Sentencing as a Human Process

Rhys Hester
Rhys Hester

Over the decades, sentencing guidelines have been both celebrated and castigated. Those who praise them point out that guidelines have been successful: they’ve reduced racial disparities, increased uniformity, and assisted some jurisdictions in effectively managing prison populations.1 Criticisms include that guidelines—or at least some guidelines (they’re not all created equal)—are overly mechanical and an unfortunate departure from treating sentencing as a human process. Some examples of the overly mechanistic characteristics relate to the way structured sentencing treats a person’s criminal history.  Guidelines frequently count prior offenses regardless of how long ago they occurred, they often include misdemeanors and juvenile offenses, and they count these prior events according to inflexible rules. What’s more, they only take account of bad things in a person’s past; they don’t give credit for efforts at rehabilitation, prior success towards desistance from crime, or other positive social strides. As Professor Michael Tonry has put it, two dimensional grids have “impoverished sentencing,” in part because the “two-axis grid reifies thinking about punishment into a calculus that takes account only of criminality.”2 As Tonry notes, this doesn’t mean grids should be abandoned, “but only that their drawbacks and dangers should be recognized.”3 After all, indeterminacy—the leading alternative to guidelines—was once described as “terrifying and intolerable for a society that professes devotion to the rule of law.”4 To mix metaphors, are we forced to choose between an overly rigid guidelines machine and lawless Wild West of indeterminacy? Maybe there are ways to soften The Machine without devolving into Madness, to make sentencing more of a human process, while still maintaining consistency and fairness. 

Criminal history scores were not always quite so narrow and mechanical. In the earliest efforts, now long forgotten, guidelines architects did include scoring mechanisms that would allow for the deduction, not just the accumulation, of criminal history points. Some of the earliest sentencing guidelines projects of the 1970s were an offshoot of the development of parole guidelines and included pilot sites in Essex County, New Jersey and Maricopa County, Arizona.5 Early models developed for Essex County provided for the deduction of a point when the offender was employed or in school, and some of the preliminary models from Maricopa County called for a one-point reduction (not just a 0) from the offender score when the individual had no prior adult convictions.6

Utah recently made some adjustments to its sentencing guidelines to provide for a similar mechanism that actually allows individuals to earn a one-point deduction in their criminal history score for successfully completing a probationary sentence.7 This new rule went into effect in 2015 as part of Utah’s Justice Reinvestment initiative. 

My recent research on criminal history enhancements in a state that never adopted sentencing guidelines supports such policies that would give credit towards positive past progress.8 I interviewed trial judges in a non-guidelines state who sometimes considered a person’s criminal history as highly relevant, but not always. The judges didn’t treat prior record in a mechanical way, as operationalized in most guidelines schemes. They were instead most concerned with a steady and recent pattern of serious offending, and an inability to successfully complete a probationary term without revoking. Except for the most serious offenses, the judges generally felt that first- and even second-time offenders deserved a second (or third) chance via probation rather than prison. But eventually, the accumulation of a criminal record would indicate that other efforts had failed, and some prison time was appropriate. The judges did not necessarily find old offenses or offenses committed while a juvenile relevant, and they appeared interested in what inferences they could make about an individual based on the age of priors, the time in between offenses, and prior successes with probation. Thus, a person could effectively earn back a clean slate if their prior offenses were not recent, or if they’d shown themselves amenable to probation.

For instance, several judges explained that while having recent priors or failure on probation might signal that the individual is not amenable to probation at the current time, they also viewed a prior successful completion of probation as a positive sign. One judge explained:

It’s not just the number of checks against them. It goes more deeply. Were they on probation? Did they live probation? How long has it been? Have you gone 10 years without getting into trouble?

The concept of “living probation” or “living one,” refers to an individual’s ability to successfully complete a probationary sentence without committing violations, and this was a recurrent theme of importance for many of the judges in the interviews. As another judge said,

I always want to know what it [the prior offense] was and the disposition—did they get an active sentence or probation, which would tell me if they could live one.

And yet another:

If they’ve been on probation in the past I’m always interested in how they did. If they flunked and got revoked it’s unlikely I’ll try again. But if they were on their third time on probation and did OK I might give it again. Same with early release from prison.

These perspectives from non-guidelines judges are consonant with the new Utah Guidelines policy of allowing a deduction for prior success on probation. The interview findings also underscore Professor Tonry’s criticism, quoted above, that guidelines impoverish the sentencing process by only taking account of criminality. If one buys into the guidelines-are-overly-mechanical critique, one avenue for reform could be to follow Utah’s recent example of reimagining how criminal history scores could become two-way streets, mirroring the more holistic approach of indeterminacy, though still through guidelines and promulgated rules to maintain consistency. For example, in addition to giving credit for prior probationary success, guidelines could allow for a one-point deduction from the score when an individual is able to demonstrate to the court that he satisfied court-ordered restitution for a prior property offense, or successfully completed substance abuse treatment following a prior drug conviction. And if most reasonable judges engaging in sentencing as a human process would disregard a 20-year-old conviction or one committed by a juvenile, guidelines themselves could also easily incorporate this intuitive limitation in the guidelines scoring rules.9 Establishing rules that reward progress or disregard facts that are no longer relevant would not only create incentives for positive social behavior, but would also go along way to enriching structured sentencing as a holistic process while maintaining a guidelines emphasis on fairness and consistency in sentencing.

  • 1See, e.g., Richard Frase, Just Sentencing: Principles and Procedures for a Workable System (2013); Michael Tonry, Sentencing Fragments: Penal Reform in America, 1975-2025 (2016).
  • 2Tonry, Sentencing Matters (p.20).
  • 3Id.
  • 4Marvin Frankel, Criminal Sentences: Law Without Order (1973: 5).
  • 5See Joseph C. Calpin, Jack M. Kress, and Arthur M. Gelman, Sentencing Guidelines: Structuring Judicial Discretion, Volume II., Analytic Basis for the Formulation of Sentencing Policy, National Institute of Justice (1982).
  • 6Id. at pp.92, 98.
  • 7See Utah Sentencing Commission, 2016 Adult Sentencing & Release Guidelines at 24. Available at
  • 8Rhys Hester, Punishing for the Past (Sometimes). Working Paper, The Robina Institute of Criminal Law & Criminal Justice, University of Minnesota Law School (2016) (draft available upon request).
  • 9These rules exist and are in fact used by some guidelines jurisdictions, but many others continue to count juveniles, misdemeanors, and felonies that were committed long ago. See generally, Richard Frase, Julian Roberts, Rhys Hester, and Kelly Mitchell, Criminal History Enhancements Sourcebook (2015), available at