« Idaho judge includes celibacy for teen sex offender on intensive probation | Main | Florida legislature finally moving toward really fixing its capital procedures after Hurst »

February 6, 2017

"A Theory of Differential Punishment"

The title of this post is the title of this notable new paper authored by John Boeglin and Zachary Shapiro now available via SSRN. Here is the abstract:

A puzzle pervades the criminal law: Why is it that two offenders who behave identically are sentenced differently when one of them, due to circumstances beyond her control, causes a harmful result? Through first proposing a novel deconstruction of this question by separating theories of punishment into two broad categories (namely, offender-facing and victim-facing justifications for punishment), the Article demonstrates that results-based “differential punishment” in the criminal law can only be justified, if at all, by victim-facing theories.

The Article then makes its central claim: while victim-facing theories may be capable of justifying results-based punishment in respect to many types of offenses, there are three distinct classes of offenses for which everyone should agree that differential punishment is unjustified.  We conclude by showing how applying our framework would reduce the unnecessary incarceration of a significant class of criminal offenders, without sacrificing any legitimate goals of the criminal justice system.

February 6, 2017 at 07:08 PM | Permalink


I believe that the best way to conceptualize this problem is to note that the word "luck" is in the dictionary. Some people get lucky and either not get caught or get caught and a jury lets them off. Other people get unlucky and they pay for a crime that they were not responsible for.

There is no need to create justifications for luck. It over-intellectualizes the issue.

Posted by: Daniel | Feb 6, 2017 7:18:49 PM

Would the victim facing punishment theory end mala prohibita crime, regulatory crime, and other risk raining crime not resulting in any physical or economic harm to a victim? I drive at 100 mph on New York's Fifth Avenue at 2 AM, no one is harmed. No punishment?

Those types of crimes are not mentioned in the article, that I could find.

Posted by: David Behar | Feb 7, 2017 7:54:30 AM

Hi David,

Jack Boeglin here, co-author of the article. Your comment raises an interesting question; We give our response on p. 56-57 of the article, where we write:

"Finally, a special case is presented by certain regulatory offenses, such as traffic violations, for which there is both no “object” of the offense—it is not “done to” anyone—and no derivative offense criminalizing the underlying behavior, but for which the underlying behavior poses sufficient risks that it should be criminally prohibited. For this narrow set of offenses, the failure to independently punish the behavior underlying them is undoubtedly a product of the reality that detecting such undesirable behavior in the absence of the statutory harm occurring would be impractical. For instance, it might be functionally impossible to know if someone was driving recklessly with regard to breaking the speed limit if they were not, in fact, speeding.

As a matter of pure theory, this is not a reason to engage in differential punishment—that is, in the (possibly rare) cases where we could be certain that an offender had engaged in the underlying behavior (e.g., recklessness with regard to the speed one is driving) without causing the statutory harm, we would have sufficient offender-facing justifications for punishing him equally as an offender who did bring about the statutory harm. In an ideal world, therefore, such crimes would be redefined solely in terms of the undesirable behavior they seek to prohibit, without regard to whether a statutory harm occurs.

Until such derivative offenses are created, however, it may be necessary in these limited circumstances to punish criminals on the basis of whether the statutory harm occurs, even when no victim-facing justifications for punishment are available. In light of the obvious offender-facing justifications that apply to these offenses, it would simply be unacceptable to allow them to go unpunished entirely."

Hopefully that is responsive to your concern,

Posted by: Jack Boeglin | Feb 7, 2017 10:48:38 AM

The article, appreciate the follow-up, notes the law provides various legitimate and illegitimate punishments where luck is a factor. For instance, if you shoot a gun in a crowded area, you might be lucky and not hit anyone. Or, the person hit might not die. etc. There is a reason, so argues the article, to still punish the person unlucky more than the one who is lucky. OTOH, there are various cases where there isn't a reason to benefit luck.

The first comment says "luck" and than cites a specific sort of luck. There are various types of luck and they probably should be handled differently.

Posted by: Joe | Feb 7, 2017 11:54:11 AM

Jack. Thank you. I just missed that passage toward the end of the article. The solution in your proposal would likely be technological. Radar tickets everyone speeding, in a mandatory way.

In your article you share one goal of mandatory sentencing guidelines, more equality of sentencing for equal culpability, with discounts for other values pre-sentencing conduct. One may see a robot on the bench applying legislatively enacted algorithms. I will admit that I have never once gotten the correct answer in exercises of the arithmetic of sentencing guidelines. I am curious if experts such as Prof. Berman ever have, come up with the same sentence as a judge obeying sentencing guidelines.

In an Israeli study, posted in this blog, defendants received lighter sentencing for matched crimes after the judges had lunch. Hypoglycemia would not be problem in the mood of a robot.

Posted by: David Behar | Feb 7, 2017 2:12:43 PM

Post a comment

In the body of your email, please indicate if you are a professor, student, prosecutor, defense attorney, etc. so I can gain a sense of who is reading my blog. Thank you, DAB