If the rule of law were brought to bear on sentencing decisions, it would not just permit, but would rather require, different penal sanctions to be imposed when offenders are found to differ in characteristics and circumstances that are relevant to the choice of penal purpose in the case, or to the plausibility of a penal strategy by which the sentence is expected to advance its purpose.
The decisional rules embedded in presumptive guideline grids have achieved some of their drafters' objectives, but fall short of a "rule of law" for sentencing, a paramount goal of the sentencing reforms Marvin Frankel set in motion 25 years ago. The rule of law requires a reasoning process that moves from penal purpose to penal measure by fair inference from relevant facts. Two-dimensional grid guideline schemes gain simplicity by reducing relevant facts to two: current offense and prior record. Even when judges are permitted to consider other facts, the presumptive guideline methodology obscures differences in the penal purposes sentencers ought to have in mind for cases that fall into the same cell of a grid. In contrast, the rule of law requires fact-finding in the individual case to reveal such differences and to test the plausibility of available penal measures. This elevates the relevance of offender characteristics and circumstances, including some that are held ordinarily irrelevant in the Federal scheme and are of uncertain relevance in many State schemes. Although this is most obvious when use of noncustodial penal measures is subjected to the rule of law, it is no less important when imprisonment is used. Disparate impacts should be avoided by more sophisticated deployment of correctional authority and resources, not by constriction of the rule of law that governs the sentencing decision itself. 31 notes and 53 references