The Supreme Court has stepped directly into one of the most important questions in federal sentencing today: what, exactly, counts as an “extraordinary and compelling” reason to reduce a lawfully imposed sentence under 18 U.S.C. § 3582(c)(1)(A)(i).
Two sets of cases now before the Court, Fernandez v. United States and the consolidated cases Rutherford v. United States / Carter v. United States, will shape how far district judges may go when considering compassionate-release motions. Fernandez asks whether alleged trial or sentencing error can be treated as an extraordinary and compelling reason. Rutherford asks whether nonretroactive changes in sentencing law, specifically the First Step Act's fix to stacked § 924(c) penalties, can justify compassionate release for people already serving decades-long sentences.
At Federal Research Group, we are following these cases closely because they will determine how much room remains for judges to correct unusually harsh or potentially flawed sentences after the usual appeals and habeas routes have run their course.
I. How We Got Here: A Short History of Compassionate Release
To understand why Rutherford and Fernandez matter so much, it helps to step back and look at how § 3582(c)(1)(A) evolved. When Congress passed the Sentencing Reform Act in 1984, it created the modern compassionate-release statute. But for decades, only the Bureau of Prisons (BOP) could file a motion asking a court to reduce a prisoner's sentence on compassionate grounds. Prisoners themselves had no direct path to court. It was a narrow, rarely used safety valve.
The Sentencing Commission eventually filled in some details by adopting Guideline § 1B1.13, which listed examples of qualifying reasons, terminal illness, serious medical conditions, advanced age with deteriorating health, and some family circumstances, plus a catch-all for “other reasons” identified by BOP. For many years, that policy and BOP's gatekeeping role kept the remedy limited to a small number of seriously ill or elderly prisoners.
During the same period, something very different was happening in federal sentencing more broadly. The Guidelines became increasingly complex and, to many critics, increasingly severe. Until 2005, courts were required to follow them. That changed when the Supreme Court decided United States v. Booker, making the Guidelines advisory rather than mandatory. Once that happened, district judges began exercising more discretion, and the Sentencing Commission adopted retroactive amendments (especially in drug cases) that allowed many people to receive sentence reductions. Presidential clemency also expanded. Still, compassionate release remained relatively marginal. Everything changed in 2018, when the First Step Act amended § 3582(c)(1)(A). Congress finally allowed prisoners, after first making a request to BOP, to file their own compassionate-release motions in court if BOP did not act. This simple but profound shift opened the door to thousands of prisoner-initiated motions.
Then came 2020 and COVID-19. The pandemic drove a surge in compassionate-release filings and, at least for a period, a greater willingness among judges to find extraordinary and compelling reasons, especially for medically vulnerable individuals in high-risk facilities. According to Sentencing Commission data, from late 2019 through mid-2025, prisoners filed over 36,000 compassionate-release motions, and courts granted roughly one out of every six. Compassionate release has moved from a rarely invoked provision to a major mechanism for sentence reduction in the federal system.
With that growth, however, came deeper disputes:
How broad is a judge's discretion to decide what counts as “extraordinary and compelling”?
Can the statute be used to address legal errors or nonretroactive changes in law that create glaring disparities?
How much weight should be given to the Sentencing Commission's policy statements, including its 2023 amendments that expressly allow consideration of certain nonretroactive changes in law?
Fernandez and Rutherford now bring those questions to the Supreme Court.
II. Fernandez v. United States: Can Alleged Errors Support Compassionate Release?
Joe Fernandez is serving a life sentence for his role in a murder-for-hire scheme tied to a large cocaine debt. At trial, a cooperating cousin testified that Fernandez was one of the shooters. The jury convicted him on all counts, and the court imposed a mandatory life sentence. His post-trial motions, direct appeal, and two rounds of habeas litigation all failed to disturb that judgment.
Years later, Fernandez returned to court under § 3582(c)(1)(A). This time, he did not argue a narrow medical or family hardship. Instead, he pointed to:
Serious doubts about the reliability of the key government witness
Comparisons between his life sentence and the substantially lower sentences imposed on codefendants who pleaded guilty.
The district court expressed “disquiet” about aspects of the evidence and about the sentencing disparity, and, relying on the compassionate-release statute, reduced Fernandez's sentence to time served. The Second Circuit reversed, holding that alleged trial or sentencing errors and generalized dissatisfaction with the evidence are not “extraordinary and compelling” reasons within the meaning of § 3582(c)(1)(A). In the court's view, such challenges belong in direct appeal or habeas, not in a compassionate-release motion.
At the Supreme Court, Fernandez argues that:
Congress deliberately gave district courts broad discretion to recognize exceptional situations that justify a lower sentence.
The statute itself imposes only one express limit: rehabilitation alone is not enough (§ 994(t)).
Nothing in the text bars a court from considering serious concerns about guilt or about the fairness of a sentence alongside other factors (rehabilitation, age, conduct in prison, etc.) when deciding whether extraordinary and compelling reasons exist.
The government responds that allowing trial-error claims into the compassionate-release framework would effectively create a parallel post-conviction system, undermining the strict rules Congress put in place for habeas and successive petitions. (Oral Argument Transcripts can be accessed here). (Listen to Oral Arguments here).
The Court's eventual ruling in Fernandez will answer a core question: Can § 3582(c)(1)(A) ever be used, even in part, to address doubts about the correctness of a conviction or sentence, or is that strictly off-limits?
III. Rutherford & Carter: Stacked § 924(c) Sentences and Nonretroactive Changes in Law
Rutherford and Carter present a different but equally important problem. They do not center on alleged trial error, but on nonretroactive changes in sentencing law and the enormous disparities those changes create.
Both men were convicted of multiple firearm offenses under 18 U.S.C. § 924(c) in connection with robberies in the mid-2000s. At that time, a second § 924(c) conviction, even within the same case, triggered a mandatory consecutive 25-year term, stacked on top of any other sentences. Courts routinely imposed decades-long sentences as a result.
Rutherford received a total sentence of roughly 42½ years.
Carter received a sentence of about 70 years.
Later, the First Step Act changed the rule: the 25-year enhancement now applies only when a second § 924(c) offense is committed after a prior § 924(c) conviction has become final. In other words, Congress ended “front-end stacking” in a single prosecution. But Congress made this change prospective only, meaning that people like Rutherford and Carter, already sentenced under the old rule, did not automatically benefit. Rutherford and Carter asked for compassionate release, arguing that:
Their sentences are dramatically higher than what similarly situated defendants would face today. That disparity, stemming from a reform Congress itself enacted, is an “extraordinary and compelling” reason to reduce their sentences, especially when combined with their age, rehabilitation, and conduct in custody.
The Third Circuit rejected this argument, relying on its earlier decision in United States v. Andrews, No. 23-8046, 2025 U.S. App. LEXIS 26904 (2d Cir. Oct. 16, 2025), which held that courts may not treat nonretroactive changes in § 924(c) penalties as extraordinary and compelling reasons under § 3582(c)(1)(A).
By contrast, several other circuits, the First, Fourth, Ninth, and Tenth, have allowed district courts to consider such nonretroactive changes as part of the compassionate-release analysis, at least in cases involving unusually long sentences served for many years. Adding another wrinkle, the Sentencing Commission's 2023 amendments to Guideline § 1B1.13 expressly permit judges, in “narrowly cabined” circumstances, to consider nonretroactive changes in law when deciding whether extraordinary and compelling reasons exist, typically where the defendant has already served a substantial portion of an unusually long sentence. Some courts (including the Sixth Circuit in Bricker) have pushed back, calling that aspect of the guideline an improper expansion of Commission authority.
The Supreme Court in Rutherford/Carter thus faces two related questions:
Statutory discretion – Does the text of § 3582(c)(1)(A) allow district courts to treat large sentencing disparities created by nonretroactive statutory reforms as extraordinary and compelling reasons for a reduction?
Commission authority – Did the Sentencing Commission act within its delegated powers when it adopted a policy statement permitting limited consideration of such nonretroactive changes?
IV. Defining “Extraordinary and Compelling”: The Core Legal Clash
Taken together, Fernandez and Rutherford require the Court to decide how flexible the phrase “extraordinary and compelling reasons” really is. A few key themes run through both sets of cases:
Text versus structure:
The defendants emphasize the plain meaning of “extraordinary and compelling,” arguing that Congress intentionally left the phrase open-ended, with only one explicit restriction (rehabilitation alone is not enough).
The government stresses how this phrase fits within the larger post-conviction framework, warning that using § 3582(c)(1)(A) to revisit alleged trial errors or nonretroactive reforms would undercut the limits Congress placed on habeas, successive petitions, and retroactivity.
A. Role of the Sentencing Commission
Defendants and many amici argue that the Commission's updated § 1B1.13 reflects what Congress always intended: a broad, evolving safety valve for unusually harsh or outdated sentences. Critics, including some courts of appeals, see the Commission's latest policy as an overreach, especially its allowance for considering changes in law in the absence of explicit retroactivity.
Finality versus mercy
The government leans on finality: once appeals and habeas are over, sentences should be altered only in narrow, traditional situations (serious illness, advanced age, family emergencies).
Defendants, and many reform advocates, emphasize that Congress amended § 3582(c)(1)(A) in the First Step Act precisely because the old system proved too rigid, and that compassionate release must be flexible enough to address not just health crises but also exceptional fairness problems created by evolving law and new understandings of a case.
B. Interaction with habeas and direct appeal
In Fernandez, the Court must decide whether compassionate release can factor in doubts about guilt or trial fairness without transforming § 3582(c)(1)(A) into a back-door habeas remedy.
In Rutherford, the Court must decide whether and how compassionate release can soften the hard edges of nonretroactive sentencing reform without negating Congress's choice to limit retroactivity.
V. What This Means for Federal Prisoners and Practitioners
The Court has already heard argument in these cases; a decision is expected in 2026. Whatever the outcome, the rulings will have practical consequences for thousands of people in federal custody and for the lawyers who assist them. Depending on how broadly or narrowly the Court reads § 3582(c)(1)(A), we may see:
A. Narrow reading of “extraordinary and compelling”
Trial-error arguments and nonretroactive changes in law are largely off-limits. Compassionate release returns to a more traditional focus on medical issues, age, family circumstances, and a very small group of “other” reasons closely aligned with those categories.
Many currently pending or planned motions grounded in sentencing disparity or possible innocence may become significantly harder to pursue.
B. Middle-ground approach
The Court might reject using compassionate release as a full substitute for habeas, while still allowing judges to consider extreme disparities from nonretroactive reforms when combined with other factors (time served, rehabilitation, low risk, etc.). That would preserve some flexibility for unusually harsh sentences while keeping ordinary trial-error claims in the habeas lane.
C. Broad reading of “extraordinary and compelling”
Judges retain wide latitude to consider both nonretroactive changes in law and serious concerns about the fairness of a conviction or sentence, so long as they also weigh public-safety factors and the § 3553(a) sentencing considerations. Compassionate release remains a robust tool for addressing outlier sentences and evolving understandings of justice, particularly for prisoners who have already served substantial time.
For now, practitioners drafting § 3582(c)(1)(A) motions should:
· Continue to develop the full picture: health, age, family responsibilities, rehabilitation, disciplinary record, programming, and support plans on release.
· Where relevant, carefully document sentencing disparities, especially in § 924(c) and other areas affected by the First Step Act and later reforms.
· Be realistic: until the Supreme Court rules, some circuits remain hostile to treating nonretroactive changes or alleged errors as extraordinary and compelling reasons. Motions in those jurisdictions should be framed with particular care.
VI. Federal Research Group's Perspective
From our vantage point, these cases are not about opening the floodgates; they are about clarifying whether § 3582(c)(1)(A) can function as the flexible safety valve Congress intended, or whether its role will be limited to a band of medical and family emergencies while other extraordinary injustices remain untouched. Until the Supreme Court speaks, we encourage defendants, families, and counsel to stay informed, preserve arguments grounded in sentencing disparity and fairness, and carefully track developments in their circuits.
As the law continues to evolve, our role at Federal Research Group is to help you understand the landscape, identify viable arguments under § 3582(c)(1)(A), and present those arguments in a clear, well-researched, and respectful manner so that when the door to compassionate release is open, you are ready to walk through it.
Director of Operations
Federal Research Group, Inc.
9100 South Dadeland Blvd., Suite 1500
Miami, FL 33156
(305) 242-0770 Office
(786) 592-1114 Fax
Document Retrieval Request:
BOP Inmate Emails:
Comments
There are no comments for this post. Be the first and Add your Comment below.
Leave a Comment