Amicus Curiae Brief — William Maxwell, Petitioner v. Sandra Dinis, Acting Warden
Supreme Court briefAug 12, 2026
Ask Donna
What actually matters in this document.
Text
No. 25-5930
IN THE
Supreme Court
of the United States
_________
WILLIAM MAXWELL,
Petitioner,
v.
SANDRA DINIS, ACTING WARDEN,
_________
Respondent.
On Writ of Certiorari to the
United States Court of Appeals
for the Fifth Circuit
_________
BRIEF OF SENATORS RICHARD J. DURBIN
AND CORY A. BOOKER AS AMICI CURIAE IN
SUPPORT OF PETITIONER
_________
DELIA SCOVILLE
HOGAN LOVELLS
CADWALADER US LLP
125 High Street
Suite 2010
Boston, MA 02110
(617) 371-1000
JO-ANN TAMILA SAGAR
REEDY C. SWANSON
Counsel of Record
MARLAN GOLDEN
HOGAN LOVELLS
CADWALADER US LLP
555 Thirteenth Street, N.W.
Washington, DC 20004
(202) 637-5600
reedy.swanson@hlc.com
Counsel for Amici Curiae
i
TABLE OF CONTENTS
Page
TABLE OF AUTHORITIES ....................................... ii
STATEMENT OF INTEREST ................................... 1
INTRODUCTION AND SUMMARY OF
ARGUMENT ............................................................ 2
ARGUMENT ............................................................... 4
I.
II.
CONGRESS DESIGNED AND
ENACTED THE FIRST STEP ACT TO
EXPAND ACCESS TO PRELEASE
CUSTODY ......................................................... 4
A.
Congress Designed The First Step
Act To Enact Comprehensive,
Bipartisan Criminal Justice Reform........4
B.
Congress Made An Earned Time
Framework Incentivizing Transfers
To Prerelease Custody A
Cornerstone Of The First Step Act ..........5
CONGRESS INTENDED FOR THE
FIRST STEP ACT’S INCENTIVE
SYSTEM TO BE ENFORCEABLE AND
REVIEWABLE ............................................... 11
A.
The First Step Act Imposes A
Mandatory Obligation On The
Bureau Of Prisons To Apply
Earned Time Credits To Prerelease
Custody Or Supervised Release ............. 11
B.
The First Step Act’s Prerelease
Custody Incentives Are Enforceable
Through Habeas Petitions ..................... 14
CONCLUSION .......................................................... 19
ii
TABLE OF AUTHORITIES
Page(s)
CASES:
Bivens v. Six Unknown Named Agents of
Federal Bureau of Narcotics,
403 U.S. 388 (1971)................................................ 18
Bufkin v. Collins,
604 U.S. 369 (2025)................................................ 12
Davis v. Mich. Dep’t of Treasury,
489 U.S. 803 (1989)................................................ 18
Duncan v. Walker,
533 U.S. 167 (2001)................................................ 12
Edwards v. Balisok,
520 U.S. 641 (1997)................................................ 15
Fogerty v. Fantasy, Inc.,
510 U.S. 517 (1994)................................................ 11
Gagnon v. Scarpelli,
411 U.S. 778 (1973)................................................ 17
González-Fuentes v. Molina,
607 F.3d 864 (1st Cir. 2010) .................................. 16
Griffin v. Oceanic Contractors, Inc.,
458 U.S. 564 (1982)................................................ 19
Hewitt v. United States,
606 U.S. 419 (2025)............................................ 4, 15
Johnson v. Pettiford,
442 F.3d 917 (5th Cir. 2006) ................................. 17
Levine v. Apker,
455 F.3d 71 (2d Cir. 2006) ..................................... 16
Me. Cmty. Health Options v. United States,
590 U.S. 296 (2020).......................................... 12, 13
iii
TABLE OF AUTHORITIES—Continued
Page(s)
Merck & Co. v. Reynolds,
559 U.S. 633 (2010)................................................ 14
Morrissey v. Brewer,
408 U.S. 471 (1972)................................................ 17
Mujahid v. Daniels,
413 F.3d 991 (9th Cir. 2005) ................................. 17
Murphy v. Smith,
583 U.S. 220 (2018).......................................... 12, 13
Osorio-Calderon v. Warden, FCI Sandstone,
179 F.4th 1158 (8th Cir. 2026) .............................. 12
Preiser v. Rodriguez,
411 U.S. 475 (1973)................................................ 15
Russello v. United States,
464 U.S. 16 (1983).................................................. 13
United States v. Hansen,
599 U.S. 762 (2023).................................................. 6
United States v. Haymond,
588 U.S. 634 (2019)................................................ 17
Wilkinson v. Dotson,
544 U.S. 74 (2005)............................................ 15, 16
Woodall v. Federal Bureau of Prisons,
432 F.3d 235 (3d Cir. 2005) ............................. 16, 17
STATUTES:
18 U.S.C. § 3621 ....................................................... 12
18 U.S.C. § 3621(b) ................................................... 12
18 U.S.C. § 3622 ....................................................... 12
18 U.S.C. § 3624(b)(1) ................................................ 8
18 U.S.C. § 3624(c)(1) ................................................. 2
iv
TABLE OF AUTHORITIES—Continued
Page(s)
18 U.S.C. § 3624(g)(1)(D)(i)-(ii) .................................. 5
18 U.S.C. § 3624(g)(4)............................................... 13
18 U.S.C. § 3632 ....................................................... 18
18 U.S.C. § 3632(a)(6) ................................................ 7
18 U.S.C. § 3632(c) ................................................... 13
18 U.S.C. § 3632(d) ................................................... 11
18 U.S.C. § 3632(d)(4)(A)(i) .................................... 7, 8
18 U.S.C. § 3632(d)(4)(A)(i) ........................................ 8
18 U.S.C. § 3632(d)(4)(B) ......................................... 13
18 U.S.C. § 3632(d)(4)(C) ....................................... 7, 9
18 U.S.C. § 3632(d)(4)(D) ......................................... 13
18 U.S.C. § 3632(d)(4)(E) ......................................... 13
31 U.S.C. § 3729(b)(1)(A)(i)-(iii) ...................................
REGULATIONS:
Annual Determination of Average Cost of
Incarceration Fee (COIF), 90 Fed. Reg.
58,059 (Dec. 15, 2025) ........................................... 10
Home Confinement Under the Coronavirus
Aid, Relief, and Economic Security (CARES)
Act, 88 Fed. Reg. 19,830 (Apr. 4, 2023)................. 10
LEGISLATIVE MATERIALS:
First Step Act of 2018, Pub. L. No. 115-391,
132 Stat. 5194 .......................................................... 2
164 Cong. Rec. S7639 (daily ed. Dec. 17,
2018):
S7642 (statement of Sen. Cornyn) .......................... 7
S7644 (statement of Sen. Durbin)........................... 7
v
TABLE OF AUTHORITIES—Continued
Page(s)
S7645 (statement of Sen. Durbin)........................... 4
S7646 (statement of Sen. Durbin)........................... 5
S7649 (statement of Sen. Grassley) ........................ 5
S7650 (vote) ............................................................. 5
164 Cong. Rec. S7740 (daily ed. Dec. 18,
2018):
S7744 (statement of Sen. Durbin)....................... 6, 8
S7745 (statement of Sen. Blumenthal) ................... 6
S7746 (statement of Sen. Cornyn) ...................... 7, 8
S7749 (statement of Sen. Leahy) ........................ 4, 7
S7778 (statement of Sen. Grassley) ........................ 4
S7781 (vote) ............................................................. 5
164 Cong. Rec. S7838 (daily ed. Dec. 19,
2018) (statement of Sen. Grassley) ....................... 11
164 Cong. Rec. H10362, H10430 (daily ed.
Dec. 20, 2018):
H10362 (statement of Rep. Collins) ........................ 6
H10430 (vote) ........................................................... 5
ADMINISTRATIVE MATERIALS:
Remarks by President Trump at Signing
Ceremony for S. 756, the “First Step Act of
2018” and H.R. 6964, the “Juvenile Justice
Reform Act of 2018,” 2018 WL 6715859
(Dec. 21, 2018) ......................................................... 5
OTHER AUTHORITIES:
Avinash Bhati, First Step Act: An Early
Analysis of Recidivism, Council on Crim.
Just. (Aug. 2023), https://perma.cc/C22NHV8C ........................................................................ 9
vi
TABLE OF AUTHORITIES—Continued
Page(s)
Avinash Bhati, First Step Act: An Early
Analysis of Recidivism, Council on Crim.
Just. (Dec. 2024), https://perma.cc/PL4ET4B4 ................................................................... 9, 10
Avinash Bhati, Time Sentenced and Time
Served: Exploring the Impact of the First
Step Act, Council on Crim. Just. (Dec.
2023), available at FSA.counciloncj.org/fsa .......... 10
Dep’t of Just., First Step Act Annual Report
(Apr. 2023), https://perma.cc/2RJF-7RUH ............. 9
Dep’t of Just., First Step Act Annual Report
(June 2024), https://perma.cc/9N8Z-JBG6 ............. 9
Press Release, Federal Bureau of Prisons,
Federal Bureau of Prisons Issues Directive
to Expand Home Confinement, Advance
First Step Act (May 28, 2025) ............................... 14
IN THE
Supreme Court of the United States
_________
No.
25-5930
_________
WILLIAM MAXWELL,
Petitioner,
v.
SANDRA DINIS, ACTING WARDEN,
Respondent.
_________
On Writ of Certiorari to the
United States Court of Appeals
for the Fifth Circuit
_________
BRIEF OF SENATORS RICHARD J. DURBIN
AND CORY A. BOOKER AS AMICI CURIAE IN
SUPPORT OF PETITIONER
_________
STATEMENT OF INTEREST1
Amici curiae Senators Richard J. Durbin and Cory
A. Booker are Members of the United States Senate
who were lead sponsors of the First Step Act of 2018.
The Act includes provisions that reformed the system
for earned time credits so that an offender who has
taken advantage of the statutory incentives to participate in evidence-based recidivism reduction programming is entitled to transfer to a less restrictive
form of custody, including community-based options
such as a halfway house or home confinement.
1
No counsel for a party authored this brief in whole or in part.
Only amici and counsel for amici funded its preparation and submission.
2
Passage of the First Step Act was a historic step forward for our Nation’s criminal justice system, which
was made possible with broad across-the-aisle support.
Amici are uniquely positioned to speak to the history
of the First Step Act and have a strong interest in ensuring that the Act is interpreted in a manner consistent with Congress’s chosen language. As reflected
in the text of the statute and overall statutory scheme,
offenders who participate in recidivism reduction programming may accrue credits for time earned. They
can then benefit from the Act’s incentive structure for
accelerating transfer to supervised release or prerelease custody, defined as custody served outside
prison under “conditions that will afford that prisoner
a reasonable opportunity to adjust to and prepare for
the reentry,” including home confinement and community correctional facilities. 18 U.S.C. § 3624(c)(1).
Amici urge the Court to hold that claims seeking
transfer to prerelease custody are cognizable in habeas, which would give full effect to the First Step
Act’s reforms in line with the Act’s text, purpose, and
underlying Congressional intent. The Fifth Circuit’s
contrary judgment should be reversed.
INTRODUCTION AND
SUMMARY OF ARGUMENT
In December 2018, a bipartisan majority of both
chambers of Congress passed, and President Donald
J. Trump signed into law, the First Step Act of 2018,
Pub. L. No. 115-391, 132 Stat. 5194 (the “First Step
Act” or the “Act”). The Act is a historic and bipartisan
piece of legislation, ushering in momentous correctional, sentencing, and criminal justice reforms.
3
At the heart of the Act is a carefully designed system
that incentivizes prisoner participation in rehabilitative and recidivism reduction activities. By participating in this productive, socially beneficial programming, offenders accumulate earned time credits that,
under the statute, must be applied to accelerate offenders’ access to community-based, alternative
means of custody such as halfway houses and home
confinement. After eight years, the myriad benefits of
this incentive system—to offenders and to the Nation—are evident: Participants in First Step Act programming reoffend at markedly lower rates, and offenders are able to more easily transition back into
productive, community-facing roles upon release.
This system also results in meaningful cost savings
for the federal government because the communitybased forms of confinement encouraged under the Act
are less expensive than full custody.
The Fifth Circuit’s judgment in this case threatens
to undermine Congress’s groundbreaking approach to
prison reform. First, the decision fails to consider that
the Act imposes a mandatory obligation on the Bureau
of Prisons to apply earned time credits to prerelease
custody or supervised release and to place offenders
into community-based alternative confinement arrangements as soon as the statutory criteria are met.
And second, the decision erects an artificial barrier to
relief for offenders like Mr. Maxwell, who seek to enforce the Bureau of Prisons’ nondiscretionary obligation. Nothing in the statutory text, congressional intent, or this Court’s precedent supports that result.
4
ARGUMENT
I. CONGRESS DESIGNED AND ENACTED
THE FIRST STEP ACT TO EXPAND ACCESS
TO PRELEASE CUSTODY.
Congress Designed The First Step Act To
Enact
Comprehensive,
Bipartisan
Criminal Justice Reform.
The First Step Act marked a watershed legislative
response to the extensive body of evidence showing
that rehabilitative programming is central to improving outcomes for people in federal custody. Enacted
by “a supermajority of Congress,” the Act is “a landmark piece of legislation” marking the culmination of
a yearslong bipartisan effort. Hewitt v. United States,
606 U.S. 419, 424 (2025). The supporters who formed
the “extraordinary political coalition” backing the
First Step Act reflected the broad constituency in favor of correctional sentencing and criminal justice reform. 164 Cong. Rec. S7639, S7645 (daily ed. Dec. 17,
2018) (statement of Sen. Durbin).
Support for the First Step Act was “not just bipartisan; it [was] nearly nonpartisan.” 164 Cong. Rec.
S7740, S7749 (daily ed. Dec. 18, 2018) (statement of
Sen. Leahy). Senate Judiciary Committee Chairman
Grassley echoed this sentiment, observing that he did
not know “whether we have had legislation like this
before * * * whereby we have put together such diverse groups of people and organizations that support
the bill.” Id. at S7778 (statement of Sen. Grassley).
The broad collection of supporters counted among its
ranks are groups ranging from the Fraternal Order of
Police to the American Civil Liberties Union, as well
5
as others like the American Correctional Association
and the Association of Prosecuting Attorneys. The
fruit of this coalition’s work was “the most significant
criminal justice reform bill in a generation.” 164
Cong. Rec. S7649 (statement of Sen. Grassley); see
also id. at S7646 (statement of Sen. Durbin) (describing the Act as “one of the most historic changes in
criminal justice legislation in our history”).
Considering the significant bipartisan coalition supporting the First Step Act, it is no surprise that it was
enacted by overwhelming bipartisan margins. The
bill was filed in the Senate in nearly its final form on
December 13, 2018. After invoking cloture on December 17 by a vote of 82 to 12, see 164 Cong. Rec. S7650,
the Senate passed the First Step Act the following day
by a vote of 87 to 12, see id. at S7781. Two days later,
the House passed the measure by a vote of 358 to 36.
See 164 Cong. Rec. H10430, H10430 (daily ed. Dec. 20,
2018). Then, on December 21, President Trump
signed the First Step Act into law, describing it as “an
incredible moment” for “criminal justice reform.” Remarks by President Trump at Signing Ceremony for
S. 756, the “First Step Act of 2018” and H.R. 6964, the
“Juvenile Justice Reform Act of 2018,” 2018 WL
6715859, at *16 (Dec. 21, 2018).
Congress
Made
An
Earned
Time
Framework Incentivizing Transfers To
Prerelease Custody A Cornerstone Of The
First Step Act.
Part of the First Step Act’s broader criminal justice
reforms were provisions that deliberately expanded
access to supervised release and other forms of prerelease custody. 18 U.S.C. § 3624(g)(1)(D)(i)-(ii). When
6
enacting the First Step Act, Congress acted against
the backdrop of sentencing laws and federal correctional practices that made it impossible to implement
the sort of evidence-based reforms that Congress embraced in the legislation. See 164 Cong. Rec. S7745
(statement of Sen. Blumenthal explaining that the
pre-enactment regime “provide[d] few incentives for
prisoners to prepare for reentry, and that is the gap
the First Step Act seeks to address”).2
When Congress surveyed the federal criminal justice landscape in 2018, it found staggering correctional expenses that showed no signs of slowing. Fiscal prudence accordingly figured into the legislative
calculus. Congress “want[ed] to focus tax dollars on
programs that have proven results” and applied that
budgetary restraint principle to the First Step Act,
which “spell[s] out exactly what [Congress was] looking for: the most effective and efficient, evidencebased recidivism-reduction programs.” 164 Cong.
Rec. S7744 (statement of Sen. Durbin). Congress legislated to “make our communities safer, save taxpayer
money, and start addressing the revolving door of incarceration and crime.” See 164 Cong. Rec. H10362
(Dec. 20, 2018) (statement of Rep. Collins). These reforms were implemented when the “cost of housing
Federal offenders” had grown so much that it “consume[d] nearly one-third of the Justice Department’s
budget”: Congress understood that “[b]ecause public
safety dollars are finite,” this extraordinary budgetary
burden “strips critical resources away from law
2
The Court looks to statutory history and context for guidance
in interpreting a statute’s terms, see United States v. Hansen,
599 U.S. 762, 775 (2023).
7
enforcement strategies that have been proven to make
our communities safer,” thus imposing both financial
and human costs. 164 Cong. Rec. S7749 (Statement
of Sen. Leahy).
Senator Cornyn explained how the First Step Act
was one effort to break the “perpetua[l] cycle of crime
that continues to plague communities across the country and to drain taxpayer dollars in the process.” See
164 Cong. Rec. S7642 (Dec. 17, 2018) (statement of
Sen. Cornyn); see also id. at S7644 (Dec. 17, 2018)
(statement of Sen. Durbin explaining that the First
Step Act set out to reduce the Federal prison population).
Prerelease custody was a key component of Congress’s design in enacting the First Step Act. Critically, Congress tied access to prerelease custody to an
incentive structure that is built into the statute: The
drafters made participation in recidivism reduction
programs the medium through which a person in custody may accrue earned time credit. See 18 U.S.C.
§ 3632(a)(6) (directing the Bureau of Prisons to “develop” a “risk and needs assessment system” that
“provide[s] incentives and rewards for successful participation in evidence-based recidivism reduction programs or productive activities”). Credits earned for
“completing evidence-based programming”—including “education, job training, drug treatment, life
skills, [and] faith-based programs”—help to make a
person eligible for prerelease custody. 164 Cong. Rec.
S7746 (statement of Sen. Cornyn); see also 18 U.S.C.
§ 3632(d)(4)(A)(i), (d)(4)(C).
Congress understood that the availability of incentives was paramount to the success of the First Step
8
Act’s new paradigm. As one Senator explained, because “we, as human beings, all operate based on incentives,” “the incentives in this program are really
important.” 164 Cong. Rec. S7746 (statement of Sen.
Cornyn). Congress thus deliberately created “incentive[s] for prisoners * * * to go through the program”—
namely, the promise that those who “gain the earned
credit * * * can be released” to community-based
placements that operate under “less confining conditions, for example, a halfway house.” Id. The statute’s
drafters understood that recidivism reduction programming would give people a direct stake in their
own correctional futures. Successful participation
would “reduce the time they serve” by allowing offenders to “be released to a halfway house or something
similar to that.” 164 Cong. Rec. S7744 (statement of
Sen. Durbin). Consistent with that end, Congress provided that federal prisoners may earn up to 54 days of
good time credit per year toward service of the overall
imposed sentence. 18 U.S.C. § 3624(b)(1).
The authors of the First Step Act paired this inducement with a set of mandatory directives to the Bureau
of Prisons. Congress directed that a “prisoner shall
earn 10 days of time credits for every 30 days of successful participation in evidence-based recidivism reduction programming or productive activities.” 18
U.S.C. § 3632(d)(4)(A)(i) (emphasis added). A prisoner categorized as medium- or low-risk is eligible to
earn additional time credits for participation. Id.
§ 3632(d)(4)(A)(ii). These earned time credits “shall
be applied toward time in prerelease custody or supervised release,” and the Bureau “shall transfer eligible
prisoners * * * into prerelease custody or supervised
9
release” when they meet the eligibility requirements
set out in the statute. Id. § 3632(d)(4)(C) (emphases
added).
In the years since the First Step Act’s enactment,
evidence shows the incentive system that Congress
designed, including prerelease custody, is working.
Recidivism among those released under the First Step
Act is down markedly, and the system’s rehabilitative
aims are bearing fruit. The most recently available
annual report detailing outcomes under the First Step
Act showed that 90.3% of the 44,673 former prisoners
released under the Act had successfully reintegrated
into the communities to which they returned. Dep’t of
Just., First Step Act Annual Report 40-44 (June
2024), https://perma.cc/9N8Z-JBG6. The 9.7% of former prisoners who recidivated after being released
under the First Step Act paled in comparison to the
44.8% overall recidivism rate of all prisoners released
from Bureau of Prisons custody during that same
year. Avinash Bhati, First Step Act: An Early Analysis of Recidivism, Council on Crim. Just. (Dec. 2024),
https://perma.cc/PL4E-T4B4. 3 Even after adjusting
for differences between federal prisoners generally
and those eligible for release under the First Step Act,
3
This divergence between the recidivism rates of former prisoners released under the First Step Act and the broader universe
of prisoners released from Bureau of Prisons custody continues a
years-long trend. See, e.g., Dep’t of Just., First Step Act Annual
Report 40-43 (Apr. 2023), https://perma.cc/2RJF-7RUH; Avinash
Bhati, First Step Act: An Early Analysis of Recidivism, Council
on Crim. Just. (Aug. 2023), https://perma.cc/C22N-HV8C (12.4%
recidivism rate among individuals released under the First Step
Act in 2023, compared to 46.2% recidivism rate of all prisoners
released from Bureau of Prisons custody that year).
10
the recidivism rate among First Step Act-released individuals is 55% lower than prisoners with similar
profiles released prior to the First Step Act’s enactment. Id. And the majority of those released under
the First Step Act had completed at least one recidivism reduction program. Avinash Bhati, Time Sentenced and Time Served: Exploring the Impact of the
First Step Act, Council on Crim. Just. (Dec. 2023),
available at FSA.counciloncj.org/fsa.
The potential fiscal benefits that animated wide bipartisan interest in the First Step Act have also
started coming to fruition since the law’s enactment.
The community-based forms of prerelease custody
that the First Step Act make possible are far more
cost-effective. The Bureau of Prisons reported that in
2024, the federal government spent $129.21 each day
to house a prisoner in a Bureau of Prisons facility,
compared to $119.73 in a residential reentry center
(halfway house). See Annual Determination of Average Cost of Incarceration Fee (COIF), 90 Fed. Reg.
58,059 (Dec. 15, 2025). The daily cost of home confinement is less than half of the federal outlay for housing
a prisoner in a Bureau of Prisons facility. See Home
Confinement Under the Coronavirus Aid, Relief, and
Economic Security (CARES) Act, 88 Fed. Reg. 19,830,
19,833, 19,839 (Apr. 4, 2023).
The taxpayer savings that helped drive support for
the First Step Act on both sides of the political aisle
will be realized only if prisoners like Mr. Maxwell are,
in fact, transferred to prerelease custody once they become eligible. If the promise of transfer to less-restrictive, community-based forms of prerelease custody
goes unfulfilled, savings will not be fully realized. By
11
reversing the decision below, the Court can uphold the
judgment of Congress that the First Step Act will
“make the system work better for the taxpayers, help
law enforcement fight crime, and put a stopper in the
revolving prison door.” 164 Cong. Rec. S7838 (daily
ed. Dec. 19, 2018) (statement of Sen. Grassley).
II. CONGRESS INTENDED FOR THE FIRST
STEP ACT’S INCENTIVE SYSTEM TO BE
ENFORCEABLE AND REVIEWABLE.
The First Step Act Imposes A
Mandatory Obligation On The Bureau
Of Prisons To Apply Earned Time
Credits To Prerelease Custody Or
Supervised Release.
“Congress clearly knows how to use mandatory language when it so desires.” Fogerty v. Fantasy, Inc.,
510 U.S. 517, 524 n.11 (1994). The First Step Act’s
system of earned time credits for federal prisoners is
a perfect example. Congress eliminated the Bureau’s
discretion about awarding earned time credits when
statutory prerequisites have been met. The text and
structure of the statute make this abundantly clear.
First, the statute’s incentive provision speaks in
mandatory terms. It instructs that a prisoner “shall”
earn time credits for participating in qualifying activities. The Bureau of Prisons “shall” apply those credits toward prerelease custody or supervised release.
And once the statutory requirements (including a sufficient number of earned time credits) have been met,
the Bureau “shall” transfer eligible prisoners to prerelease custody or supervised release. See supra, p. 8
(describing statutory language); 18 U.S.C. § 3632(d).
12
“It is undisputed that the word ‘shall’ imposes a mandatory command.” Bufkin v. Collins, 604 U.S. 369,
379 (2025).
Second, § 3632’s “adjacent provisions * * * underscore its mandatory nature.” Me. Cmty. Health Options v. United States, 590 U.S. 296, 310 (2020). Contrast the language in the incentive system provision
with, for instance, § 3622, which provides that the Bureau of Prisons “may release a prisoner * * * for a limited period” for certain enumerated purposes; or
§ 3621(b), under which the Bureau “may designate
any available penal or correctional facility * * * that
the Bureau determines to be appropriate and suitable.”4 These provisions—which also concern prisoner
housing and relocation—authorize, but do not obligate, the Bureau of Prisons to act. Cf. Murphy v.
4
18 U.S.C. §§ 3621(b), 3622 (emphases added). The First Step
Act’s structure also does not support treating prerelease custody
eligibility as a decision regarding a mere “place of imprisonment.” Contra Osorio-Calderon v. Warden, FCI Sandstone, 179
F.4th 1158, 1164 (8th Cir. 2026). The First Step Act has a separate provision outlining the Bureau’s authority to determine a
person’s “place of imprisonment”; that section grants the Bureau
discretion to make certain housing designations. See 18 U.S.C.
§ 3621. But the statutory criteria governing whether someone
has become eligible for prerelease custody in the first place is different. Consequently, the Bureau of Prisons has not been vested
with the same level of discretion when it comes to that initial
eligibility inquiry—to conclude otherwise would read the incentive system’s critical, mandatory language out of the statute. See
Duncan v. Walker, 533 U.S. 167, 174 (2001) (it is the court’s duty
“to give effect, if possible, to every clause and word of a statute,”
particularly “when the term occupies so pivotal a place in the
statutory scheme” (citation omitted)).
13
Smith, 583 U.S. 220, 226-227 (2018) (“If Congress had
wished to confer the same discretion in [one section of
a statute] that it conferred in [another section], we
very much doubt it would have bothered to * * * replace those old discretionary words with new mandatory ones.”); Russello v. United States, 464 U.S. 16, 23
(1983) (rejecting argument that “differing language”
in two separate provisions of the same statute could
nevertheless be read to reveal “the same meaning in
each”). Accordingly, “[w]hen, as is the case here, Congress distinguishes between ‘may’ and ‘shall,’ it is generally clear that ‘shall’ imposes a mandatory duty.”
Me. Cmty. Health, 590 U.S. at 311 (cleaned up).
Third, outside of the specifically enumerated, limited exceptions to the earned time credit system, no
other qualifying language appears in that statutory
provision. See 18 U.S.C. § 3632(d)(4)(B), (D), (E).
Again, that absence is telling. And again, it stands in
contrast to other portions of the statute, such as the
provisions governing housing decisions or the use of
less restrictive conditions, which qualify the Bureau
of Prisons’ obligation to act only “to the extent practicable.” Id. §§ 3632(c), 3624(g)(4).
The First Step Act thus plainly provides that the Bureau of Prisons must apply earned time credits toward
prerelease custody or supervised release. That obligation lies at the heart of the statutory incentive
scheme. Indeed, the Bureau itself understands its
statutory mandate to be essential to the First Step
Act’s purpose. As recently as May 2025, it issued a
directive to “[e]xpand [h]ome [c]onfinement[] [and]
[a]dvance [the] First Step Act,” emphasizing its “responsibility to * * * ensure that eligible incarcerated
14
individuals * * * are transferred to home confinement
as soon as statutorily possible.” Press Release, Federal Bureau of Prisons, Federal Bureau of Prisons Issues Directive to Expand Home Confinement, Advance First Step Act (May 28, 2025). The agency explained that this directive “honors the bipartisan vision behind the FSA.” Id.
The First Step Act’s Prerelease Custody
Incentives Are Enforceable Through
Habeas Petitions.
Habeas petitions like Mr. Maxwell’s permit the
courts to give effect to congressional intent and ensure
that the First Step Act remains a meaningful tool for
prison reform. Mr. Maxwell’s petition alleges that he
was denied transfer to prerelease custody despite
meeting all statutory eligibility requirements. Notably, he does not challenge the Bureau of Prisons’
choice of which type of community-based confinement
to assign him—he challenges the denial of access to
any prerelease custody arrangement. Pet. for Certiorari, at 1. By failing to transfer him, the Bureau of
Prisons is unlawfully confining Mr. Maxwell using a
meaningfully different—and more restrictive—form
of custody in violation of its statutory obligations under the First Step Act. Accordingly, Mr. Maxwell’s petition is properly understood as a challenge to the lawfulness of his custodial arrangement—and is cognizable in habeas.
“[W]hen Congress enacts statutes, it is aware of relevant judicial precedent.” Merck & Co. v. Reynolds,
559 U.S. 633, 648 (2010). The First Step Act is no exception to this presumption; the text Congress chose
in the Act should be interpreted against the relevant
15
backdrop. See Hewitt, 606 U.S. at 431 (explaining, in
a case interpreting provisions of the First Step Act,
that “we recognize that Congress legislates against
the backdrop of certain unexpressed presumptions”)
(cleaned up). Decades of this Court’s precedent establish that habeas relief encompasses situations, like
the one here, that challenge the legality of a person’s
level of confinement. Had Congress intended to close
the door to habeas relief despite this precedent, it
would have said so. There was no need to adopt a statute-specific enforcement mechanism to enforce the
Act’s nondiscretionary duties.
In Preiser v. Rodriguez, 411 U.S. 475 (1973), the
Court explained that “the writ of habeas corpus
evolved as a remedy available to effect discharge from
any confinement contrary to the Constitution or fundamental law.” Id. at 485 (emphasis added). The
Court understood the scope of this remedy to include
both unlawful “confine[ment] in the wrong institution” and unlawful revocation of parole. Id. at 486.
And it applied that historical understanding to the respondents’ claim that “the deprivation of their goodconduct-time credits was causing or would cause them
to be in illegal physical confinement,” concluding it
was “beyond doubt” that such a claim sounded in habeas. Id. at 487-488. Likewise, in Edwards v. Balisok,
520 U.S. 641 (1997), the Court held that habeas was
the proper vehicle for challenging a “procedural defect” that would “necessarily imply the invalidity of
the deprivation of [petitioner’s] good-time credits.” Id.
at 646. And in Wilkinson v. Dotson, 544 U.S. 74
(2005), Justice Scalia recognized that “permissible habeas relief * * * includes ordering a ‘quantum change
16
in the level of custody,’ such as release from incarceration to parole.” Id. at 86 (Scalia, J., concurring) (citation omitted).
The majority of the courts of appeals to have considered the issue have correctly understood this Court’s
precedent to permit challenges to the level of custody
via a habeas petition.
In keeping with Preiser and Wilkinson, the First Circuit applied the “quantum change in the level of custody” framework to analyze whether a challenge was
cognizable in habeas. González-Fuentes v. Molina,
607 F.3d 864, 873 (1st Cir. 2010). In Molina, the court
considered the practical effects of the electronic supervision program at issue, in which participants “were
able to live with family members, work daily jobs, attend church, and reside in their own homes.” Id. This
constituted a “difference in the level of custody” from
institutional confinement “of sufficient magnitude to
justify the use of habeas corpus.” Id. at 873 n.7.
Similarly, in Levine v. Apker, 455 F.3d 71 (2d Cir.
2006), the Second Circuit concluded that habeas petitions were proper vehicles to challenge “the degree of
physical restriction.” Id. at 78. And in Woodall v. Federal Bureau of Prisons, 432 F.3d 235 (3d Cir. 2005),
the Third Circuit applied the Court’s reasoning to a
petitioner’s challenge to the Bureau’s regulations limiting prisoner placements in Community Corrections
Centers (CCCs). As the court explained, “[c]arrying
out a sentence through detention in a CCC is very different from carrying out a sentence in an ordinary penal institution.” Id. at 243. CCCs “satisfy different
goals from other types of confinement,” meaning
“placement in a CCC represents more than a simple
17
transfer.” Id. Accordingly, the prisoner’s challenge to
the “execution” of the sentence was cognizable in habeas. See id. at 241.
Precedent regarding challenges to parole revocation
should also guide the Court’s treatment of challenges
to community-based placement eligibility denials.
There are critical similarities between parole, supervised release, and prerelease custody. Indeed, “Congress established the system of supervised release”
“[i]n parole’s place.” United States v. Haymond, 588
U.S. 634, 652 (2019) (plurality op.). And the former
federal parole system, like the alternative forms of
confinement found in § 3632, was designed “to help individuals reintegrate into society as constructive individuals as soon as they are able, without being confined for the full term of the sentence imposed.” Morrissey v. Brewer, 408 U.S. 471, 477 (1972).
This Court long understood challenges to parole decisions to be cognizable in habeas. E.g., id.; Gagnon
v. Scarpelli, 411 U.S. 778, 791 (1973) (permitting petitioner to challenge parole revocation through writ of
habeas corpus). And courts of appeals, including the
Fifth Circuit, have recognized that the same is true
for challenges to a term of supervised release. See,
e.g., Johnson v. Pettiford, 442 F.3d 917 (5th Cir. 2006)
(per curiam) (permitting prisoner to challenge duration of term of supervised release via habeas petition);
Mujahid v. Daniels, 413 F.3d 991 (9th Cir. 2005) (recognizing that petitioner could bring habeas petition
challenging the Bureau of Prisons’ calculation of
earned time credits as applied to his term of supervised release). In other words, each time someone
seeks to move a rung up the ladder toward full
18
release—from full custodial confinement to parole or
supervised release, or from supervised release to full
release—habeas is the proper vehicle.
The First Step Act’s structure supports treating parole, supervised release, and prerelease custody similarly for purposes of judicial review. The statute
clearly identifies prerelease custody and supervised
release as related forms of community-based confinement that are distinct in form and purpose from full
custodial confinement. See 18 U.S.C. § 3632. And
they are the only two alternative confinement methods to which the Bureau of Prisons must apply earned
time credits. Thus, prerelease custody should be considered through the same lens. See Davis v. Mich.
Dep’t of Treasury, 489 U.S. 803, 809 (1989) (“It is a
fundamental canon of statutory construction that the
words of a statute must be read in their context and
with a view to their place in the overall statutory
scheme.”).
*
*
*
The First Step Act’s text and this Court’s precedent
establish that Mr. Maxwell may challenge the Bureau
of Prisons’ refusal to transfer him to prerelease custody through a habeas petition. Any other result
would undermine congressional intent: It would
make little sense to insulate from judicial review the
Bureau’s noncompliance with the carefully designed,
nondiscretionary system outlined in § 3632. Determining that a claim like Mr. Maxwell’s must be
brought in an action under Bivens v. Six Unknown
Named Agents of Federal Bureau of Narcotics, 403
U.S. 388 (1971), would create artificial barriers to reviewability, see Petitioner Br. 30, thus “produc[ing] a
19
result demonstrably at odds with the intentions of
[the First Step Act’s] drafters.” Griffin v. Oceanic
Contractors, Inc., 458 U.S. 564, 571 (1982).
CONCLUSION
The Fifth Circuit’s judgment should be reversed.
DELIA SCOVILLE
HOGAN LOVELLS
CADWALADER US LLP
125 High Street, Suite 2010
Boston, MA 02110
(671) 371-1000
JO-ANN TAMILA SAGAR
REEDY C. SWANSON
Counsel of Record
MARLAN GOLDEN
HOGAN LOVELLS
CADWALADER US LLP
555 Thirteenth Street, N.W.
Washington, DC 20004
(202) 637-5600
reedy.swanson@hlc.com
Counsel for Amici Curiae
AUGUST 2026
This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.