rule of lenity back to the benefit of clergy cases. Their accounts have been accepted and recited by modern rule of lenity scholars.” (citations omitted)).
. Id. Perhaps this state of affairs—presumably not all too displeasing to the legislature, which could reap the benefits of passing such statutes without any of the political downsides—could eventually lead to a “sort of prescriptive validity.” See Antonin Scalia, Assorted Canards of Contemporary Legal Analysis, 40 Case W. Rsrv. L. Rev. 581, 583 (1989). After the charade has gone on long enough, “the legislature presumably has [it] in mind when it chooses its language.” Id. But that does not clarify the justification for the rule of lenity at its inception.
. See, e.g., John Calvin Jeffries, Jr., Legality, Vagueness, and the Construction of Penal Statutes, 71 Va. L. Rev. 189, 198 (1985) (“Faced with a vast and irrational proliferation of capital offenses, judges invented strict construction to stem the march to the gallows.”); Zachary Price, The Rule of Lenity as a Rule of Structure, 72 Fordham L. Rev. 885, 897 (2004) (“The rule of lenity has its oldest origins in the efforts of common law courts in the seventeenth and eighteenth centuries to limit the brutality of English criminal law.”); Romantz, supra note 3, at 527 n.12 (“The rule of lenity first developed in England with the decided goal of frustrating the intent of the legislature. English courts resolved to chart a more humane path despite the legislature’s facility to enact capital crimes.”); Lawrence M. Solan, Law, Language, and Lenity, 40 Wm. & Mary L. Rev. 57, 87 (1998) (“The courts, doing what they could to frustrate the legislative will, developed the principle that penal statutes were to be construed strictly.”). In fact, as mentioned above, the legislature’s abrogation of the benefit of clergy might have itself been a move in this tug-of-war, a response to the courts’ improper extension of the benefit to all literate citizens in the first place. See Romantz, supra note 3, at 526–27 (“Keenly aware that the courts were frustrating its legislative prerogative to kill the nation’s criminals, Parliament responded by enacting more and more capital felonies, while excluding increasing numbers of felonies from the benefit of the clergy.”).
. Id.; see also id. at 6 (“In the long and sanguinary reign of Henry VIII, it is stated by Hollinshed that 72,000 persons died by the hands of the executioner, which is at the rate of 2,000 in every year.”). To be sure, Romilly had an agenda of his own, as an advocate for repealing the death penalty for various crimes. See generally Charles Noble Gregory, Sir Samuel Romilly and Criminal Law Reform, 15 Harv. L. Rev. 446 (1902) (providing an example of Romilly’s advocacy against the death penalty).
. Id. Among others, Blackstone and Justice Scalia have referenced the judicial treatment of this statute, the latter somewhat scornfully. See Scalia, supra note 32, at 582 (“I doubt, for instance, that any modern court would go to the lengths described by Blackstone in its application of the rule that penal statutes are to be strictly construed.”).
. Other felonies from which the benefit of clergy was withdrawn was the “robbing of anny Parsone or parsons in the highe waye or nere to the highe waye,” and the “felonious taking of anny good out of anny pishe Churche or other Churche or Chapell.” 1 Edw. VI c. 12. That the statute made the distinction between “Parson” and “parsons” might have also cast doubt on whether the term “horses” should be read to also include a single horse.
. 2 & 3 Edw. VI c. 33. But even then, the legislature seemed to admit that the initial statute was unclear. The preface to its clarifying act stated that it was made necessary because “it is and hathe been ambyguous and doubtfull . . . whether that any p[er]son being in due fourme of the lawes found gyltye . . . [of] felonyous stealinge of one horse geldynge or mare ought to be admytted to have and enjoye the priviledge and benefyte of his Clergie . . . .” Id.
- Dan M. Kahan, Lenity and Federal Common Law Crimes, 1994 Sup. Ct. Rev. 345, 357 (1994). That statute, according to Professor Kahan, was “the very first piece of criminal legislation enacted by Congress.” Id.
. Shaw v. United States, 137 S. Ct. 462, 469 (2016) (internal quotation marks and citation omitted); see also United States v. Shabani, 513 U.S. 10, 17 (1994) (“The rule of lenity, however, applies only when, after consulting traditional canons of statutory construction, we are left with an ambiguous statute.”)
. See Reno v. Koray, 515 U.S. 50, 65 (1995) (“The rule of lenity applies only if, after seizing everything from which aid can be derived, we can make no more than a guess as to what Congress intended.” (internal quotation marks and citation omitted)); see also Brannon, supra note 65, at 31 n.317 and accompanying text (“Consequently, most courts will not apply the substantive canons [such as the rule of lenity] unless they conclude that after consulting other interpretive tools, the statute remains ambiguous.”).
. Id. at 11 n.8. This discussion of the rule of lenity in Leocal is dicta. Clark v. Martinez, 543 U.S. 371, 397 (2005) (Thomas, J., dissenting). But from the viewpoint of the lower courts, “there is dicta and then there is dicta, and then there is Supreme Court dicta.” Schwab v. Crosby, 451 F.3d 1308, 1325 (11th Cir. 2006).
. Leocal, 543 U.S. at 11 n.8; see also Kasten v. Saint-Gobain Performance Plastics Corp., 563 U.S. 1, 16 (2011) (“[W]e have said that the rule of lenity can apply when a statute with criminal sanctions is applied in a noncriminal context.”); Martinez, 543 U.S. at 380 (“It is not at all unusual to give a statute’s ambiguous language a limiting construction called for by one of the statute’s applications, even though other of the statute’s applications, standing alone, would not support the same limitation.”).
. Id. at 103; see also United States v. Beck, 957 F.3d 440, 450 (4th Cir. 2020) (noting that “lenity doesn’t support [Beck’s] interpretation,” because his interpretation “would generally be against defendants’ interests”).
. See Shular v. United States, 140 S. Ct. 779, 787 (2020) (Kavanaugh, J., concurring) (“[A] court may invoke the rule of lenity only after consulting traditional canons of statutory construction.” (internal quotation marks and citation omitted)); id. at 787 n.1 (listing cases).
. See id. (noting that “the Supreme Court’s advice that it only serves as an aid for resolving an ambiguity . . . provides little more than atmospherics, since it leaves open the crucial question—almost invariably present—of how much ambiguousness constitutes an ambiguity”).
. United States v. Bass, 404 U.S. 336, 348 (1971); see also United States v. Wiltberger, 18 U.S. (5 Wheat.) 76, 95 (1820) (“The rule that penal laws are to be construed strictly . . . is founded . . . on the plain principle that the power of punishment is vested in the legislative, not in the judicial department.”).
. United States v. Nasir, 17 F.4th 459, 473 (3d Cir. 2021) (Bibas, J., concurring). On this point, courts repeatedly cite to Judge Henry Friendly’s observation of “our ‘instinctive distaste against men languishing in prison unless the lawmaker has clearly said they should.’” Id. (quoting Henry J. Friendly, Mr. Justice Frankfurter and the Reading of Statutes, in Benchmarks 196, 209 (1967)); see, e.g., United States v. R.L.C., 503 U.S. 291, 305 (1992) (plurality opinion); United States v. Pembrook, 609 F.3d 381, 391 (6th Cir. 2010); Sash v. Zenk, 439 F.3d 61, 65 n.2 (2d Cir. 2006); United States v. Latimer, 991 F.2d 1509, 1514 (9th Cir. 1993).
. See Marinello v. United States, 138 S. Ct. 1101, 1108 (2018) (explaining that if Congress had intended to make minor violations of tax laws a felony, it would have used clearer language to do so).
Two reasons can be found for the decline in importance of the lenity canon. First, the criminal law has been used more and more, not just to condemn evil behavior, but to regulate economic activity. Jail sentences and stigmas are less likely to attach, either by law or in practice. In that setting, a generalized tilt toward the accused loses some of its attraction. Second, as public concern about crime increases, the inclination to adopt an across-the-board presumption in favor of the accused weakens.
William D. Popkin, Statutes in Court: The History and Theory of Statutory Interpretation 204 (1999).
. See, e.g., Maine v. Millett, 203 A.2d 732, 734 (Me. 1964) (quoting Maine v. Blaisdell, 105 A. 359, 360 (Me. 1919)) (noting that a statute declaring a felony “calls for a more strict construction than one which declares an act to be a misdemeanor”); Mo., K. & T. Ry. Co. v. State, 100 S.W. 766, 767 (Tex. 1907) (“It is a well-established principle of statutory construction that penal statutes must be strictly construed in determining the liability of the person upon whom the penalty is imposed, and the more severe the penalty, and the more disastrous the consequence to the person subjected to the provisions of the statute, the more rigid will be the construction of its provisions in favor of such person and against the enforcement of such law.”). In United States v. Plaza Health Laboratories, 3 F.3d 643 (2d Cir. 1993), the Second Circuit declined to construe the Clean Water Act (“CWA”) in the same way that it did the Rivers and Harbors Act (“RHA”). Id. at 647–48. The court “view[ed] with skepticism the government’s contention that [it] should broadly construe the greatly magnified penal provisions of the CWA based upon RHA cases that did so in the context of strict-liability and misdemeanor penalties.” Id. at 648. See generally 1 Wayne R. LaFave, Substantive Criminal Law § 2.2(d) (3d ed. 2021) (footnote omitted) (citing Millett, 203 A.2d 732) (“No doubt some criminal statutes deserve a stricter construction than others. Other things being equal, felony statutes should be construed more strictly than misdemeanor statutes; those with severe punishments more than those with lighter penalties.”).
. See, e.g., Massachusetts v. St. Hilaire, 21 N.E.3d 968, 979 (Mass. 2015) (quoting McBoyle v. United States, 283 U.S. 25, 27 (1931)); United States v. Gallaher, 624 F.3d 934, 941(9th Cir. 2010) (quoting United States v. Bass, 404 U.S. 336, 348 (1971)); Sash v. Zenk, 439 F.3d 61, 65 n.2 (2d Cir. 2006) (quoting Bass, 404 U.S. at 348)).
. See, e.g., Dowling v. United States, 473 U.S. 207, 213 (1985); United States v. Boston & M. R.R., 380 U.S. 157, 160 (1965) (“A criminal statute is to be construed strictly, not loosely. Such are the teachings of our cases from United States v. Wiltberger down to this day.” (citation omitted)); United States v. Canelas-Amador, 837 F.3d 668, 674 (6th Cir. 2016); United States v. Valle, 807 F.3d 508, 527 (2d Cir. 2015); United States v. Parker, 762 F.3d 801, 807 (8th Cir. 2014); United States v. Winchester, 916 F.2d 601, 607 (11th Cir. 1990) (“Lenity, the quality of being lenient or merciful, is an application of the common law principle that criminal statutes are to be strictly construed, a rule which ‘is perhaps not much less old than construction itself.’” (quoting Wiltberger, 18 U.S. (5 Wheat.) at 95)).
. See United States v. Komzinski, 487 U.S. 931, 951 (1988) (denouncing “the arbitrariness and unfairness of a legal system in which the judges would develop the standards for imposing criminal punishment on a case-by-case basis”).
. See United States v. King, 423 F. Supp. 3d. 481, 484 (M.D. Tenn. 2019) (“To date, it does not appear that any Court of Appeals has weighed in on the issue, but the vast majority of district court[s] to have addressed the matter have concluded that the count of the conviction controls . . . .”).
. See, e.g., United States v. Henderson, 399 F. Supp. 3d 648, 653–54 (W.D. La. 2019). The Supreme Court has since clarified that the phrase “statutory penalties” refers to “a violation of a Federal criminal statute.” Terry v. United States, 141 S. Ct. 1858, 1862 (2021).
. See Henderson, 399 F. Supp. 3d at 654; see also King, 423 F. Supp. 3d at 484–85; United States v. Hardnett, 417 F. Supp. 3d 725, 737 (E.D. Va. 2019); United States v. Willis, 417 F. Supp. 3d 569, 575 (E.D. Pa. 2019); United States v. Williams, 402 F. Supp. 3d 442, 448 (N.D. Ill. 2019); United States v. Askins, No. CR-02-00645-001, 2019 WL 3800227, at *3 (D. Ariz. Aug. 6, 2019); United States v. White, No. 99-CR-628-04, 2019 WL 3228335, at *4 (S.D. Tex. July 17, 2019); United States v. Martin, No. 03-CR-795, 2019 WL 2571148, at *2 (E.D.N.Y. June 20, 2019); United States v. Rose, 379 F. Supp. 3d 223, 229 (S.D.N.Y. 2019); United States v. Allen, 384 F. Supp. 3d 238, 242 (D. Conn. 2019); United States v. Pierre, 372 F. Supp. 3d 17, 22 (D.R.I. 2019).
. See, e.g., United States v. Jackson, No. 03-0642, 2019 U.S. Dist. LEXIS 109993, at *7–8 n.3 (E.D. Pa. June 26, 2019) (“To the extent the Government suggests that the meaning of ‘violation’ in § 404(a) is ambiguous, the Court disagrees. As a result, the Court’s interpretation of ‘violation’ in the First Step Act is not subject to the rule of lenity, which requires courts to construe ambiguities in criminal statutes in favor of defendants.” (citation omitted)).
. United States v. Jones, 962 F.3d 1290, 1297 (11th Cir. 2020); see also United States v. Puentes, 803 F.3d 597, 605–06 (11th Cir. 2015) (“The law is clear that the district court has no inherent authority to modify a sentence; it may do so only when authorized by a statute or rule.”).
. Johnson v. United States, 576 U.S. 591, 630 (2015) (Alito, J., dissenting) (noting that fair notice concerns “have less force when it comes to sentencing provisions, which come into play only after the defendant has been found guilty of the crime in question”).
. Dillon v. United States, 560 U.S. 817, 828 (2010); see also United States v. Padilla-Diaz, 862 F.3d 856, 861 (9th Cir. 2017) (“As acts of lenity, such sentence reductions are not constrained by the general policies underlying initial sentencing or even plenary resentencing proceedings.”); United States v. Maiello, 805 F.3d 992, 1000 (11th Cir. 2015); United States v. Johnson, 703 F.3d 464, 469 (8th Cir. 2013).
. Holloway v. United States, 526 U.S. 1, 21 (1999) (Scalia, J., dissenting). This statement is especially striking from Justice Scalia, who along with Bryan Garner has been credited with maintaining the rule of lenity’s significance in our law. See Intisar A. Rabb, Response, The Appellate Rule of Lenity, 131 Harv. L. Rev. F. 179, 180 (2018) (“Justice Scalia and Professor Bryan Garner have helped elevate the rule of lenity by including it in a set of fifty-seven recommended canons of construction in their widely read treatise on interpretation.”).