Antonin Scalia famously observed in his dissent in United States v. Booker that an advisory sentencing guidelines regime would result in a “discordant symphony” where similarly situated offenders would receive ad hoc sentences. As this article demonstrates through a statistical survey, he was right. Federal sentencing practice is in chaos. The fundamental goals of the guidelines—uniformity, proportionality, and certainty—have been undermined. Nonetheless, this does not mean the guidelines should be abandoned or a wholesale redrafting is required. As it turns out, the guidelines continue to be a useful tool, not for determining the ultimate sentence per se, but for identifying similarly situated offenders. Using the total offense level, criminal history category, and other relevant factors discussed in the article, a sentencing judge can identify those individuals within the U.S. Sentencing Commission’s datafiles matching those criteria. From there, a judge can determine various relevant statistics to consider when imposing a sentence: most importantly, the Interquartile Range (IQR). The IQR defines the central range of sentences imposed on similarly situated offenders. Sentencing within the IQR will necessarily promote the guidelines fundamental goals and thus harmonize federal sentencing. But the upshot of this approach also demonstrates that it is the sentencing table, and not the guidelines overall, that are in need of fundamental revision. This article concludes that the Commission should recalibrate the sentencing table downward to match current sentencing practice.
Research Article| February 01 2020
Sentencing in Chaos: How Statistics Can Harmonize the “Discordant Symphony”
Federal Sentencing Reporter (2020) 32 (3): 128–137.
- Views Icon Views
- PDF LinkPDF
- Share Icon Share
- Search Site
Mark H. Allenbaugh; Sentencing in Chaos: How Statistics Can Harmonize the “Discordant Symphony”. Federal Sentencing Reporter 1 February 2020; 32 (3): 128–137. doi: https://doi.org/10.1525/fsr.2020.32.3.128
Download citation file: