United States v. David McCall, Jr.

RECOMMENDED FOR PUBLICATION
Pursuant to Sixth Circuit I.O.P. 32.1(b)
File Name: 22a0274p.06

UNITED STATES COURT OF APPEALS
FOR THE SIXTH CIRCUIT

┐
UNITED STATES OF AMERICA,
│
Plaintiff-Appellee, │
> No. 21-3400
│
v. │
│
DAVID E. MCCALL, JR., │
Defendant-Appellant. │
┘

On Petition for Rehearing En Banc.
United States District Court for the Northern District of Ohio at Cleveland;
No. 1:13-cr-00345-41—Christopher A. Boyko, District Judge.

Argued En Banc: June 8, 2022

Decided and Filed: December 22, 2022

Before: SUTTON, Chief Judge; MOORE, COLE, CLAY, GIBBONS, GRIFFIN,
KETHLEDGE, WHITE, STRANCH, DONALD, THAPAR, BUSH, LARSEN,
NALBANDIAN, READLER, and MURPHY, Circuit Judges.*
_________________

COUNSEL
ARGUED EN BANC: Vanessa F. Malone, OFFICE OF THE FEDERAL PUBLIC
DEFENDER, Akron, Ohio, for Appellant. Eric J. Feigin, UNITED STATES DEPARTMENT
OF JUSTICE, Washington, D.C., for Appellee. ON SUPPLEMENTAL BRIEF: Vanessa F.
Malone, OFFICE OF THE FEDERAL PUBLIC DEFENDER, Akron, Ohio, for Appellant.
Rebecca Chattin Lutzko, Matthew B. Kall, UNITED STATES ATTORNEY’S OFFICE,
Cleveland, Ohio, for Appellee.
NALBANDIAN, J., delivered the opinion of the court in which SUTTON, C.J., and
GRIFFIN, KETHLEDGE, THAPAR, BUSH, LARSEN, READLER, and MURPHY, JJ., joined.
MOORE, J. (pp. 25–36), delivered a separate dissenting opinion in which COLE, CLAY,

*
Judge White, who took senior status on June 13, 2022, and Judge Donald, who took senior status on
September 28, 2022, were in regular active service at the time that this case was heard en banc. They have
continued to participate in the decision of this case.
 No. 21-3400 United States v. McCall Page 2

WHITE, STRANCH, and DONALD, JJ., joined. GIBBONS, J. (pp. 37–39), also delivered a
separate dissenting opinion.

_________________

OPINION
_________________

NALBANDIAN, Circuit Judge. David McCall, a federal prisoner with a long drug-
dealing career, pleaded guilty to a conspiracy charge involving heroin possession and
distribution in 2015. Five years into his 235-month sentence, McCall moved for compassionate
release under 18 U.S.C. § 3582(c)(1). Although he cited several “extraordinary and compelling
reasons” justifying a sentence reduction under that statute, the heart of his motion rested on our
opinion in Havis. Invoking that opinion, McCall argued that if he were sentenced today, he
would receive a shorter sentence than he received in 2015. The district court denied his motion,
reasoning that a nonretroactive change in sentencing law could not amount to an “extraordinary
and compelling” reason for a sentence reduction. We agree and affirm.

I.

David McCall served as a middleman in a sprawling drug-trafficking conspiracy. From
2011 to 2013, he supplied lower-level dealers in Cleveland with heroin and cocaine smuggled in
from Chicago and Atlanta. The United States indicted him, along with 59 of his coconspirators,
in a 196-count indictment in 2013. Faced with multiple charges, McCall struck a deal with the
government. He pleaded guilty to one count of conspiracy to possess with intent to distribute
heroin. And in exchange, the United States agreed to drop its remaining facilitation and
substantive distribution charges.

This plea proved only the latest chapter in McCall’s drug-dealing career. Beginning in
1994, McCall racked up multiple Ohio felony convictions, many of them for drug trafficking.
For sentencing purposes, these convictions cemented McCall’s status as a career offender and
raised his base-offense level from 24 to 34. A few more adjustments resulted in a Sentencing
Guidelines range of 188–235 months. The government, emphasizing McCall’s extensive
criminal history, urged the district court to sentence McCall to 235 months. McCall, for his part,
 No. 21-3400 United States v. McCall Page 3

didn’t object to the career-offender classification. In the end, the district court sentenced McCall
to 235 months’ imprisonment and to four years of supervised release.

McCall served five years of that sentence before he moved for a sentence reduction under
the compassionate-release statute. That statute allows a district court to lower a defendant’s
sentence if (among other things) “extraordinary and compelling reasons” warrant a reduction.
18 U.S.C. § 3582(c)(1)(A)(i).

Two changes in law spurred McCall’s motion. First, Congress amended the
compassionate-release statute. Historically, only the Bureau of Prisons could move for a
sentence reduction. See United States v. Ruffin, 978 F.3d 1000, 1003 (6th Cir. 2020). The First
Step Act of 2018 changed this by allowing prisoners to file their own requests when the “Bureau
refused to do so.” Id. at 1003–04. Second, our en banc court decided United States v. Havis,
927 F.3d 382 (6th Cir. 2019) (en banc) (per curiam). That decision held that attempted drug-
trafficking offenses are not controlled substance offenses sufficient to trigger the career-offender
enhancement. See id. at 387; see also United States v. Cordero, 973 F.3d 603, 626 (6th Cir.
2020) (concluding that drug-conspiracy convictions do not fall within the Guidelines’ definition
of “controlled substance offense” after Havis for purposes of determining whether an individual
is a career offender).

Invoking these changes, McCall raised five “extraordinary and compelling reasons” that
he believed merited a lower sentence, three of which are relevant here. (R. 2109, Mot. For
Sentence Reduction, PageID 17033.) First, his risk of contracting COVID-19. Second, his
rehabilitative efforts in prison. And third (and most importantly), Havis’s effect on his status as
a career offender. McCall argued that had he been sentenced after Havis, most of his prior
convictions would not have qualified as predicate offenses for the career-offender enhancement.
And without that enhancement, McCall asserted his 235-month sentence stood in “stark contrast
. . . to the sentence he would likely receive . . . today.” (R. 2134, Supp. Mot., PageID 17219.)

The district court denied McCall’s petition. Although McCall mentioned the COVID-19
pandemic, he supplied no “health concern that put[] him at risk . . . .” (R. 2143, Order, PageID
17307.) As for McCall’s Havis argument, the court noted that Havis “[was] not retroactive, nor
 No. 21-3400 United States v. McCall Page 4

would it support a claim on collateral relief.” (Id.) And it declined to let McCall “sidestep
normal post-conviction requirements” with a compassionate-release motion. (Id.) That left
rehabilitation, which could not support McCall’s motion on its own. See 28 U.S.C. § 994(t)
(“Rehabilitation of the defendant alone shall not be considered an extraordinary and compelling
reason.”).

McCall appealed, arguing that a “subsequent legal clarification” could satisfy the
“extraordinary and compelling” reason requirement. (Appellant Br. at 7.)

By the time we resolved his appeal, at least five of our published decisions had
considered, albeit with some nuance and distinctions, whether a nonretroactive change in
sentencing law could support a finding of “extraordinary and compelling reasons” warranting
relief. 18 U.S.C. § 3582(c)(1)(A)(i). Four of those decisions answered no, and one decision
answered a qualified yes. Compare United States v. Tomes, 990 F.3d 500, 505 (6th Cir. 2021)
(no), cert. denied, 142 S. Ct. 780 (2022), United States v. Wills, 997 F.3d 685, 688 (6th Cir.
2021) (no), United States v. Jarvis, 999 F.3d 442, 443–44 (6th Cir. 2021) (no), cert. denied,
142 S. Ct. 760 (2022), and United States v. Hunter, 12 F.4th 555, 564 (6th Cir. 2021) (no), cert.
denied, 142 S. Ct. 2771 (2022), with United States v. Owens, 996 F.3d 755, 763 (6th Cir. 2021)
(explaining that the “combination” of a nonretroactive change in sentencing and other factors
could “warrant[]” compassionate release). When it came to McCall’s appeal, a divided panel
answered yes, reversing and remanding so that the district court could consider “the actual
impact of Havis with respect to McCall’s prior state convictions.” United States v. McCall,
20 F.4th 1108, 1115 (6th Cir. 2021).

We granted en banc review to resolve the “intractable” “intra-circuit split” created by the
decision. Id. at 1116 (Kethledge, J., dissenting).

II.

We first stop to consider the history of compassionate release, along with its current
framework. Hardly a new remedy, the modern-day version of the statute dates to the Sentencing
Reform Act of 1984. Pub. L. No. 98-473, Title II, ch. II, § 212, 98 Stat. 1837. That Act marked
 No. 21-3400 United States v. McCall Page 5

the culmination of a reform movement dedicated to reducing perceived uncertainty and disparity
that had been a part of federal sentencing law for decades.

The parole system was partially to blame for these problems. First established by
Congress in 1910, the system vested discretion in federal parole authorities, not judges, to decide
when a prisoner’s sentence ended. See An Act to Parole United States Prisoners, and for Other
Purposes, ch. 387, 36 Stat. 819, 819–21 (1910) (creating a federal parole system). Designed with
rehabilitation in mind, the goals of the parole system were twofold: provide “the incentive for
rehabilitation” and “the possibility of an expert determination that a prisoner had been
rehabilitated and should be released from confinement.” Kate Stith & José A. Cabranes, Fear of
Judging: Sentencing Guidelines in the Federal Courts 18 (1998); see also Mistretta v. United
States, 488 U.S. 361, 363 (1989) (“Both indeterminate sentencing and parole were based on
concepts of the offender’s possible, indeed probable, rehabilitation . . . .”).

Most agree that these goals, though noble in theory, failed in fact. Rehabilitation, long
the cornerstone of federal sentencing and the parole system, fell out of favor with scholars,
reformers, and critics of all stripes by the 1970s.1 Critiques abounded. Some blamed the parole
system’s uncertainty and indeterminacy for increasing anxiety among prisoners. See Kate Stith
& Steve Y. Koh, The Politics of Sentencing Reform: The Legislative History of the Federal
Sentencing Guidelines, 28 Wake Forest L. Rev. 223, 227 (1993). And many faulted it for
producing “[s]erious disparities” in sentences. Mistretta, 488 U.S. at 365; see also Stith & Koh,
supra, at 227.

Congress responded with the Comprehensive Crime Control Act of 1984. A watershed
statute, it left few areas of federal criminal law untouched. Sentencing law proved no exception.
Title II of the scheme, the Sentencing Reform Act mentioned above, lived up to its name and

1
See, e.g., Mistretta, 488 U.S. at 365 (“Rehabilitation as a sound penological theory came to be questioned
and, in any event, was regarded by some as an unattainable goal for most cases.”); Francis A. Allen, The Decline of
the Rehabilitative Ideal: Penal Policy and Social Purpose (1981); Jon O. Newman et al., Parole Release
Decisionmaking and the Sentencing Process, 84 Yale L.J. 810, 828 (1975) (describing the Board of Parole’s
“abandonment of rehabilitation as a primary goal of the parole system”); Robert Martinson, What Works? –
Questions and Answers About Prison Reform, 35 Pub. Int. 22, 25 (1974) (“With few and isolated exceptions, the
rehabilitative efforts that have been reported so far have had no appreciable effect on recidivism.” (emphasis
altered)); Norval Morris, The Future of Imprisonment 14 (1974) (“‘Rehabilitation’ . . . must cease to be a purpose of
the prison sanction.”).
 No. 21-3400 United States v. McCall Page 6

represented the results of a movement determined to eliminate rehabilitative goals and
indeterminacy in sentencing. See Stith & Cabranes, supra, at 40 (describing the features of the
Sentencing Reform Act of 1984 that showed Congress’s intent to “eliminat[e] rehabilitation as a
purpose of incarceration”). To that end, the Act abolished parole and prohibited a district court
from “modify[ing] a term of imprisonment once it has been imposed . . . .” 18 U.S.C. § 3582(c).

Against this general prohibition, Congress carved out a few exceptions. The one relevant
here—which we call compassionate release—allowed a district court to lower a prisoner’s
sentence if four conditions (three substantive, one procedural) were met. On the procedural side,
the Sentencing Reform Act gave the Bureau of Prisons exclusive power to decide who could
apply for compassionate release. A court could reduce a sentence only “upon motion of the
Director of the Bureau of Prisons.” Id. § 3582(c)(1)(A) (2002) (amended 2018). So without the
Bureau of Prisons’ support, a prisoner could not obtain relief.

As for substance, the statute required a district court to make three findings. First, that
“extraordinary and compelling reasons warrant[ed]” a sentence reduction. Id. § 3582(c)(1)(A)(i).
Second, the “reduction [was] consistent with applicable policy statements issued by the
Sentencing Commission . . . .” Id. § 3582(c)(1)(A). And third, the § 3553(a) factors supported a
lower sentence. Id. As for what qualified as “extraordinary and compelling,” Congress didn’t
say. Instead, it delegated the task to the Sentencing Commission, instructing it to issue “general
policy statements . . . describ[ing] what should be considered extraordinary and compelling
reasons for a sentence reduction, including the criteria to be applied and a list of specific
examples.” 28 U.S.C. § 994(t). Consistent with the Act’s reform-oriented goals, Congress
instructed that “[r]ehabilitation of the defendant alone shall not be considered an extraordinary
and compelling reason.” Id.

This framework stayed intact for close to 30 years. But during that time, compassionate-
release motions were rare. This slow trickle was due to the action (and inaction) of two entities.
First, the Sentencing Commission took its time issuing the “general policy statement[]” Congress
assigned to it. See id. Twenty-two years, in fact. When the policy statement finally arrived in
2006, it did little but parrot the text of the compassionate-release statute. And it wasn’t until
2007 that the Sentencing Commission added commentary describing “specific examples” of
 No. 21-3400 United States v. McCall Page 7

“what should be considered extraordinary and compelling reasons” for relief. Id.; see U.S.S.G.
§ 1B1.13 cmt. n.1(A)–(D).2 Second, the Bureau of Prisons, although tasked with processing
applications and filing motions for meritorious ones, did so on rare occasion. On average, only
24 prisoners a year benefited from compassionate release. See United States v. Elias, 984 F.3d
516, 518 (6th Cir. 2021).

This brings us to the First Step Act of 2018. See Pub. L. 115-391, Title VI, 132 Stat.
5194. Sweeping in its coverage, the First Step Act made many changes to sentencing law. But it
modified only one aspect of the compassionate-release statute. Recall that under the original
version of the statute, only the Bureau of Prisons could move for compassionate release. See
18 U.S.C. § 3582(c)(1)(A) (2002) (amended 2018). The First Step Act, in a section titled
“Increasing the Use and Transparency of Compassionate Release,” eliminated this gatekeeping
role. See First Step Act of 2018, Pub. L. No. 115-391, 132 Stat. 5194, 5239. It allowed prisoners
to move for relief when the Bureau of Prisons declined to act. See 18 U.S.C. § 3582(c)(1)(A).

Although the First Step Act altered the process for filing a compassionate-release motion,
it left the substantive requirements for obtaining relief intact. We describe this process as a
“three-step inquiry.” United States v. Jones, 980 F.3d 1098, 1101 (6th Cir. 2020); see Ruffin,
978 F.3d at 1004–05. Prisoners still must show that “extraordinary and compelling” reasons
warrant the reduction. 18 U.S.C. § 3582(c)(1)(A)(i). Courts must confirm that any sentence
reduction “is consistent with applicable policy statements issued by the Sentencing
Commission.” Id. § 3582(c)(1)(A). And defendants must persuade the district judge to grant the
motion after the court considers the § 3553(a) factors. Id.

2
The Application Notes list the following examples: (1) medical condition of the defendant, where the
defendant is suffering from “a terminal illness,” a “serious physical or medical condition,” a “serious functional or
cognitive impairment,” or the defendant is “experiencing deteriorating physical or mental health because of the
aging process,” and this condition “substantially diminishes the ability of the defendant to provide self-care within
the environment of a correctional facility and from which he or she is not expected to recover”; (2) age of the
defendant, if the defendant is “at least 65 years old,” is “experiencing a serious deterioration in physical or mental
health because of the aging process,” and “has served at least 10 years or 75 percent of his or her term of
imprisonment, whichever is less”; (3) family circumstances, including the “death or incapacitation of the caregiver
of the defendant’s minor child,” the “incapacitation of the defendant’s spouse or registered partner when the
defendant would be the only available caregiver for the spouse or registered partner”; and (4) “[o]ther”
“extraordinary and compelling” reasons “[a]s determined by the Director of the Bureau of Prisons.” U.S.S.G.
§ 1B1.13 cmt. n.1(A)–(D).
 No. 21-3400 United States v. McCall Page 8

As of today, the Sentencing Commission has not adopted a policy statement that applies
when a “defendant (as opposed to the Bureau of Prisons) files” a motion. United States v.
McKinnie, 24 F.4th 583, 586 (6th Cir. 2022); see Jones, 980 F.3d at 1109 (holding that the
existing policy statement, found at U.S.S.G. § 1B1.13, is not applicable to defendant-filed
motions). So in practice, a district court need only consider whether extraordinary and
compelling reasons exist and whether the § 3553(a) factors support a sentence reduction. See
Jones, 980 F.3d at 1108.

With this framework in mind, we turn to the merits.

III.

The First Step Act changed compassionate release in two ways. The first change affected
the process of filing motions by “add[ing] prisoners to the list of persons who may file . . . .” See
United States v. King, 40 F.4th 594, 596 (7th Cir. 2022). The second change—an unintended
consequence of the first—affected substance. Until the Sentencing Commission updates its
policy statement to include defendant-filed motions, district courts retain discretion to define
“extraordinary and compelling” without reference to the Sentencing Commission’s guidance.
See Elias, 984 F.3d at 519–20.3

Everyone agrees the sum of these changes enhanced district courts’ discretion to grant
compassionate release. The key question is how much. McCall pushes for an approach that
would allow a district court to find that any change in federal sentencing law—even one
Congress or a court made expressly nonretroactive—can qualify as an extraordinary and
compelling reason for a sentence reduction (or a district court may at least consider it in
combination with other factors to add up to extraordinary and compelling reasons). The
government, on the other hand, advocates for a narrower reading. It contends that a district court

3
Congress has delegated to the Commission the task of “describ[ing] what should be considered
extraordinary and compelling reasons . . . .” 28 U.S.C. § 994(t). Whether the Commission could issue a new policy
statement that describes “extraordinary and compelling reasons” in a way that is inconsistent with our interpretation
of the statute’s language is a question that we need not resolve in the absence of an applicable statement. See Nat’l
Cable & Telecomms. Ass’n v. Brand X Internet Servs., 545 U.S. 967, 982 (2005); cf. United States v. Home
Concrete & Supply, LLC, 566 U.S. 478, 487–90 (2012) (plurality opinion); see also Ruffin, 978 F.3d at 1006–08; cf.
United States v. Riccardi, 989 F.3d 476, 485 (6th Cir. 2021) (explaining that the Commission’s commentaries do not
deserve deference unless the Guidelines are “genuinely ambiguous” (internal quotation omitted)).
 No. 21-3400 United States v. McCall Page 9

cannot consider nonretroactive changes in sentencing law either alone or in combination with
other factors to find extraordinary and compelling reasons exist.

The district court sided with the government, and we do too. Consistent with the text of
the compassionate-release provision, along with the principles, structure, and history of federal
sentencing law, we hold that nonretroactive changes in sentencing law cannot be “extraordinary
and compelling reasons” that warrant relief.

IV.

Text. Our analysis starts, as it must, with the text of the compassionate-release statute.
See Jimenez v. Quarterman, 555 U.S. 113, 118 (2009). With no statutory definition of
“extraordinary and compelling reasons” to guide us, Elias, 984 F.3d at 518–20, we interpret the
phrase “in accord with the ordinary public meaning of its terms at the time of its enactment,”
Bostock v. Clayton Cnty., 140 S. Ct. 1731, 1738 (2020); see also Tanzin v. Tanvir, 141 S. Ct.
486, 491 (2020) (“Without a statutory definition, we turn to the phrase’s plain meaning at the
time of enactment.”). And to do so, we rewind the clock to the time of the Sentencing Reform
Act’s adoption, here 1984. At that time, most understood “extraordinary” to mean “most
unusual,” “far from common,” and “having little or no precedent.” Extraordinary, Webster’s
Third New International Dictionary 807 (1971). “Compelling,” for its part, referred to “forcing,
impelling, driving.” Id. at 463.

At first glance, these common-sense definitions only seem to reiterate what we already
know. Of course, an “extraordinary and compelling reason” is one that is unusual, rare, and
forceful. But in a vacuum, the phrase “extraordinary and compelling” does little to illuminate
the specific type of unique or rare reason that might justify relief. This leads us back to the
question with which we began. Does a district court’s discretion to define “extraordinary and
compelling” encompass any reason—legal or factual—it finds convincing?

Two background principles of federal sentencing law help to provide an answer. The
first is finality. “[E]ssential to the operation of our criminal justice system,” finality gives
criminal law its “deterrent effect.” Teague v. Lane, 489 U.S. 288, 309 (1989) (plurality op.).
Recognizing the importance of this principle, Congress gave it statutory weight. Because a
 No. 21-3400 United States v. McCall Page 10

sentence of imprisonment “constitutes a final judgment,” federal law generally prohibits a
district court from “modify[ing]” it “once it has been imposed.” 18 U.S.C. § 3582(b)–(c); see
also Dillon v. United States, 560 U.S. 817, 824 (2010). The second is nonretroactivity. Federal
sentencing law presumes that changes in sentencing law are not retroactive. McKinnie, 24 F.4th
at 587. Drawing from these two principles, the Supreme Court has explained that the “ordinary
practice is to apply new penalties to defendants not yet sentenced, while withholding that change
from defendants already sentenced.” Dorsey v. United States, 567 U.S. 260, 280 (2012).

Framed against this background, the text of the compassionate-release statute gives way
to a basic inference: What is “ordinary” and routine cannot also be extraordinary and compelling.
See Wills, 997 F.3d at 688. After all, prospective changes in federal sentencing law are far from
an extraordinary event. See United States v. Crandall, 25 F.4th 582, 586 (8th Cir. 2022). And
unless Congress expressly says so, those changes do not apply to “defendants already
sentenced.” Dorsey, 567 U.S. at 280. Likewise, new caselaw often gives fresh meaning to
statutes or the Sentencing Guidelines. But the Supreme Court has repeatedly reiterated that
judicial decisions announcing new rules of criminal procedure ordinarily do not provide
retroactive relief on collateral review. See Edwards v. Vannoy, 141 S. Ct. 1547, 1555 (2021).
That new statutes and caselaw apply only to “defendants not yet sentenced” is the expected
outcome in our legal system. Dorsey, 567 U.S. at 280. And what is expected cannot be
extraordinary.

What’s more, we find little compelling about the duration of a lawfully imposed
sentence. This is because such a sentence represents “the exact penalt[y] that Congress
prescribed and that a district court imposed for [a] particular violation[] of a statute.” United
States v. Thacker, 4 F.4th 569, 574 (7th Cir. 2021), cert. denied, 142 S. Ct. 1363 (2022). That a
district court might impose a different sentence than one of its predecessors hardly seems the
kind of “forceful, impelling, [or] driving” reason that could justify compassionate release. See
Compelling, Webster’s Third New International Dictionary 463 (1971); see also United States v.
Andrews, 12 F.4th 255, 260–61 (3d Cir. 2021); United States v. Maumau, 993 F.3d 821, 838
(10th Cir. 2021) (Tymkovich, C.J., concurring) (“[T]he imposition of a sentence that was not
 No. 21-3400 United States v. McCall Page 11

only permissible but statutorily required at the time is neither an extraordinary nor a compelling
reason to now reduce that same sentence.”).

Viewed in this light, the phrase “extraordinary and compelling reasons” comes into
sharper focus. What is ordinary—the nonretroactivity of judicial precedent announcing a new
rule of criminal procedure like Havis—is not extraordinary. And what is routine—a criminal
defendant like McCall serving the duration of a lawfully imposed sentence—is not compelling.

Structure. The structure of federal sentencing law reinforces our conclusion. Viewed as
a whole, that body of law makes one thing clear: When Congress wants a change in sentencing
law to have retroactive effect, it explicitly says so. To see this reality at work, we need look no
further than the Sentencing Reform Act of 1984 and the First Step Act of 2018.

Consider the Sentencing Reform Act of 1984 first. It listed two exceptions to its general
prohibition on sentence modifications. See 18 U.S.C. § 3582(c). The first was the compassionate-
release provision. Id. § 3582(c)(1). The second allowed for the retroactive application of certain
Sentencing Guideline amendments that provide for shorter terms of imprisonment. Id.
§ 3582(c)(2); see also Dillon, 560 U.S. at 821. One of these exceptions is not like the other.
Why would the same Congress, skeptical of sentence modifications as a general rule, provide for
the retroactive application of specific changes in sentencing law in § 3582(c)(2) if the
compassionate-release exception already covered all legal developments, retroactive or not, in
§ 3582(c)(1)(A)?

The First Step Act begs a similar question. Aside from amending the compassionate-
release statute, the First Step Act reduced the penalties for certain drug crimes in § 401, limited
the stacking of mandatory minimums for certain repeat offenses in § 403, and provided for the
retroactive application of lower sentences for other drug crimes in § 404(b). See Pub. L. No.
115-391, §§ 401, 403, 404, 132 Stat. 5220–22. When it came to § 401 and § 403, Congress
expressly chose not to apply these changes to defendants sentenced before the passage of the
Act. Id. §§ 401(c), 403(b), 132 Stat. at 5220–22. To hold that these changes could still amount
to an extraordinary and compelling reason for relief “fails to grapple with congressional design,
expressed through the text of the statute, in which Congress chose not to make [certain]
 No. 21-3400 United States v. McCall Page 12

sentencing amendments retroactive.” Jarvis, 999 F.3d at 444. “Why would the same Congress
that” carefully delineated between retroactive and nonretroactive changes to criminal penalties
“somehow mean to use a general sentencing statute from 1984 to unscramble that approach?”
Id. “If every defendant who received a longer sentence than the one he would receive today
became eligible for compassionate release, the balance Congress struck would come to naught.”
Id.

The same reasoning applies with “equal force here.” Hunter, 12 F.4th at 564. True,
McCall’s case involves a nonretroactive judicial decision, rather than a nonretroactive statute.
But we see no reason to take a different approach.4 So a “contrary conclusion” would allow
defendants to avoid “the principal path . . . Congress established for federal prisoners to
challenge their sentences.” Thacker, 4 F.4th at 574. That path, found in 28 U.S.C. § 2255,
“embodie[s] . . . [a] specific statutory scheme authorizing post-conviction relief . . . .” Id. And
that scheme places strict limits on post-conviction claims.

Those limits would prove fatal to McCall for two reasons. He first attacked his
conviction under § 2255 in 2016. This attack proved unsuccessful, so he would need express
authorization to bring a “second or successive” petition. See 28 U.S.C. § 2244(b). But this he
could not get. Congress permits such a motion only if it invokes a “new rule of constitutional
law, made retroactive to cases on collateral review by the Supreme Court” or includes “newly
discovered evidence.” 28 U.S.C.§ 2255(h)(1)–(2). McCall presents no new “evidence” and “no
new rule of constitutional law,” making any second petition dead on arrival. But even without
these “second or successive” limits, id. § 2244(b)(1), McCall could not obtain post-conviction
relief. The Supreme Court has long held that “newly recognized rules of criminal procedure do
not normally apply in collateral review.” Ramos v. Louisiana, 140 S. Ct. 1390, 1407 (2020)
(citing Teague, 489 U.S. at 311–12). And we have similarly explained that a misapplication of

4
As we noted above, judicial decisions that change sentencing practices or criminal penalties are
presumptively nonretroactive. So like the panel below, “[w]e assume . . . that the difference between nonretroactive
[statutory changes to the First Step Act] and nonretroactive changes in sentencing law more broadly is ‘immaterial’”
to this question. McCall, 20 F.4th at 1113 n.4; cf. Hunter, 12 F.4th at 565 (“Nothing in § 3582(c) purports to change
the judicial doctrine that ‘a decision announcing a new rule of criminal procedure ordinarily does not apply
retroactively’ to final judgments.” (citing Edwards, 141 S. Ct. at 1551)).
 No. 21-3400 United States v. McCall Page 13

an advisory guidelines range, including a Havis error, does not present a “cognizable” claim
under § 2255. See Bullard v. United States, 937 F.3d 654, 659, 661 (6th Cir. 2019).

McCall cannot avoid these restrictions on “post-conviction relief” by “resorting to a
request for compassionate release instead.” Crandall, 25 F.4th at 586. We assume Congress
knew of its “specific statutory scheme authorizing post-conviction relief” when it adopted the
compassionate-release statute in 1984 and amended it in 2018. Thacker, 4 F.4th at 574. And we
likewise “assume” Congress was “aware of relevant judicial precedent” limiting the retroactive
application of new rules of criminal procedure when it set the extraordinary and compelling
reasons threshold. Guerrero-Lasprilla v. Barr, 140 S. Ct. 1062, 1072 (2020) (quoting Merck &
Co. v. Reynolds, 559 U.S. 633, 648 (2010)). So if Congress intended the compassionate-release
statute to act as an exception to this post-conviction framework, “it would have made ‘that intent
specific.’” McKinnie, 24 F.4th at 588 (quoting Hunter, 12 F.4th at 565); see also Banister v.
Davis, 140 S. Ct. 1698, 1707 (2020). Yet we can find no such intent in the phrase “extraordinary
and compelling reasons.” This phrase “does not remotely suggest that Congress intended to
effect the monumental change of giving district courts the discretion to treat non-retroactive
precedent as a basis to alter a final judgment (and release a prisoner).” Hunter, 12 F.4th at 566.

Judge Moore questions why we’re discussing habeas because habeas and compassionate
release “serve entirely different purposes . . . .” (Moore Dissent p. 33.) She continues: “When a
court grants a habeas petition, it deems the sentence invalid. . . . By contrast, a grant of
compassionate release does not invalidate the relevant sentence; rather, it recognizes that a
holistic consideration of the defendant’s circumstances entitles them to early release . . . .”
(Moore Dissent p. 33–34.)

But arguing that an intervening change to sentencing law provides an extraordinary and
compelling reason for early release necessarily implicates the validity of the relevant sentence.
See United States v. Jenkins, 50 F.4th 1185, 1202–03 (D.C. Cir. 2022). If someone like McCall
were to prevail on his Havis claim, a court would alter the duration of his sentence because his
sentence would be different today, thereby implying the unlawfulness of his sentence as
originally imposed. See id. at 1202, 1204.
 No. 21-3400 United States v. McCall Page 14

Put another way, a compassionate release petitioner argues that, although his sentence is
lawful, other unrelated factors, such as health or age, counsel in favor of early release. A habeas
petitioner argues that his sentence is unlawful. Entertaining an argument that implies a sentence
is unlawful at the threshold level of a compassionate release analysis blurs these two forms of
relief. And it is precisely because, as Judge Moore notes, habeas and compassionate release
serve different purposes, that we won’t blur these lines in a way that ignores context.

The habeas-channeling rule that we alluded to earlier supports our view. Because habeas
serves as the avenue to attack the lawfulness of confinement, “an inmate may not rely on a
generally worded statute to attack the lawfulness of his imprisonment . . . .” Jenkins, 50 F.4th at
1202; see Preiser v. Rodriguez, 411 U.S. 475, 485–86 (1973). This includes attacks seeking “a
judicial determination that necessarily implies the unlawfulness of the State’s custody.”
Wilkinson v. Dotson, 544 U.S. 74, 81 (2005). All to say, habeas is the appropriate place to bring
challenges to the lawfulness of a sentence. And as we’ve discussed, we do not read
“extraordinary and compelling” to provide an end run around habeas.

This is not to say that “extraordinary and compelling” doesn’t have a “broad range of
meaningful application . . . .” Jenkins, 50 F.4th at 1203. Most obviously, “extraordinary and
compelling” “covers factors like those enumerated by the Sentencing Commission—health, age,
and family circumstances . . . .” Id. And “for those defendants who can show some []
‘extraordinary and compelling’ reason for a sentencing reduction,” considering other factors not
encompassed by “extraordinary and compelling” when analyzing the § 3553(a) factors is a
“superior means” for promoting fairness and case-by-case sentencing decisions “and honor[ing]
the choices Congress [or the courts] made . . . .” Jarvis, 999 F.3d at 445.

Return back to our original insight about the structure of federal sentencing law. When
Congress expects a change in sentencing law to have retroactive application, deviating from both
its principal path for post-conviction relief and general prohibition on sentence modifications, it
makes that intent clear. The term “extraordinary and compelling reasons” displays no such
intent. Without a more explicit instruction, defendants cannot use compassionate release as an
“end run around Congress’s” or courts’ retroactivity decisions. Tomes, 990 F.3d at 505.
 No. 21-3400 United States v. McCall Page 15

History. What text and structure show, history confirms. Since its enactment,
compassionate release has never been understood to cover nonretroactive changes in sentencing
law.

An early prototype of the compassionate-release statute, found in the Parole Commission
and Reorganization Act (the “Parole Act”), supplies our starting point. See Pub. L. No. 94-233,
90 Stat. 219, 223 (1976). It provided that “[a]t any time upon motion of the Bureau of Prisons,
the court may reduce any minimum term to the time the defendant has served.” 18 U.S.C.
§ 4205(g) (repealed).

Broad in scope, this provision gave the Bureau of Prisons wide discretion over the who
and the why of compassionate release. See Turner v. U.S. Parole Comm’n, 810 F.2d 612, 615
(7th Cir. 1987). The Bureau of Prisons, in turn, provided two examples of circumstances
meriting use of that discretion: (1) prison-overcrowding and (2) “particularly meritorious or
unusual circumstances which could not have been reasonably foreseen by the court at the time of
sentencing,” including “an extraordinary change in an inmate’s personal or family situation or if
an inmate becomes severely ill.” Id. at 617 (quoting 28 C.F.R. § 572.40(a) (1986)). Neither of
those scenarios contemplated nonretroactive legal developments. And in practice, the few
defendants who received relief often did so because of their rehabilitative efforts. See United
States v. Banks, 428 F. Supp. 1088, 1089 (E.D. Mich. 1977) (granting reduction when the
defendant displayed “outstanding institutional adjustment,” “[h]is conduct record [was] clear,”
and his work was “uniformly competent”); cf. Turner, 810 F.2d at 615–16, 618 (recognizing the
Bureau’s role in initiating sentencing modification motions under the Parole Act).

The Sentencing Reform Act of 1984 replaced the Parole Act’s provision with a new-and-
improved compassionate-release statute. Perhaps mirroring the Bureau of Prisons’ old
regulations, the updated version limited its scope to inmates whose circumstances presented
“extraordinary” reasons for relief. 18 U.S.C. § 3582(c)(1)(A)(i). As we know, the Sentencing
Commission took close to 22 years to flesh out the contours of that requirement. Still, even
without a policy statement, the Bureau of Prisons continued to move for compassionate release
and courts continued to grant it, even if in narrow circumstances. And, for example, the only
circumstance that the Bureau’s 1998 Program Statement explicitly discusses is “the inmate’s
 No. 21-3400 United States v. McCall Page 16

health.” See Fed. Bureau of Prisons, U.S. Dep’t of Justice, Change Notice No. 5050.46,
Program Statement Concerning Compassionate Release; Procedures for Implementation of
18 U.S.C. 3582(c)(1)(A) & 4205(g), 1(a)(1)(d); 4(c) (1998). To our knowledge, for at least the
first two decades of compassionate release, the “extraordinary and compelling reasons”
requirement did not encompass nonretroactive changes in sentencing law.

The next decade or so followed a similar course. The Sentencing Commission’s long-
awaited policy statement and commentary expanded on the kinds of “extraordinary and
compelling reasons” that could justify a sentence reduction. See U.S.S.G. § 1B1.13 cmt. n.1(A)–
(D). But those reasons never included nonretroactive legal developments. Instead, the growing
categories sounded in variations on a theme. The Sentencing Commission added old age, family
caretaking emergencies, and a catch-all provision for other circumstances approved by the
Bureau of Prisons. Id. § 1B1.13 cmt. n.1(B)–(D). Its policy statement also expanded on the
kinds of medical conditions that amounted to an “extraordinary and compelling” reason for
relief. Id. § 1B1.13 cmt. n.1(A). Of course, this policy statement no longer applies to defendant-
filed motions. But still, it “sheds light on the meaning of extraordinary and compelling reasons.”
Andrews, 12 F.4th at 260; see also Tomes, 990 F.3d at 503 n.1 (describing the policy statement
“as relevant, even if no longer binding”). For over thirty years, that meaning encompassed
changes in fact—like terminal illness and family emergencies—largely unforeseeable at
sentencing.

Congress, no doubt, had this meaning and practice in mind when it amended the
compassionate release statute in 2018. See Parker Drilling Mgmt. Servs., Ltd. v. Newton, 139
S. Ct. 1881, 1890 (2019) (“It is a commonplace of statutory interpretation that Congress
legislates against the backdrop of existing law.” (internal citation and quotation marks omitted)).
Yet its amendment focused on process, not substance. See King, 40 F.4th at 596 (“The First Step
Act did not create or modify the ‘extraordinary and compelling reasons’ threshold for eligibility;
it just added prisoners to the list of persons who may file motions.”). And because Congress left
the substance of the “extraordinary and compelling reasons” requirement intact, we assume the
phrase retained “the meaning it had under the previous version of the statute.” Andrews,
12 F.4th at 260; see also Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of
 No. 21-3400 United States v. McCall Page 17

Legal Texts 322 (2012) (“The clearest application of the prior-construction canon occurs with
reenactments: If a word or phrase . . . has been given a uniform interpretation by inferior courts
or the responsible agency, a later version of that act perpetuating the wording is presumed to
carry forward that interpretation.”). That meaning has never covered nonretroactive changes in
sentencing law. Indeed, nothing in the 30-odd year history of compassionate release “hint[s] that
the sort of legal developments routinely addressed by direct or collateral review [could] qualify a
person for compassionate release.” King, 40 F.4th at 595–96. And nothing in the First Step Act
of 2018 suggests Congress intended to change this substantive status quo with a process-oriented
amendment.5

***

This conclusion brings us to McCall’s appeal. He argues that our decision in Havis, his
risk of contracting COVID-19, and his rehabilitative efforts supply “extraordinary and
compelling” reasons to reduce his sentence. We disagree. Havis, a nonretroactive judicial
decision announcing a new rule of criminal procedure, cannot serve as a basis for relief. Nor,
with “vaccinations [widely] available to federal prisoners,” can McCall’s claims about the
dangers of COVID-19. McKinnie, 24 F.4th at 588. That leaves rehabilitation, which cannot by
itself justify a sentence reduction. See 28 U.S.C. 994(t). Because none of McCall’s reasons
meets the “extraordinary and compelling” threshold for relief, the district court did not err when
it denied his petition for compassionate release.

V.

McCall sees it differently. He raises a host of arguments, invoking new Supreme Court
precedent, the text of the statute, legislative history, and notions of disparity. None succeeds.

5
True, the effect of the First Step Act’s process-oriented amendment was that it gave district courts
discretion without a corresponding policy statement when evaluating defendant-filed motions. See supra p. 8. But,
practically speaking, that’s the case because “[t]he Sentencing Commission lacked a quorum from January 2019
until August 2022, so it has been unable to promulgate a policy statement that applies to such motions.” Jenkins, 50
F.4th at 1193; see supra n.3. And so it’s unlikely that this process-oriented amendment tells us anything about
Congress’s intent to change the meaning of “extraordinary and compelling” or its desire to vest more discretion in
the district courts to do so.
 No. 21-3400 United States v. McCall Page 18

First, he contends that the Supreme Court’s recent decision in Concepcion v. United
States, 142 S. Ct. 2389 (2022), mandates a different result. That case considered another
provision of the First Step Act, § 404(b), which “empowers district courts to lower sentences
imposed for crack-cocaine offenses as if” certain reduced penalties “had been the law during the
original sentencing hearing.” United States v. Maxwell, 991 F.3d 685, 688 (6th Cir. 2021)
(internal citation and quotation marks omitted). The Supreme Court held that if a prisoner
qualifies for resentencing under § 404(b), a district court may consider “intervening changes of
law or fact” in deciding whether and how much to reduce the sentence. Concepcion, 142 S. Ct.
at 2404. And in doing so, the Court emphasized district courts’ historical exercise of “broad
discretion to consider all relevant information” during an initial sentencing or sentence-
modification hearing. Id. at 2398.

McCall claims Concepcion’s holding buoys his reading of the extraordinary and
compelling reason requirement, but we disagree. To start, Concepcion concerned a different and
unrelated provision of the First Step Act that explicitly applied retroactively. The decision said
nothing about the “threshold question [of] whether any given prisoner has established an
‘extraordinary and compelling’ reason for release.” King, 40 F.4th at 596; see United States v.
Beckford, No. 22-10638, 2022 WL 4372553, at *3 (11th Cir. Sept. 22, 2022); United States v.
Bledsoe, No. 22-2022, 2022 WL 3536493, at *2 (3d Cir. Aug. 18, 2022).6

Next, to the extent that Concepcion sheds any light on this case, it supports the
government’s position. Concepcion’s insight goes to what a court may consider after it finds a
defendant meets the threshold requirement for a sentence modification. If that threshold is met,
Concepcion teaches that a district court may consider any number of changes in law and fact
when exercising its discretion to grant or deny the defendant’s motion. See 142 S. Ct. at 2404;
see also id. at 2402 n.6.

6
Judges Moore and Gibbons think that Concepcion has a broader mandate. True, Concepcion articulates a
“broad and expansive view of sentencing discretion.” (Gibbons Dissent p. 37.) But, as both Judges Moore and
Gibbons acknowledge, a district court’s “discretion” is “limit[ed]” by what is “set forth by Congress in a statute or
by the Constitution.” Concepcion, 142 S. Ct. at 2400. And “the compassionate-release statute imposes just such a
limit[] in authorizing a reduced term of imprisonment only for extraordinary and compelling reasons.” Jenkins,
50 F.4th at 1200.
 No. 21-3400 United States v. McCall Page 19

Our approach to compassionate release runs a parallel course. A defendant must first
satisfy the provision’s threshold requirement, showing “some [] ‘extraordinary and
compelling’ reason” justifies “a sentencing reduction.” Jarvis, 999 F.3d at 445. With that hurdle
cleared, Concepcion’s holding comes into play. A district court may “consider subsequent
developments,” legal or factual, “in deciding whether to modify the original sentence and, if so,
in deciding by how much.” Maxwell, 991 F.3d at 691; see also Jarvis, 999 F.3d at 445 (adopting
Maxwell’s approach). “We take the Supreme Court at its word that Concepcion is about the
matters that district judges may consider when they” exercise their discretion to “resentence
defendants,” not whether those defendants qualify for resentencing in the first place. King, 40
F.4th at 596.7

Second, McCall asserts that the text of the compassionate-release statute supports his
broad reading of its terms. To back up his assertion, he relies on what we call the “negative-
implication canon.” Scalia & Garner, supra, at 107. That canon stands for the idea “that the
expression of one thing (i.e., the specified exception) implies the exclusion of other things of the
same sort (i.e., other exceptions).” Caleb Nelson, Statutory Interpretation 89 (2011).

Here, McCall suggests that canon plays out as follows. Although Congress provided no
official definition of what qualifies as an “extraordinary and compelling” reason, instead
delegating the task to the Sentencing Commission, it did articulate a limit on its scope.
“Rehabilitation of the defendant alone” cannot “be considered an extraordinary and compelling

7
Judge Gibbons believes Concepcion “incorporates consideration of nonretroactive changes at the
threshold stage . . . .” (Gibbons Dissent p. 37.) And she relies on the following sentence for support: “In many
cases, a district court is prohibited from recalculating a Guidelines range in light of nonretroactive Guidelines
amendments, but the court may find those amendments to be germane when deciding whether to modify the
sentence at all . . . .” Concepcion, 142 S. Ct. at 2400. This sentence simply supports the holding of Concepcion:
intervening changes in the law can be considered if a prisoner qualifies for resentencing. Id. at 2404. It doesn’t tell
us anything about when, if at all, changes in sentencing law should be considered in the compassionate-release
context. Placing this sentence in context doesn’t lend support to Judge Gibbons’s reading either. See id. at 2400
(referencing three cases that discussed intervening changes to law in the § 3553(a) context); id. at n.6 (“A district
court cannot, however, recalculate a movant’s benchmark Guidelines range in any way other than to reflect the
retroactive application of the Fair Sentencing Act. . . . The district court may then consider postsentencing conduct
or nonretroactive changes in selecting or rejecting an appropriate sentence . . . .”); see also Bledsoe, 2022 WL
3536493, at *2 (recognizing that Concepcion “harmonizes” with the acknowledgement that “‘[i]f a prisoner
successfully shows extraordinary and compelling circumstances, the current sentencing landscape may be a
legitimate consideration for courts at the next step of the analysis when they weigh the § 3553(a) factors.’” (quoting
Andrews, 12 F.4th at 262)).
 No. 21-3400 United States v. McCall Page 20

reason.” 28 U.S.C. 994(t). That Congress articulated one limit on what might be considered
“extraordinary and compelling,” McCall posits, implies the exclusion of any other limit. Put
another way, had Congress wanted to limit the use of nonretroactive changes in sentencing law,
like it did rehabilitation, it would have said so. Without such a statement, McCall finishes, our
limit on nonretroactive changes in sentencing law “is in direct conflict with the statutory text,”
which contains no “support . . . for the categorical exclusion of non-retroactive legislation or
decisions.” (Appellant Supp. Br. at 8, 11.)

As with all arguments that rely on the negative-implication canon, we approach McCall’s
with “great caution.” Scalia & Garner, supra, at 107. This is because “its application depends so
much on context.” Id. Properly understood, the canon comes into play “only when” the limit or
exception specified “can reasonably be thought to be an expression of all that shares in the grant
or prohibition involved.” Id. “[C]ommon sense often suggests when this is or is not so.” Id.;
see also Nelson, supra, at 91 (explaining that courts do not apply the negative-implication canon
“reflexively”; “[i]nstead, they think about whether the relevant statutory language really does
carry the negative implication that application of the [canon] would suggest”).

Here, context and common sense undermine McCall’s argument. Historical context
provides our first clue. Recall the reform movement that led to the Sentencing Reform Act of
1984. That movement rejected rehabilitation “as a sound penological theory” and dismissed the
idea as an “unattainable goal for most cases.” Mistretta, 488 U.S. at 365. Congress designed the
Sentencing Reform Act with this movement in mind, erecting several barriers designed to
eliminate rehabilitation’s outsized influence. It began by abolishing the parole system and
generally prohibiting sentence modifications. Against that general prohibition, Congress left a
few doors ajar. One of those doors, compassionate release, had a barrier of its
own: “Rehabilitation . . . alone” could not meet the “extraordinary and compelling” threshold.
28 U.S.C. § 994(t). This history leads us to a common-sense inquiry. Did Congress’s restriction
on rehabilitation signify an intent to exclude all other limits on what could qualify as an
“extraordinary and compelling” reason? Or did that limit simply reinforce Congress’s statutory
goal of “eliminat[ing] [] rehabilitation as a purpose of incarceration”? See Stith & Cabranes,
 No. 21-3400 United States v. McCall Page 21

supra, at 40. Given the historical context of the Sentencing Reform Act, we find the latter more
likely.

Textual context also supports our holding. McCall asserts Congress placed “no textual
. . . limit”—aside from rehabilitation—on the reasons that may warrant a sentence reduction.
(Appellant Supp. Br. at 4.) But this assertion ignores the explicit textual limitations at the heart
of this appeal—that is, that the reasons be “extraordinary and compelling.” As we have
explained, there is nothing extraordinary about the ordinary operation of our legal system, which
assumes new statutes and caselaw have no retroactive effect. Thus it is the text of the
compassionate-release statute itself, rather than our “re-writing” of it, which “support[s] . . . the
categorical exclusion of non-retroactive legislation or decisions.” (Appellant Supp. Br. at 8, 11.)
So understood, Congress’s restriction on rehabilitation cannot be “an expression of all that shares
in the . . . prohibition involved.” Scalia & Garner, supra, at 107.

Third, McCall next argues that legislative history supports his vast construction of the
compassionate-release statute. But “legislative history is not the law.” Epic Sys. Corp. v. Lewis,
138 S. Ct. 1612, 1631 (2018). And even when courts consider legislative history, they do so
only when it “shed[s] a reliable light on the enacting Legislature’s understanding of otherwise
ambiguous terms.” Exxon Mobil Corp. v. Allapattah Servs., Inc., 545 U.S. 546, 568 (2005); see
also United States v. Woods, 571 U.S. 31, 46 n.5 (2013) (“Whether or not legislative history is
ever relevant, it need not be consulted when . . . the statutory text is unambiguous.”). There is no
such ambiguity here. The text of the compassionate-release statute, informed by the principles,
history, and structure of sentencing law forecloses McCall’s argument.

But even if it did not, McCall’s recitation of legislative history does little to advance his
case. He first cites a Senate Report explaining the Senate Judiciary Committee’s view that “there
may be unusual cases . . . justified by changed circumstances,” that would permit a sentence
reduction, including: “cases of severe illness, cases in which other extraordinary and compelling
circumstances justify a reduction of an unusually long sentence, and some cases in which the
sentencing guidelines for the offense . . . have been later amended to provide a shorter term of
imprisonment.” S. Rep. No. 98-225, at 55–56 (1983).
 No. 21-3400 United States v. McCall Page 22

Put in context, this portion of the Report does not interpret or reference the proposed
compassionate release statute, 18 U.S.C. § 3582. See id. These Committee comments respond
to the Parole Commission’s concerns about its diminished place within the new sentencing
regime and refer to 18 U.S.C. § 3583(c), which vests the courts with authority to determine
whether and when supervised release is appropriate. See id. at 56. To the extent that the
Committee comments here sound in compassionate-release language, the comments distinguish
“cases in which the sentencing guidelines for the offense . . . have been later amended to provide
a shorter term of imprisonment” from “extraordinary and compelling” cases. See id. at 55–56.
So the reference to guideline changes most naturally points to Congress’s other exception to its
general prohibition on sentence modifications, namely: the retroactive application of certain
sentencing guideline amendments that provide for shorter terms of imprisonment. See 18 U.S.C.
§ 3582(c)(2). And what kinds of “other extraordinary and compelling circumstances” would
justify the reduction of an “unusually long sentence”? The Committee doesn’t elaborate here.
“There being, alas, no legislative history to this legislative history, we are left with more room
for conjecture” than clarity. Am. Fin. Grp. & Consol. Subsidiaries v. United States, 678 F.3d
422, 427 (6th Cir. 2012).

McCall’s second piece of legislative history is no more illuminating. It describes
compassionate release as a “safety valve” for cases “in which the defendant’s circumstances are
so changed, such as by terminal illness, that it would be inequitable to continue [] confinement.”
S. Rep. No. 98-225, at 121. This piece of legislative history hurts, rather than helps, McCall’s
cause. The few concrete examples provided by the Senate Report—like “severe” or “terminal
illness”—have little in common with nonretroactive legal developments. S. Rep. No. 98-225, at
55, 121; see also id. at 173 (describing the compassionate-release provision as “permitting” a
“reduction of a term of imprisonment in a compelling case, such as terminal cancer”). If
anything, these examples reinforce the notion that the phrase “extraordinary and compelling
reasons” covers changes in a defendant’s personal circumstances, not changes in law. Taken
altogether, the legislative history is ambiguous at best and damaging to McCall’s case at worst.

Fourth, McCall seems to concede that nonretroactive legal developments cannot,
standing alone, qualify as an “extraordinary and compelling” reason for relief. (Appellant Supp.
 No. 21-3400 United States v. McCall Page 23

Br. at 11 (quoting United States v. McGee, 992 F.3d 1035, 1047–48 (10th Cir. 2021).) But he
nonetheless argues that a new statute or judicial decision can enter the picture when used to
“illuminate the disparity as a component of the extraordinary and compelling reasons criteria,
and [can permit] the district court to consider that disparity along with other individual
circumstances.” (Appellant Supp. Br. at 13 (emphasis added).)

He’s not alone in this position. The First, Ninth, and Tenth Circuits have held that
nonretroactive legal developments can contribute to a finding of “extraordinary and compelling
reasons” when viewed “in combination” with a defendant’s “unique circumstances.” McGee,
992 F.3d at 1048; see United States v. Chen, 48 F.4th 1092, 1098 (9th Cir. 2022); United States
v. Ruvalcaba, 26 F.4th 14, 28 (1st Cir. 2022). Judge Moore advocates for this approach as well.
(Moore Dissent pp. 29–30.) The Fourth Circuit’s position goes a step further. It held that a
nonretroactive statutory change, coupled with the resulting “disparity” between the sentence the
defendant received and “the sentence a defendant would receive today,” may satisfy the
“extraordinary and compelling reason” standard on its own. United States v. McCoy, 981 F.3d
271, 285 (4th Cir. 2020). Different around the edges, all three of these decisions seem to rest on
the common goals of “alleviating unfair and unnecessary sentences as judged by today’s
sentencing laws . . . and of promoting ‘individualized, case-by-case’ sentencing decisions.”
Jarvis, 999 F.3d at 445 (quoting McGee, 992 F.3d at 1047, citing McCoy, 981 F.3d at 285–86).

Although we appreciate these ends, we cannot reconcile this approach with the plain text
of the compassionate-release statute. Congress prospectively amends or updates its criminal-
penalty scheme. The nonretroactivity of judicial precedent like Havis is the rule, not the
exception. That a defendant might receive a different sentence today than he received years ago
represents the routine business of our legal system. These ordinary happenings “cannot supply
an extraordinary and compelling reason to reduce a lawful sentence whose term Congress
enacted, and the President signed, into law.” Thacker, 4 F.4th at 574; see also Crandall, 25 F.4th
at 586 (“The views of a present-day Congress, like those of a present-day sentencing judge,
about the appropriate punishment for a present-day offense do not establish an ‘extraordinary
and compelling reason’ for reducing a sentence imposed years ago.”).
 No. 21-3400 United States v. McCall Page 24

And the reality is that the circuits that agree with Judge Moore’s approach “recognize that
a court may never grant compassionate release based solely on prospective sentencing changes”
because “[s]uch reasoning always runs headlong into Congress’s judgment that the unamended
statute remains appropriate for previously sentenced defendants . . . .” Jenkins, 50 F.4th at 1199;
see McGee, 992 F.3d at 1047–48 (“[T]he possibility of a district court finding the existence of
‘extraordinary and compelling reasons’ based, in part, on a defendant’s pre-First Step Act . . .
sentence . . . does not, in our view, necessarily usurp Congressional power. That said, we also
agree with the Sixth Circuit’s decision in Tomes that the fact that a defendant is serving a pre-
First Step Act mandatory life sentence . . . cannot, standing alone, serve as the basis for a
sentence reduction under § 3582(c)(1)(A).”); cf. Ruvalcaba, 26 F.4th at 27 (“[I]f a district court
were to reduce a sentence solely because . . . [a statute’s] non-retroactive amendments would
have lowered a defendant’s sentence, it might be seen as substituting its own judgment on
retroactivity for Congress’s judgment . . . .”).

Our conclusion, on the other hand, holds true in the singular and in the aggregate.
Nonretroactive legal developments, considered alone or together with other factors, cannot
amount to an “extraordinary and compelling reason” for a sentence reduction. McCall’s case
shows why. He contends that, even if Havis does not amount to an extraordinary and compelling
reason on its own, it meets the standard when combined with COVID-19 and his rehabilitative
efforts. But this line of argument falls flat. “[A]dding a legally impermissible ground to [other]
insufficient factual considerations” cannot “entitle a defendant to a sentence reduction.” Jarvis,
999 F.3d at 444; see also McKinnie, 24 F.4th at 588 (“[W]hy would combining unrelated factors,
each individually insufficient to justify a sentence reduction, amount to more than the sum of
their individual parts?”). Nonretroactive legal developments do not factor into the extraordinary
and compelling analysis. Full stop.8

VI.

For these reasons, we affirm the district court.

8
Of course, we have held that district courts remain free to consider nonretroactive changes in sentencing
law when “balancing the § 3553(a) factors” to decide whether and how much to reduce a qualifying defendant’s
sentence. See Jarvis, 999 F.3d at 445. The government is not contesting that determination here.
 No. 21-3400 United States v. McCall Page 25

_________________

DISSENT
_________________

KAREN NELSON MOORE, Circuit Judge, dissenting. David McCall has served more
than half of a nearly twenty-year sentence. At sentencing, the district court found that McCall’s
sentencing guideline range was 188–235 months and sentenced him to a term of imprisonment of
235 months, the top end of that range. Several years later, a decision of our en banc court made
it likely that McCall would have faced a far shorter guideline range—in McCall’s estimation, a
range of 77–96 months—if he had been sentenced after that decision. See United States v.
Havis, 927 F.3d 382 (6th Cir. 2019) (en banc) (per curiam).1 Today, the majority holds that the
district court could not consider that intervening change in law when evaluating whether McCall
had shown that “extraordinary and compelling reasons” warranted his compassionate release
under 18 U.S.C. § 3582(c)(1)(A). Because the majority’s decision rests on a cramped view of
compassionate release that cannot be reconciled with the expansive statutory scheme Congress
has adopted, I respectfully dissent.

The Supreme Court’s recent decision in Concepcion v. United States, 142 S. Ct. 2389
(2022), provides a roadmap for resolving the issue before us. In that case, the Court addressed a
similar issue in a distinct context: whether district courts could consider nonretroactive changes
in law when adjudicating motions raised under a provision of the First Step Act that allowed for
resentencing of imprisoned individuals who had been convicted of certain offenses involving
crack cocaine. Id. at 2396. The Court began its decision by noting the “longstanding tradition in
American law” that “a judge at sentencing considers the whole person before him or her ‘as an
individual.’” Id. at 2395 (quoting Koon v. United States, 518 U.S. 81, 113 (1996)). Consistent
with that tradition, the Court observed that “[i]t is only when Congress or the Constitution limits
the scope of information that a district court may consider in deciding whether, and to what
extent, to modify a sentence, that a district court’s discretion to consider information is

1
In Havis, our en banc court held that “[t]he [Sentencing] Guidelines’ definition of ‘controlled substance
offense’ does not include attempt crimes.” 927 F.3d at 387. If McCall had been sentenced after Havis, he would not
have received the career-offender enhancement that he argues increased his sentencing range from 77–96 months to
188–235 months.
 No. 21-3400 United States v. McCall Page 26

restrained.” Id. at 2396. Finding that Congress “did not contravene this well-established
sentencing practice” in the First Step Act, the Court held that district courts were bound to
consider (though free to accept or reject in their discretion) nonfrivolous arguments premised on
nonretroactive intervening changes in law when adjudicating motions under the relevant
provision of the Act. Id. at 2401–05.

Although the majority questions the decision’s relevance, Maj. Op. 18, the principles
recognized in Concepcion help frame the issue presented in this case. The en banc majority
belabors the fact that the provision of the First Step Act addressed in Concepcion is different
from the one we confront here, see id., but the Court’s analysis implicated broader principles
relevant to all “initial sentencings” and “sentencing modification hearings” irrespective of the
statute authorizing the proceedings, Concepcion, 142 S. Ct. at 2399. Indeed, as Judge Gibbons’s
separate dissent observes, Concepcion speaks directly to what a district court may consider
“when deciding” both “whether to modify a sentence at all, and if so, to what extent.” Id. at
2400 (emphasis added). In line with the principles recognized in Concepcion, our review of
McCall’s request for a sentence modification must begin with the default rule that “[t]he only
limitations on [the] [district] court’s discretion to consider any relevant materials . . . in
modifying that sentence are those set forth by Congress in a statute or by the Constitution.” Id.;
see also United States v. Chen, 48 F.4th 1092, 1095–96 (9th Cir. 2022) (adopting a similar frame
of analysis); United States v. Arriola-Perez, No. 21-8072, 2022 WL 2388418, at *2 (10th Cir.
July 1, 2022); United States v. Brice, No. 21-6776, 2022 WL 3715086, at *2 (4th Cir. Aug. 29,
2022). Here, all parties agree that the only limitation relevant to McCall’s motion is supplied by
§ 3582(c)(1)(A)’s phrase “extraordinary and compelling reasons.” The issue is whether that
phrase prohibits district courts from considering nonretroactive changes in law on a case-by-case
basis.2

2
Despite its suggestion that Concepcion is inapplicable here, the majority agrees that the only possible
limitation on the district courts’ ability to consider nonretroactive changes in law when deciding whether to grant
compassionate release under § 3582(c)(1)(A) comes from the phrase “extraordinary and compelling reasons.” See
Maj. Op. 18 n.6; cf. United States v. Jenkins, 50 F.4th 1185, 1200 (D.C. Cir. 2022) (asserting that Congress limited
district courts’ discretion to award compassionate release by authorizing compassionate release only for
“extraordinary and compelling reasons”).
 No. 21-3400 United States v. McCall Page 27

The majority’s detailed historical account of § 3582(c)(1)(A) and the First Step Act’s
recent amendments to the provision demonstrates the error in reading inflexible limitations into
the phrase “extraordinary and compelling reasons.” As the majority explains, § 3582(c)(1)(A)
originated with the passage of the Sentencing Reform Act of 1984, which abolished federal
parole and constrained federal courts from “modify[ing] a term of imprisonment once it has been
imposed.” Sentencing Reform Act of 1984, Pub. L. No. 98–473, Title II, ch. 2, § 212(a), 98 Stat.
1837, 1998 (enacting 18 U.S.C. § 3582(c)). Congress provided an exception to that general rule
in the form of compassionate release, allowing federal courts to reduce a sentence when
“warrant[ed]” by “extraordinary and compelling reasons[.]” 18 U.S.C. § 3582(c)(1)(A) (1984).

By design, § 3582(c)(1)(A) provides targeted relief to particular individuals whose
unique circumstances warrant a “second look at resentencing[.]” Shon Hopwood, Second Looks
& Second Chances, 41 CARDOZO L. REV. 83, 102 (2019); see also United States v. Ruffin, 978
F.3d 1000, 1005 (6th Cir. 2020) (noting the “discretionary nature of a reduction-of-sentence
decision” under § 3582(c)(1)(A)). Whereas the prior federal parole system tasked the Parole
Commission with “review[ing] every federal sentence” for evidence that the imprisoned person
had been rehabilitated before deciding whether to grant that person release from confinement,
§ 3582(c)(1)(A) allows “courts to decide, in individual cases, if ‘there is justification for
reducing a term of imprisonment.’” Hopwood, supra, at 102 (quoting S. Rep. No. 98-225, at 56
(1983)). As a result of this penological shift, “extraordinary and compelling reasons” replaced
rehabilitation as the lodestar for those empowered to decide whether to release certain
imprisoned individuals.

Congress has never provided, nor sought to provide, a comprehensive definition of
“extraordinary and compelling reasons” under § 3582(c)(1)(A). For over three decades,
Congress instead delegated the power to define the phrase to the Bureau of Prisons (“BOP”) and
the Sentencing Commission. The Sentencing Commission, for its part, was tasked with issuing
general policy statements “describ[ing] what should be considered extraordinary and compelling
reasons for sentence reduction, including the criteria to be applied and a list of specific
examples.” 28 U.S.C. § 994(t). Congress limited this expansive delegation in only one respect,
dictating that “[r]ehabilitation of the defendant alone shall not be considered an extraordinary
 No. 21-3400 United States v. McCall Page 28

and compelling reason” for a sentence reduction. Id. The ability to define “extraordinary and
compelling reasons” under § 3582(c)(1)(A) was otherwise left to the discretion of the Sentencing
Commission.

The Sentencing Commission has done little to advance our understanding of what is
“extraordinary and compelling” under § 3582(c)(1)(A). “Ignoring Congress’s edict for twenty-
two years, the Commission issued its first policy statement regarding compassionate release only
in 2006.” United States v. Jones, 980 F.3d 1098, 1104 (6th Cir. 2020) (citing U.S.S.G. § 1B1.13
(U.S. Sent’g Comm’n 2006)). That policy statement, as the majority observes, did little to
explicate what amounted to extraordinary and compelling reasons for a sentence reduction. See
Maj. Op. 6–7. Instead, it merely “parroted the ‘extraordinary and compelling reasons warrant [a
sentence] reduction’ language from § 3582(c)(1)(A).” Jones, 980 F.3d at 1104. The Sentencing
Commission later amended the commentary’s application notes in 2007 and 2018 to describe
certain circumstances under which “extraordinary and compelling reasons” exist. See United
States v. Brooker, 976 F.3d 228, 232–33 (2d Cir. 2020) (walking through the amendments). By
2018, the Commission’s definition of “extraordinary and compelling reasons” included “events
relating to the traditional categories of the imprisoned person’s health, age, or family
circumstances, and to clarify that such circumstances did not have to be unforeseen at the time of
sentencing.” Id. (citing U.S.S.G. § 1B1.13 nn.1, 2 (U.S. Sent’g Comm’n 2018)). Even then,
however, the Sentencing Commission did not define “criteria to be applied” when evaluating the
reasons, cf. 28 U.S.C. § 994(t), and left open the possibility that other, undefined reasons would
warrant a sentence reduction under § 3582(c)(1)(A), see Brooker, 976 F.3d at 233.3

The Sentencing Commission’s inaction left the BOP “to fill the void and create the
standards for extraordinary and compelling reasons that warrant resentencing.”
Hopwood, supra, at 103. In its original form, § 3582(c)(1)(A) allowed only the BOP’s Director
to file motions for compassionate release. Jones, 980 F.3d at 1104. The BOP rarely exercised
this power, however, and subsequent investigation revealed that the BOP did “not properly

3
The Sentencing Commission recently signaled its interest in revisiting its existing policy statement,
releasing a list of proposed policy priorities that included “[c]onsideration of possible amendments to §1B1.13 . . . to
. . . further describe what should be considered extraordinary and compelling reasons for sentence reductions under
18 U.S.C. § 3582(c)(1)(A).” Proposed Priorities for Amendment Cycle, 87 Fed. Reg. 60,438, 60,439 (Oct. 5, 2022).
 No. 21-3400 United States v. McCall Page 29

manage the compassionate release program,” its “implementation of the program [was]
inconsistent and result[ed] in ad hoc decision making,” and it “ha[d] no timeliness standards for
reviewing . . . requests.” U.S. DEP’T OF JUST. OFFICE OF THE INSPECTOR GENERAL, THE FEDERAL
BUREAU OF PRISONS’ COMPASSIONATE RELEASE PROGRAM 11 (2013),
https://www.oversight.gov/sites/default/files/oig-reports/e1306.pdf. As we have observed, the
BOP’s mismanagement of the compassionate-release process had a direct and consequential
impact on imprisoned individuals by preventing them from obtaining relief under
§ 3582(c)(1)(A). See United States v. Elias, 984 F.3d 516, 518–19 (6th Cir. 2021) (remarking
that BOP rarely exercised its power to bring compassionate release motions on behalf of
imprisoned individuals); Jones, 980 F.3d at 1104 (observing that BOP approved only 6% of
5,400 compassionate-release applications received between 2013 and 2017).

In 2018, Congress moved to increase the availability of compassionate release by
reforming § 3582(c)(1)(A) in the First Step Act. The Act altered the compassionate-release
landscape by shifting power from the BOP and Sentencing Commission to the district courts in
two ways. First, the Act “ousted the BOP from its preclusive gatekeeper position” by
empowering imprisoned individuals to file motions for compassionate release without the
support of the BOP under certain circumstances. Jones, 980 F.3d at 1105. Second, until the
Sentencing Commission issues guidance for defendant-filed motions, the Act freed the district
courts to define the phrase “extraordinary and compelling” without reference to the Sentencing
Commission’s guidance. See United States v. Sherwood, 986 F.3d 951, 953 (6th Cir. 2021). The
sum of these changes was to empower district courts to “define ‘extraordinary and compelling’
on their own initiative” when evaluating motions filed directly by incarcerated persons. Elias,
984 F.3d at 519–20.

Read against this historical backdrop, there is nothing in the First Step Act or
§ 3582(c)(1)(A) that precludes a district court from finding that a nonretroactive change in law is
an “extraordinary and compelling reason” for a sentence reduction under the appropriate
circumstances. As recounted above, Congress has imposed and later removed several limitations
on the district courts’ discretion to define the phrase “extraordinary and compelling reasons”
since § 3582(c)(1)(A)’s enactment. Today, Congress imposes only one operative textual
 No. 21-3400 United States v. McCall Page 30

limitation on the phrase: “[r]ehabilitation of the defendant alone” cannot constitute “an
extraordinary and compelling reason” for a sentence reduction. 28 U.S.C. § 994(t). Congress
has otherwise eschewed providing a rigid definition of the reasons that may be “extraordinary
and compelling” under § 3582(c)(1)(A), allowing the district courts to resolve in the first
instance whether a change in law, or any other relevant fact, constitutes an extraordinary and
compelling reason for a sentence reduction in the context of an imprisoned person’s unique
circumstances. Given this historical context, we should join the First, Fourth, Ninth, and Tenth
Circuits in declining to read additional limitations into § 3582(c)(1)(A). See United States v.
Ruvalcaba, 26 F.4th 14, 25–28 (1st Cir. 2022); United States v. McCoy, 981 F.3d 271, 286 (4th
Cir. 2020); Chen, 48 F.4th at 1098–1101; United States v. McGee, 992 F.3d 1035, 1047–48 (10th
Cir. 2021); see also Concepcion, 142 S. Ct. at 2402 (cautioning against “[d]rawing meaning from
silence” in the sentencing context (quoting Kimbrough v. United States, 552 U.S. 85, 103
(2007)).4

None of this is to say that nonretroactive changes in law will amount to “extraordinary
and compelling reasons” for a sentence reduction in all cases under § 3582(c)(1)(A). Indeed, the
role of a district court in resolving a motion for compassionate release filed under
§ 3582(c)(1)(A) is not to decide whether a change in law is itself extraordinary or compelling.
The district court must instead evaluate whether there are “extraordinary and compelling
reasons” to “reduce the sentence when a particular statutory change is considered in the context
of the defendant’s individualized circumstances.” Ruvalcaba, 26 F.4th at 32 (Barron, J.,
concurring). This individualized analysis will necessarily lead to different results in different
cases, just as it does in many cases arising under § 3582(c)(1)(A), as district courts work through
the particular facts raised by differently situated imprisoned individuals. Compare, e.g., United
States v. Lawrence, No. 17-20259, 2021 WL 859044, at *1 (E.D. Mich. Mar. 8, 2021) (finding
extraordinary and compelling reasons warranting release where the defendant was obese and

4
The majority’s decision puts us in conflict with these courts and further entrenches the circuit split on this
issue. See United States v. Andrews, 12 F.4th 255, 260–62 (3d Cir. 2021) (adopting the majority’s position); United
States v. Thacker, 4 F.4th 569, 573–75 (7th Cir. 2021) (same); United States v. Crandall, 25 F.4th 582, 585–86 (8th
Cir. 2022) (same); Jenkins, 50 F.4th at 1197–1200 (same). Unless and until the Supreme Court takes up the issue,
McCall and those like him who are within these jurisdictions will be prevented from having district courts consider
the totality of their circumstances when deciding whether to modify their sentences.
 No. 21-3400 United States v. McCall Page 31

demonstrated that his guideline range would have been significantly lower following Havis),
with United States v. Lewis, No. CR 0:14-362-JFA, 2021 WL 4596450, at *9 (D.S.C. Oct. 6,
2021) (finding that a nonretroactive change in law did not amount to extraordinary and
compelling reasons for a sentence reduction in part because the district court would have
imposed the same sentence irrespective of the change in law).

Turning to the present case, I would hold that the district court was required to consider
each of the “extraordinary and compelling reasons” identified in McCall’s motion, including the
impact of the COVID-19 pandemic, his rehabilitation efforts since sentencing, his reentry plan,
and the fact that he would face a shorter sentencing guideline range if he were sentenced after
our decision in Havis. Of course, the district court was not required to accept McCall’s
arguments or to order a modification of his sentence. See Concepcion, 142 S. Ct. at 2405. The
district court could have found, for instance, that McCall’s relatively brief motion for
compassionate release merely identified various facts and failed to demonstrate that these factors
were extraordinary and compelling in the context of his own circumstances. Or the district court
could have reasoned that McCall’s conditions of confinement, when considered alongside the
fact that he had already served a sentence longer than recommended under the current guidelines,
warranted taking a second look at his sentence to account for the developments that had occurred
since his original sentencing. For our purposes, it would be sufficient that the district court
consider these arguments and provide a reasoned explanation for its decision. Jones, 980 F.3d at
1112–16.

The majority, by contrast, holds that the district court was precluded as a matter of law
from even considering McCall’s arguments related to our decision in Havis. The majority’s
primary argument in support of its contrary decision rests on a textual analysis of the phrase
“extraordinary and compelling reasons.” Drawing on a dictionary definition from the time of
§3582(c)(1)(A)’s adoption, the majority observes that “extraordinary and compelling reasons”
are those that are “unusual, rare, and forceful.” Maj. Op. 9. In the majority’s view, the situation
in which McCall finds himself is neither extraordinary nor compelling. Nonretroactive changes
in law, the argument goes, are not “extraordinary” because “in federal sentencing the ordinary
practice is to apply new penalties to defendants not yet sentenced, while withholding that change
 No. 21-3400 United States v. McCall Page 32

from defendants already sentenced.” Dorsey v. United States, 567 U.S. 260, 280 (2012). And
the fact that McCall is serving a sentence imposed based on a prior interpretation of the law is
not “compelling” because the sentence was lawful at the time it was imposed. The majority’s
argument, if appealing at a superficial level, falls apart in the compassionate-release context.

As a preliminary matter, the majority’s free-floating textual analysis demonstrates the
peril of divorcing statutes from their context. Compassionate-release decisions do not exist in a
vacuum. When an imprisoned person moves for compassionate release based on their advanced
age or “deteriorating physical or mental health because of the aging process,” U.S.S.G. § 1B1.13
n.1(A)(ii)(III), (B), § 3582(c)(1)(A) does not direct the district court to decide whether aging or
its associated challenges are themselves unusual or rare. No doubt they are not, especially given
today’s aging prison population. See Daniel Litwok et al., Causes of Aging in the Federal Prison
Population and a Comparison to States, 45 Crim. Just. Rev. 129, 129 (2020) (observing that “the
population of federal inmates over the age of 50 continues to be the fastest growing age segment
among prison populations”). Rather, the district court must evaluate whether the person’s age or
health challenges, considered in the context of that person’s particular circumstances, present
extraordinary and compelling reasons to reevaluate the person’s sentence. The same is true
when an imprisoned person argues that a nonretroactive change in law supports their
compassionate-release motion. By ignoring the context in which compassionate-release
decisions are made, the majority walls itself off from the context that should guide our
interpretation of Congress’s work and substitutes its own inflexible judgment for the reasoned
discretion of the district courts.

There is an even more fundamental problem with the majority’s textual analysis: it
writes into § 3582(c)(1)(A) what Congress declined to include. “Congress has only placed two
limitations directly on extraordinary and compelling reasons: the requirement that district courts
are bound by the Sentencing Commission's policy statement, which does not apply here, and the
requirement that ‘[r]ehabilitation . . . alone’ is not extraordinary and compelling.” Chen, 48
F.4th at 1098. The majority purports to read § 3582(c)(1)(A) to impose a third limitation:
nonretroactive changes in law, whether alone or in combination with other factors, cannot be
extraordinary and compelling reasons for compassionate release. The problem, of course, is that
 No. 21-3400 United States v. McCall Page 33

Congress has had more than three decades to impose such a limitation, but never has. Instead,
Congress adopted a broad and ambiguous phrase and left it to different institutions—first the
Sentencing Commission and BOP, and now the district courts—to sort out what the phrase
means in practice. Our circuit has long declined to add restrictions to Congress’s chosen text.
See, e.g., Gen. Med., P.C. v. Azar, 963 F.3d 516, 521 (6th Cir. 2020) (stating that courts “should
not add language that Congress has not included”). Today, the majority does exactly that.

To support its reading of § 3582(c)(1)(A), the majority also asserts several arguments
under the banner of what it calls the structure of federal sentencing law. According to the
majority, allowing district courts to find that nonretroactive changes in law constitute
extraordinary and compelling reasons under § 3582(c)(1)(A) would subvert Congress’s choice
not to make those changes retroactive. See Maj. Op. 11–12. As the Fourth Circuit has aptly
explained, however, “there is a significant difference between automatic vacatur and
resentencing of an entire class of sentences—with its avalanche of applications and inevitable
resentencings—and allowing for the provision of individual relief in the most grievous cases.”
McCoy, 981 F.3d at 286–87 (internal quotation marks and citation omitted); see also Chen, 48
F.4th at 1100 (same). In other words, the ability of some incarcerated individuals to access
discretionary relief in no way subverts Congress’s broader decision not to confer a broader,
automatic entitlement to all such individuals.

The majority’s second structural argument is even more problematic. For several pages,
the majority labors to explain why McCall would not be able to obtain the benefit of our Havis
decision through collateral proceedings under 28 U.S.C. § 2255. See Maj. Op. 12–14. Of
course, the issue of whether McCall is entitled to post-conviction relief is not before us. And
more to the point, McCall’s entitlement to relief under § 2255 has no bearing on whether he has
shown that extraordinary and compelling reasons warrant a sentence reduction under
§ 3582(c)(1)(A). A comparison between the two statutes illustrates the flaw in the majority’s
analysis. Although the majority groups collateral attacks and compassionate release under the
phrase “post-conviction relief,” Maj. Op. 13, the two serve entirely different purposes and are
governed by entirely different procedural and substantive rules. When a court grants a habeas
petition, it deems the sentence invalid. See Daniels v. United States, 532 U.S. 374, 379 (2001).
 No. 21-3400 United States v. McCall Page 34

By contrast, a grant of compassionate release does not invalidate the relevant sentence; rather, it
recognizes that a holistic consideration of the defendant’s circumstances entitles them to early
release—a remedy that Congress specifically codified in § 3582(c)(1). Ruvalcaba, 26 F.4th at
27; McCoy, 981 F.3d at 287; United States v. Maumau, 993 F.3d 821, 837 (10th Cir. 2021);
Appellant Supp. Br. at 17–18.

The majority attempts to bridge the gap between the two forms of relief in two ways.
First, the majority posits that Congress knew of the Supreme Court’s limitations on the
retroactive application of new rules of law in post-conviction proceedings but did not clearly
abrogate them in § 3582(c)(1)(A). Maj. Op. 13. According to the majority, if Congress intended
for § 3582(c)(1)(A) to operate outside the bounds of the retroactivity rules applicable in
collateral-review proceedings, it would have made that intention clear. But there is simply no
question that motions filed under § 3582(c)(1)(A) are not subject to the retroactivity rules
applicable in collateral proceedings. That is clear from Concepcion, 142 S. Ct. 2389, a decision
that does not once address the limitations the majority suggests are applicable here. In fact, prior
to the majority’s opinion and the decisions it relies upon for this point, Congress would have had
no reason to think that it needed to distance § 3582(c)(1)(A) from the rules applicable in post-
conviction proceedings. Thus, Congress’s silence is meaningful, but not for the reasons
identified by the majority.5

Second, the majority justifies its reliance on post-conviction rules by asserting that
permitting district courts to find that a nonretroactive change in law constitutes an “extraordinary
and compelling reason” for release under § 3582(c)(1)(A) would necessarily cast doubt on the
validity of the imprisoned person’s sentence, an issue that the Supreme Court has said must be
raised through “habeas corpus (or similar state) remedies.” Wilkinson v. Dotson, 544 U.S. 74, 81
(2005). There are multiple shortcomings in this argument. To begin with, the majority once

5
The majority’s overreading of Congress’s inaction shows why “[d]rawing meaning from silence is
particularly inappropriate’ in the sentencing context, ‘for Congress has shown that it knows how to direct sentencing
practices in express terms.’” Concepcion, 142 S. Ct. at 2402 (quoting Kimbrough, 552 U.S. at 103). The majority’s
position is that Congress’s purported silence on this issue amounts to an implicit adoption of a set of strict
procedural rules developed in an entirely different area of law. Congress may have the power to make such a
dramatic change to the compassionate-release process, but nothing in § 3582(c)(1)(A)’s history or Congress’s
approach to sentencing issues suggests that it would do so through mere silence.
 No. 21-3400 United States v. McCall Page 35

again misapprehends the significance of compassionate release. A finding that extraordinary and
compelling reasons exist in part due to a nonretroactive change in law does not—necessarily or
otherwise—invalidate an imprisoned person’s conviction. Rather, as in the case where the
person’s age or health are found to be extraordinary and compelling, the finding simply reflects
that an intervening event provides grounds to reconsider the length of the person’s sentence and
allows the district court to proceed to consider whether a sentence modification is in fact
warranted.

Further, the majority’s argument on this score is fatally undercut by a point it makes just
a few paragraphs later. The majority concedes, as it must, that district courts may consider
nonretroactive legal developments after finding that extraordinary and compelling reasons exist.
See Maj. Op. 14; see also Concepcion, 142 S. Ct. at 2402 n.6. The majority does not—and
cannot—explain why allowing district courts to consider nonretroactive changes in law at the
“extraordinary and compelling” stage would circumvent the post-conviction process, but
allowing them to consider those same legal developments when weighing the 18 U.S.C.
§ 3553(a) factors would not. The majority’s inability to reconcile this fundamental conflict in its
reasoning demonstrates the error in basing a decision on post-conviction rules that have no
bearing here.

Lastly, the majority offers an historical argument that is at odds with the history
recounted earlier in its decision. Based on two-page historical survey of a prior version of the
compassionate-release statute, a decision by a federal district court in Michigan from over four
decades ago, and the action (and inaction) of the BOP and Sentencing Commission discussed
earlier, the majority declares that compassionate release has never been used to reduce sentences
based on nonretroactive changes in law. Maj. Op. 15–17. To be sure, the majority and I agree
that the resolution of the issue before us should be informed by historical context and practice.
But the value of the historical record in this case is derived from its power to explain Congress’s
motivation in passing the First Step Act, not in defining the universe of extraordinary and
compelling reasons warranting compassionate release under § 3582(c)(1)(A). The majority turns
 No. 21-3400 United States v. McCall Page 36

history on its head by suggesting that Congress shackled the district courts to an understanding
of compassionate release that it explicitly and emphatically repudiated in the First Step Act.6

In the end, nothing in § 3582(c)(1)(A)’s history, text, or purpose suggests that Congress
imposed an unwritten yet unyielding limitation on the district courts’ ability to find that
nonretroactive changes in law constitute extraordinary and compelling reasons for a sentence
reduction under the appropriate circumstances. I therefore decline to join the majority in
rewriting § 3582(c)(1)(A) in a way that overrides Congress’s efforts to expand access to
compassionate release, limits the ability of imprisoned individuals to present the true account of
their individual circumstances at resentencing, and breaks with the “longstanding tradition in
American law” that “a judge at sentencing considers the whole person before him or her as an
individual.” Concepcion, 142 S. Ct. at 2395 (internal quotation marks omitted). I respectfully
dissent.

6
The majority defends its historical analysis in a footnote asserting that Congress intended for the First Step
Act to alter only the procedural, not the substantive, aspects of the compassionate-release process. Maj. Op. 17 n.5.
But that point is impossible to reconcile with observations made elsewhere in the majority’s opinion. Indeed, there
is no dispute that Congress passed the First Step Act in part to increase the use of compassionate release. See id. at 7
(citing a section of the First Step Act entitled “Increasing the Use and Transparency of Compassionate Release”);
see also Jones, 980 F.3d at 1104–05. Nor is there any dispute that Congress sought to achieve that substantive end
through procedural changes to § 3582(c)(1)(A). See Maj. Op. 8 (describing the “two ways” in which the First Step
Act changed compassionate release). Lastly, “[e]veryone agrees” that the practical effect of Congress’s procedural
changes was to “enhance[] district courts’ discretion to grant compassionate release.” Id. The majority’s
inconsistency on this point is emblematic of its broader failure to appreciate how Congress has determined the
availability of compassionate release not by circumscribing the phrase “extraordinary and compelling reasons” but
by vesting discretion to define the phrase in different institutions over time.
 No. 21-3400 United States v. McCall Page 37

_________________

DISSENT
_________________

JULIA SMITH GIBBONS, Circuit Judge, dissenting. At the time we heard argument in
this case, my best assessment of the law was that nonretroactive changes in sentencing law could
not constitute “extraordinary or compelling reasons” for granting compassionate release. See
United States v. Hunter, 12 F.4th 555, 563 (6th Cir. 2021), cert. denied, 142 S. Ct. 2771 (2022).
In my judgment, the broad language of the Supreme Court’s recent opinion in United States v.
Concepcion, 142 S. Ct. 2389 (2022), makes adherence to that position impossible. Because the
majority opinion fails to fairly apply the scope and spirit of Concepcion, I respectfully dissent.

Although Concepcion deals with § 404(b) of the First Step Act, it very specifically
undertakes in Part II a broad and expansive view of sentencing discretion. 142 S. Ct. at 2398-
401. Concepcion warns that the “only limitations on a court’s discretion to consider any relevant
materials at an initial sentencing or in modifying that sentence are those set forth by Congress in
a statute or by the Constitution.” Id. at 2400. That “venerable tradition of discretion” applies in
full to both “initial sentencings and sentence modification proceedings.” Id. at 2401 n.4.
Following Concepcion’s instruction, we should be reluctant to manufacture our own limits on a
district court’s discretion in determining “extraordinary or compelling reasons” for
compassionate release without a clear statement from Congress.

The majority reads Concepcion more narrowly. It claims that Concepcion’s broad
statement of sentencing powers does not apply to the threshold question of whether a prisoner
has established an extraordinary and compelling reason for compassionate release. Maj. Op. at
22. In the majority’s interpretation, this broad discretion only arises in the resentencing itself,
not the threshold decision of whether “extraordinary and compelling” circumstances exist. But
such framing contradicts the language of Concepcion, which incorporates consideration of
nonretroactive changes at the threshold stage:

In many cases, a district court is prohibited from recalculating a Guidelines range
in light of nonretroactive Guidelines amendments, but the [district] court may find
 No. 21-3400 United States v. McCall Page 38

those amendments to be germane when deciding whether to modify a sentence at
all, and if so, to what extent.

Concepcion, 142 S. Ct. at 2400 (emphasis added). If a district court finds a nonretroactive
change in law “germane” to its decision to resentence the defendant, then that change is a factor
that the district court can properly consider when determining whether “extraordinary and
compelling reasons” exist for compassionate release.1

These inherent sentencing powers are the backdrop against which Congress must carve
out exceptions. However, “Congress has created only two relevant limitations that control the
district court’s reading of the [“extraordinary and compelling”] standard: One for controlling
Sentencing Commission policy statements, and the other for ‘rehabilitation . . . alone.’” United
States v. Jenkins, 50 F.4th 1185, 1210 (D.C. Cir. 2022) (Ginsburg, J., concurring in part,
dissenting in part). The majority here cites no further congressional limitation against
nonretroactive changes in sentencing law. Instead, it relies on silence. See, e.g., Maj. Op. at 19
(explaining that the Sentencing Commission’s policy statement “never included nonretroactive
legal developments”); id. (inferring that “Congress, no doubt, had this practice [of only referring
to unforeseeable circumstances for compassionate release] in mind when it amended . . . the
statute in 2018”). Silence is insufficient. “Nothing in the text and structure of the First Step Act
expressly, or even implicitly, overcomes the established tradition of district courts’ sentencing
discretion.” Concepcion, 142 S. Ct. at 2401. The majority would have us use every possible tool
of interpretation to fashion a high bar for what constitutes “extraordinary and compelling
reasons” for compassionate release. However, such manipulation effectively circumvents the
Supreme Court’s directive of broad sentencing discretion.

At each turn, the majority is contradicted by Concepcion. It emphasizes background
principles like finality, see Maj. Op. at 11, but “[n]o one doubts the importance of finality. Here,
however, the Court interprets a statute whose very purpose is to reopen final judgments.”

1
The majority suggests that my reasoning takes this sentence out of context in the Concepcion opinion.
Maj. Op. at 23 n.7. Yet, Concepcion clearly makes a broad and undeniable statement about resentencing discretion
which the opinion does not limit to one part of the First Step Act. Concepcion’s background presumption of
discretion, in which the Court expressly includes retroactive changes in sentencing law, is not overcome by
speculating about how Congress would fill in the definition of “extraordinary and compelling reasons.”
 No. 21-3400 United States v. McCall Page 39

Concepcion, 142 S. Ct. at 2398 n.3. It advises looking to the structure and history of federal
sentencing law, see Maj. Op. at 13-20, but then fails to give weight to the broad historical
sentencing power of the judiciary which “dates back to before the founding.” Concepcion,
142 S. Ct. at 2399. And it draws interpretive meaning from congressional silence, despite a
warning against doing so “in the sentencing context, for Congress has shown that it knows how
to direct sentencing practices in express terms.” Concepcion, 142 S. Ct. at 2402 (internal
quotations and citation omitted).

When the Supreme Court speaks this broadly, we cannot ignore it. Because I see no way
to both adhere to Concepcion and join my colleagues in the majority, I respectfully dissent.