Current Issue Archive  Submissions Subscribe About BJCL Contact Us Volume Five Index

Volume Five Masthead


Jamie Blain

John Kelly


Managing Editor
Jennifer Schlotterbeck


Executive Articles Editor

Scott Bennett


Submissions Editor

Judy An


Articles Editor

Austin Ku


Web Editor

John Kelly


Assistant Editors

Sloan Johnston

Mary Dabiza

Sasha Abrams

Craig Sieverding



Shelia Quinlan

Vishal Sahni

Maya Sinha

Tim Aoyagi

Benjamin Kim

San Won Lee

Michele Casey

Eric Broxmeryer

Meredith Jones

Elizabeth Zirker

Julia Clayton

Jordan Cunningham


Watching Legislatures for Apprendi’s Effects on Plea Bargaining

Darryl K. Brown*

Cite as 5 Cal. Crim. Law Rev. 1
Pincite using paragraph numbers, e.g. 5 Cal. Crim. Law Rev. 1, ¶11

     I. Introduction

¶1  In Apprendi v. New Jersey,[1] decided just two years ago, the United States Supreme Court held that any facts that work to increase a criminal defendant’s sentence above the statutory maximum must be treated as elements of the crime, and thus determined by juries, rather than as sentencing factors to be determined by the judge.  With this decision, the Court purported to restrict legislatures’ ability to aid law enforcement in circumventing constitutional criminal procedure rules like the criminal burden of proof--thereby aiding criminal defendants.  With a recent article in the Yale Law Journal, Stephanos Bibas joined other scholars in observing how easily legislatures can rewrite statutes to avoid Apprendi’s prosecutorial and legislative restrictions, if indeed they are substantive restrictions at all.[2]  Uniquely and provocatively, Bibas then argued that Apprendi, as a practical result of its interaction with the other rules and incentives of criminal litigation, will actually hurt criminal defendants, by shifting more power to prosecutors.  Bibas argued that Apprendi acts to deprive defendants, who overwhelmingly plead guilty rather than face jury trials, of the only meaningful, real-world hearings they are likely to receive--judicial sentencing hearings.[3]

¶2  Scholars have already responded to and challenged Bibas’s assumptions and predictions about how Apprendi will affect the real-world strategic behavior of both prosecutors and defendants.  In this article, I want to discuss why it will be the legislative reaction to Apprendi, circumvention or not,that will prove or disprove the accuracy of Bibas’s theses.  Over the last four decades, legislatures have responded, often indirectly but still very effectively, to a broad range of judicial criminal procedure decisions that have attempted to impede law enforcement in various ways.  If Bibas’s analysis is accurate, and Apprendi in fact acts to significantly disadvantage criminal defendants, legislators will have little incentive to circumvent it or otherwise to effect statutory change as a result, no matter how easy such circumvention would be.  If however, contrary to Bibas’s predictions, Apprendi in practice does afford defendants its intended benefit (or otherwise increases the price of adjudication), legislators will revise statutes to avoid its implications. 

     II. Apprendi’s Holding

¶3  In Apprendi, the Court focused on a New Jersey statute that subjected offenders in a wide range of crimes to higher sentences if their crimes were committed with a racial motive.  The statute, however, did not make racial motive an element of Apprendi’s crime  (Apprendi pled to weapons possession counts), or of any of the crimes to which the sentencing enhancement applied; it made racial motive a mere sentencing factor.[4]  This allowed prosecutors to prove the fact to the judge instead of a jury, by preponderance of the evidence instead of beyond a reasonable doubt. 

¶4  In the Court’s view, the New Jersey legislature had improperly labeled a substantive element of the crime as a sentencing factor, making it functionally a civil component of adjudication.[5]  With Apprendi, the Court attempted to force legislatures to treat facts that affect punishment as crime elements, and thus subject to criminal procedure rules, rather than as functionally civil sentencing factors.  However, Apprendi is, on its face, a weak constraint on legislatures.  The Court left open the option for legislatures to raise maximum sentences for lower grades of an offense, allowing judges to distinguish among offenders by using mitigating factors to lower the inflated sentences rather than using aggravating factors to raise them above their maximums.  Unless and until the Court expands Apprendi, it stands as an example of a larger problem of courts’ unwillingness to regulate the legislative definition of substantive criminal law.

     III. Legislative Responses to Procedural Entitlements

¶5  Only by constraining the definition of crimes can courts check legislative efforts to undermine the criminal procedure rules that courts, through the Constitution, impose on police and prosecutors.  Criminal law scholars will recognize this observation of the link between substantive criminal law and constitutional criminal procedure as the central insight that William Stuntz has brought to attention in recent years, and extended in an essay last year.[6] Stuntz has lamented, as have other scholars, the legislative trend towards over-criminalization.  Ever-expanding statutes have made an increasingly broad range of conduct[7]—even petty or innocuous conduct,[8] or risky conduct we once dealt with civilly[9]—the subject of criminal law. 

¶6  Much of criminal procedure aims to restrict invasion of personal privacy and liberty by police, preventing them from stopping and searching us for no reason while walking or driving, and from searching our homes and personal effects without good cause.  By expanding substantive criminal law to ensure, in effect, that everyone occasionally violates a criminal law, legislatures allow police to avoid such procedural constraints.  Police can now pick and choose among us all as we commit petty violations in order to search, on a hunch, for evidence of major ones, as criminal conduct justifies invasions not allowed for civil violations.[10] 

¶7  Notice that the foregoing describes the problem of legislative over-criminalization, which many (including Stuntz) would like to see constrained by the courts.  Apprendi, in contrast, is about legislative under-criminalization.  By making the racial motive a “sentencing factor” rather than a “factual element” of the crime itself, the New Jersey legislature evaded two central requirements of constitutional criminal procedure: the burden of proof and jury adjudication of the elements of non-petty crimes.

¶8  How can both over- and under-criminalization be legislative strategies for avoiding procedural mandates?  The answer lies in the traditional division of criminal procedure rules into investigative processes and adjudicative processes.  Scholars like Stuntz, who focus mostly on investigative criminal procedure, naturally see over-criminalization as the legislative means to aid police in circumventing the set of procedural rules arising primarily from the Fourth and Fifth Amendments (due process, search and seizure).  In contrast, Apprendi falls on the adjudication side, seeking to bolster the rights provided by the Sixth Amendment (such as the right to a jury trial, and to fair notice of charges and sentence).  Under-criminalization is the legislature’s means to grant more power and discretion to prosecutors than would otherwise be allowed by the restrictive procedural rules.  This is why legislatures draft statutes like the one under which Apprendi was charged—facts of his conduct can be proven more easily to a  judge than a jury, due to the lower proof standard, and more cheaply and quickly.

¶9  The legislative strategy with regard to both types of criminal procedure is to trust law enforcement officials (both prosecutors and police officers) with a lot of discretion, while granting limited resources.  With leeway to search the populace at-large granted by broad over-criminalization, first police will follow hunches, sort through petty offenders, and (in theory) efficiently find the worst offenders we really want to prosecute.  (Of course, one obvious downside is they may target suspects on racial or other illicit grounds; another is the potentially high social cost of privacy invasion.)[11]  Once criminals are caught, legislatures have expedited the process of disposition by easing prosecutors’ hurdles to conviction and sentencing, trusting their discretion to employ tools like sentencing enhancements only when they are truly warranted (which, ideally, would coincide with those cases in which the jury would find those facts anyway). 

¶10  Bibas first argues that Apprendi, because of its weak and formalistic nature, can be easily circumvented by creative drafting by legislatures, returning us (effectively) to the pre-Apprendi regime of judicial finding on key facts that affect sentences.  On this point he has much company.  Apprendi’s current weaknesses make it an adjudication-side equivalent of recent doctrines that have largely failed to restrain legislatures from undermining investigation-side entitlements.  The Court’s most likely constitutional means to limit over-criminalization all have in fact had only minimal reach, and Bibas argues that Apprendi is similarly ineffective in limiting under-criminalization. 

¶11  Robinson v. California[12] once promised Eighth Amendment requirements for criminal responsibility, [13] but it has since been read narrowly merely as a ban on punishing the “status” of addiction.[14] Lambert v. California[15] suggested a constitutional requirement of mens rea that has since gone nowhere; even though the Court disfavors strict liability crimes by interpreting statutes to contain mens rea elements, no due process requirement has evolved from Lambert to seriously limit strict liability.[16]  Papachristou v. City of Jacksonville[17] provided a basis in vagueness doctrine for overturning the broadest of loitering and public order offenses, but legislators have easily avoided its restrictions by enacting an array of specific-but-broad offenses that still allow police to stop a large proportion of citizens in public.[18] 

¶12  To make such doctrines, as well as Apprendi, real limits on legislatures’ efforts to expand police investigative discretion, the Court would have to take a bold step into regulating substantive due process.  As King and Klein point out in another Apprendi article published this year,[19] any rule stronger than Apprendi designed to similarly protect adjudication rights (using a constitutional standard to dictate what statutory facts must count as crime elements) would have to be relatively invasive in its substantive restraint of legislatures.[20]  However, the Court is unlikely to make such an anti-democratic encroachment on legislative power in this context, even though such a move would be aimed at ameliorating a decidedly anti-democratic process defect (i.e., majoritarian disrespect for the entitlements of a disfavored minority--criminal defendants).[21]

¶13  Thus, Apprendi shows every sign of being an equally marginal limit on legislative under-criminalization, and only time will tell whether Apprendi does anything more than Papachristou or Lambert to restrain legislative strategies to expand the discretionary power of law enforcement officials.  But judging from the developments in the investigative side of criminal procedure, prospects for preserving the efficacy of adjudicative rights without substantive limits on legislative power are not promising.  The same political dynamics that allow for an ever-broadening scope of criminal statutes,[22] thereby increasing police power, will likely drive legislatures to attempt to evade Apprendi’s prosecutorial restrictions; legislatures will continue to find ways to “un-criminalize” the determination of key facts, in a continued effort to make them functionally civil issues for judges to decide.

¶14  Bibas, however, is more worried about enhanced prosecutorial power after Apprendi, assuming first that it does restrict (even if only marginally) legislatures’ ability to grant tools for use in discretionary strategies.  Counterintuitively, he sees prosecutors gaining from Apprendi’s nominal constraints on legislative drafting freedom.  His analysis fits within an important modern trend in the scholarship of criminal law (and other areas) to look at how real-world dynamics are likely to shape the practical effect of formal rules, and to assess whether rules in practice accomplish what courts intend them to do.[23], [24]  Bibas argues that, because most defendants plead guilty, Apprendi’s grant of a jury determination of facts formerly used as sentencing factors is worse than meaningless.  Perversely, he argues, prosecutors will in fact use the rule to deprive defendants of the sentencing hearings that ordinarily follow guilty pleas, and will coerce defendants into waiving challenges to sentences; post-Apprendi prosecutors will be able drive harder bargains and even reduce judicial supervision of prosecutorial sentencing decisions.[25]

¶15  Like all forecasts of future behavior, Bibas’s thesis rests on several assumptions.  He assumes that judicial findings of sentencing factors (such as drug quantity or racial motive) were once a substantial check on prosecutorial power and thus a meaningful right for defendants to begin with.[26]  He also assumes that prosecutors who formerly declined to seek sentence enhancements based on recidivism for defendants who pled guilty will now use that power to coax pleas to sentencing factors as well as base elements.[27]  If these assumptions are refuted, Apprendi is likely to work, in fact, much as the Court intends.[28]  To be sure, every prosecutorial strategy that concerns Bibas existed well before Apprendi--prosecutors have long been coercing pleas to aggravating facts with recidivism enhancements and, despite Bibas’s post-Apprendi concern about arbitrarily different treatment of similar offenders, prosecutors have been distinguishing between offenders by charging different grades of offenses for just as long.[29]

¶16  Unless Bibas’s views of legislative motives are as counterintuitive as his analysis of post-Apprendi litigation practice, he presumably does not believe that legislatures will take any action in response to a decision that he believes to be beneficial to prosecutors.  In his view, Apprendi makes it significantly easier and cheaper for prosecutors to obtain plea bargains with harsher sentences; thus, redrafting statutes to circumvent Apprendi’s nominal legislative restrictions could only benefit criminal defendants in strengthening their procedural entitlements, bargaining leeway, and the judicial supervision of prosecutors’ sentencing preferences.  Since his take on Apprendi leads to increased prosecutorial leverage, Bibas must believe that legislatures are unlikely to worry about any marginal, theoretical legislative implications.

¶17  What this means is that we can judge whether Bibas is right by simply watching legislative reaction to Apprendi.  If Bibas is right, law enforcement has already gained the upper hand with Apprendi, and legislatures will see little incentive to decrease prosecutorial discretion by returning to the pre-Apprendi state of affairs, or otherwise risking a statutory response to Apprendi.  But if Apprendi works as the Court and defendants hope—if real-world dynamics do not turn the expansion of a right into a practical disadvantage—we should gradually see legislatures marginalizing Apprendi bymoving toward higher statutory maxima, with judicial findings of mitigating factors to lower sentences rather than aggravating factors to raise them.

* Associate Professor and Alumni Faculty Fellow, Washington and Lee University School of Law.

[1] 530 U.S. 466 (2000).

[2] Stephanos Bibas, Judicial Fact-Finding and Sentence Enhancements in a World of Guilty Pleas, 110 Yale L.J. 1097 (2001).

[3] Id.

[4] Apprendi, 530 U.S. at 470.

[5] Id.

[6] See William J. Stuntz, O.J. Simpson, Bill Clinton, and the Transsubstantive Fourth Amendment, 114 Harv. L. Rev. 842 (2001) (arguing that excesses in search rules arise from the failure to calibrate them with substantive crime seriousness); Stuntz, The Uneasy Relationship Between Criminal Procedure and Criminal Justice, 107 Yale L.J. 1 (1997) (discussing the relationship of procedural rules to substantive criminal law and funding); Stuntz, Substance, Process, and the Civil-Criminal Line, 7 J. Contemp. Legal Issues 1 (1996).

[7] An estimated 300,000 federal statutes and regulations are enforceable criminally. See Susan L. Pilcher, Ignorance, Discretion and the Fairness of Notice: Confronting “Apparent Innocence” in the Criminal Law, 33 Am. Crim. L. Rev. 1, 32 (1995).

[8] Municipal ordinances are great sources for criminalization of innocuous conduct. See, e.g., People v. Kail, 501 N.E.2d 979 (Ill.App.Ct. 1986) (crime of riding a bicycle without a bell).  My current favorite among federal crimes is unauthorized use of “Smokey the Bear” imagery.  18 U.S.C. § 711 (1994).

[9] Driving under the influence of alcohol was once widely sanctioned as a civil offense. See, e.g., Welsh v. Wisconsin, 466 U.S. 740 (1984).  DUIs probably have been criminalized in reaction to social changes that make us view such conduct more seriously, but Welsh gave states another incentive to criminalize it: police in “hot pursuit” can follow offenders into their homes for criminal offenses but not civil ones.  Id. at 41.

[10] See Id.

[11] See Tracey Maclin, Race and the Fourth Amendment, 51 Vand. L. Rev. 333 (1998) and David Harris, Driving While Black and All Other Traffic Offenses: The Supreme Court and Pretextual Traffic Stops, 87 J. Crim. L. & Criminology 544 (1997) (seemingly biased use of investigative leeway arising from criminalization of petty traffic offenses gives rise to complaints of “driving while black”).  See also Whren v. United States, 517 U.S. 805 (1996) (foreclosing Fourth Amendment inquiry into the subjective motivation of officers making traffic stops).  For further analysis of this topic, see Pamela S. Karlan, Race, Rights and Remedies in Criminal Adjudication, 96 Mich. L. Rev. 2001 (1998).

[12] 370 U.S. 660 (1962).

[13] Id. at 667.

[14] See Powell v. Texas, 392 U.S. 514, 535 (1968) (justifying a restrictive reading of Robinson that does not cover the crime of being intoxicated in a public place because, on a broader reading, “it is difficult to see any limiting principle that would prevent this Court from becoming … the ultimate arbiter of the standards of criminal responsibility, in diverse areas of criminal law”).

[15] 355 U.S. 225 (1957).

[16] See Richard G. Singer, The Proposed Duty to Inquire as Affected by Recent Supreme Criminal Law Decisions in the United States Supreme Court, 3 Buff. Crim. L. Rev. 701 (2000) (discussing four cases that suggest that the Court strives to read mens rea requirements into crimes whenever possible: Ratzlaf v. United States, 510 U.S. 135 (1994); United States v. X-Citement Video, 513 U.S. 64 (1994); United States v. Staples, 511 U.S. 600 (1994); and Cheek v. United States, 498 U.S. 192 (1991)).  For an insightful analysis of when the Court does allow strict liability crimes, see Alan Michaels, Constitutional Innocence,112 Harv. L. Rev. 828 (1999).

[17] 405 US. 156 (1972). 

[18]The Court recently reaffirmed  Papachristou’s  vagueness doctrine when it  found a gang-loitering ordinance unconstitutional. City of Chicago v. Morales,527 U.S. 41 (1999). However, neither decision does much to limit legislative crime-definition power or police investigative discretion.  See Debra Livingston, Gang Loitering, The Court, and Some Realism About Police Patrol, 1999 S. Ct. Rev. 141.

[19] Nancy J. King & Susan R. Klein, Essential Elements, 54 Vand. L. Rev. 1467 (2001).

[20]  Id.

[21] Id.  Klein and King are, in my opinion, rather sanguine about Apprendi’s effect on legislatures, arguing for “[l]egislative freedom to accommodate the criminal law to the times … except in the most extreme circumstances,” and against “[t]oo much regulation of the evolution of criminal law by courts”).  Id. at 1542.  They propose a more moderate re-development of Apprendi, in the form of a multi-factor test that would prohibit only “the worst legislative excesses” and otherwise protect “the supremacy of the legislature in defining substantive criminal law.”  Id. at 1536-41.

[22] For more on these political forces, see Stuntz, Substance, Process, and the Civil-Criminal Line, supra note 6.

[23] Stuntz’s analyses are prominent in this respect. See supra note 6.  See also Daniel C. Richman, Old Chief v. United States: Stipulating Away Prosecutorial Accountability?, 83 Va. L. Rev. 939 (1997) (discussing federal prosecutors’ response to the Old Chief rule, which gives defendants a right to stipulate to prior convictions).

[24] See Michael Klarman, Brown, Racial Change, and the Civil Rights Movement, 80 Va. L. Rev. 7 (1994) (an example in constitutional law); Russell Korobkin, The Status Quo Bias and Contract Default Rules, 83 Cornell L. Rev. 608 (1998) (an example in private law, drawing on behavioral economics).

[25] See Bibas, supra note 2, at 1152-59.

[26] Id. at 1155-56.

[27] Id. at 1161-64.

[28] This is, in fact, the thrust of Nancy King and Susan Klein’s recent rebuttal to Bibas. Nancy J. King & Susan R. Klein, Apprendi and Plea Bargaining,Stan. L. Rev. (forthcoming 2001).

[29]Id. at notes 14-15 & 45-51.





© 2005 Boalt Journal of Criminal Law, All Rights Reserved

Current Issue Archive  Submissions Subscribe About BJCL Contact Us Vol 5 Index