1. “Determining the Retroactive Reach of Decriminalization and Diminished Punishment” by Harold J. Krent

View PDF

May 22nd, 2013


Harold J. Krent*Φ

Legislative decriminalization often leaves those previously convicted of the underlying offense in jail.  Similarly, when legislatures reduce the punishment for a particular offense, those convicted in the past do not generally benefit if their conviction has become final.

Focusing on a tragic incident in which a seventeen year-old in Georgia was sentenced to a ten-year term of imprisonment for consensual sex with a fifteen-year-old,[1] Professor S. David Mitchell reexamines whether those convicted should benefit from legislative decriminalization or decisions to lessen punishment.[2]  In the Georgia case, the legislature reacted to the national furor over the sentencing by eliminating consensual sex among minors from the definition of aggravated child molestation, but it did not make the change retroactive.  Accordingly, Professor Mitchell urges recognition of a rule of complete retroactive amelioration, which would prompt release of all those previously incarcerated after decriminalization of an offense.  Similarly, Professor Mitchell recommends lessening punishment for those previously convicted when the legislature reduces a criminal penalty. Professor Mitchell argues that neither consequentialist nor retributivist theories of punishment support continuing punishment for those, like the seventeen-year-old, whose actions subsequently are decriminalized or punished less severely. [3]

From a consequentialist perspective, Professor Mitchell is on solid ground.  Little can be gained from a deterrence perspective by maintaining the prior punishment scheme after the legislature decriminalizes an offense or changes a legislative classification.  Deterrence will not disappear if offenders know that they will benefit should the legislature later decriminalize the offensive conduct or lessen the punishment.  The prospect is way too remote.  From the perspective of specific deterrence, there is no need to continue punishment if the offense is decriminalized or the punishment exceeds the cap now placed by the legislature.  Finally, there is little need for extended rehabilitation when the legislature has stated that, henceforth, the underlying conduct should not be punished as severely or at all.[4]

From a retributivist perspective,[5] however, Professor Mitchell’s argument falters in at least two contexts.  First, consider what some have termed “legalistic” retribution – retribution arising because an offender knowingly transgresses a rule of the community.[6]  The fact that norms later change in no way undermines the conclusion that the individual knowingly (depending on the mens rea required) violated a rule of the community.

Take a simple example.  Suppose that I have been convicted of reckless driving after being clocked at eighty miles per hour in a fifty-five mile per hour zone on a highway.  (Luckily, that has not happened to me in a long while!)  The jurisdiction later changes the law to permit eighty miles per hour on that stretch of highway.  Should my punishment be lessened?  To be sure, the community no longer views my prior speed as inherently wrongful, but I chose to travel that speed knowing that the community at that time regarded fifty-five miles per hour as the appropriate speed.  Indeed, my behavior may have been designed to gain some kind of advantage over my fellow citizens, such as by arriving at a destination prior to others.[7]  At times, laws prevent one individual from gaining an unfair advantage over the other.[8]  Irrespective of any improper advantage, why shouldn’t I suffer just desserts for willfully or recklessly violating the social order?

To be sure, it would be difficult to gauge how much punishment is appropriate, but surely some punishment is warranted for knowingly violating a social command.  In essence, part of each punishment is attributable to the blameworthiness of the underlying act and part due to the willingness to violate a social command.  Accordingly, despite the empathetic situation in Georgia that presumably triggered Professor Mitchell’s thoughts on retroactive amelioration, the criminal defendant there as well knowingly violated what society deemed wrongful, despite its controversial underpinnings.  Such offenders arguably should be held responsible even when society later clarifies those very commands.

This is not to suggest that the legalistic retribution perspective should carry the day, but it does counsel that the case for complete ameliorative retroactivity is not airtight.  When a crime is invalidated as unconstitutional, for instance, we long have been comfortable in unlocking the jailhouse doors, as Professor Mitchell relates.[9]  Perhaps, the willfulness in violating law can also be counteracted by the legislature’s changing views of the wrongfulness of the underlying conduct, particularly because the proportionality of the punishment is so hard to gauge.  The legislature’s altered view could in essence excuse the intentional violation of the social order – the (legalistic) retributive case for punishment might be trumped, but not eliminated.

Even then, Professor Mitchell’s dismissal of the retribution argument runs into further trouble.  He views retribution from a static perspective: any contemporary decision to reduce punishment is tantamount to a determination that the original punishment was mistaken.  But, the same antisocial conduct may “merit” different punishment – whether more[10] or less—in the legislature’s eyes because of changed factual circumstances as opposed to morality.  The appropriateness of particular punishment or, indeed, whether to punish at all, ebbs and flows with greater societal changes that may not be directly linked to the underlying conduct at stake.[11]

For instance, it may be one thing to conspire to ship unlicensed liquor over state lines now, and quite another thing to have committed the same conduct during Prohibition.  Congress may have lowered the punishment for interstate shipment of unlicensed liquor after Prohibition, but that does not suggest that its predecessor somehow erred in ascribing a more serious penalty during Prohibition.  And, from a retribution perspective, there is no persuasive reason to excuse interstate shipment of liquor during Prohibition.

Similarly, consider someone convicted of importation of marijuana into the United States before the legalization of medicinal marijuana in California, Colorado and Washington..  As in the Prohibition context, the social opprobrium we place on such conduct may be less now than before, but that does not mean we should ease the punishment for conduct when there was no legal use for marijuana at the time of the offense.

Or, an individual convicted of trafficking in bald eagles (if that were indeed a crime) should not be subject to lesser penalties given that bald eagles later are removed from the endangered species list.[12]  In the speeding example itself, the legislature likely raised the speed limit because of safer roads and cars, which would not excuse speeding in an era in which such speeds were riskier.  Blanket retroactive amelioration ignores that diminished legislative punishment or decriminalization may take into account circumstances that were not in effect at the time the crime was committed.[13]

Similarly, legislatures may reduce the penalties for particular crimes not because of changed norms or views of the wrongfulness of the underlying conduct, but rather for instrumental reasons of the rising cost of incarceration or the social costs of incarcerating too many young men.  Professor Mitchell states as much.[14]  Such decisions to ameliorate punishment do not necessarily lead to the conclusion that those previously convicted also should have their punishments reduced.  A rational legislature could conclude that the social or other benefits of the lightened punishment are more important with respect to those to be charged in the future than those charged in the past.  For instance, if the goal is minimizing incarceration of young offenders, then there is more reason to limit future incarceration than to open the jailhouse doors for those sentenced under the prior regime.

In short, Professor Mitchell’s analysis powerfully points out the injustice in many cases when a legislative diminution in punishment or decriminalization is not applied retroactively.  But, not in all.  Legislatures should still be able to determine, on a case by case basis, whether prior offenders warrant more severe punishment than current offenders due to the changed circumstances or changed perceptions of the social benefits of incarceration.  It is not inconsistent for a legislature to benefit future as opposed to past offenders in light of changed circumstances.  Thus, although I agree that presumptive retroactivity is appropriate, the allure of blanket retroactive amelioration founders on retributivist principles that many hold dear.


* Dean and Professor, IIT Chicago-Kent College of Law.  A version of this essay appears in a larger work: Retroactivity and Crack Sentencing Reform, 47 U. Mich. J.L. Reform (2013).

Φ Suggested citation: Harold J. Krent, Determining the Retroactive Reach of Decriminalization and Diminished Punishment, 40 Fordham Urb. L.J. City Square 7 (2013), http://urbanlawjournal.com/?p=1207.

        [1].  See Wilson v. State, 652 S.E. 2d 501 (Ga. 2007).  See also Chandra R. Thomas, Why is Genarlow Wilson in prison?, Atlanta Mag., Jan. 2006, at 64-69.

        [2].  S. David Mitchell, In with the New, Out with the Old:  Expanding the Scope of Retroactive Amelioration, 37 Am. J. Crim. L. 1 (2009).

        [3].  Some have argued that retroactive application of decriminalization to offenders whose offenses are final would violate the executive’s pardon power.   I am inclined to disagree but focus in this response only on the retributivist case for retroactive amelioration.

        [4].  Mitchell, supra note 2, at 14-16.

        [5].  Prof. Mitchell argues that “an ameliorative change represents a legislative acknowledgement, and by proxy a societal acknowledgement as well, that the prior penalty was disproportionate to the conduct.” Id. at 16.

        [6].  Russell L. Christopher, Deterring Retributivism: The Injustice of “Just” Punishment, 96 Nw. U. L. Rev. 843, 885 (2002).

        [7].  Alternatively, my violation of the law might have prompted others to reconsider their own law- abiding ways, even if I did not cause danger on the highways.

        [8].  Consider the repeal of an anti-gouging statute.  Even if a legislature decides to decriminalize raising prices of commodities during an emergency, those who previously were convicted of the offense attempted, by raising prices, to benefit at the expense of competitors as well as consumers.  A similar dynamic would arise from decriminalization of particular antitrust laws.  See also Herbert Morris, On Guilt and Innocence: Essays in Legal Philosophy and Moral Psychology 33-34 (1976).

        [9].  Id. at 33.

      [10].  Of course, legislative decisions to impose greater punishment are barred retroactive effect due to the Ex Post Facto Clauses.

      [11].  I am not arguing that we should block retroactive amelioration to defer to the morality of prior legislatures but rather that some crimes are now considered socially less blameworthy because of changed factual circumstances and not merely because of changed morality per se.

      [12].  Peter Slevin, Bald Eagle to be Taken Off Engandered List, Wash. Post
(Dec. 25, 2006), http://www.washingtonpost.com/wpdyn/content/article/2006/12/24/‌AR2006122400666.html.

      [13].  The dynamic can be reversed – changes in larger society often make conduct worse in retrospect than at the time the conduct took place.  Theft of large amounts of fertilizer became more threatening only when the potential for its use in homemade explosives was more understood, and the same is true for the unlicensed importation of pseudoephedrine (i.e. Sudafed), which over time has become a critical ingredient of crystal meth.

      [14].  Mitchell, supra note 2, at 41.