Ellingburg v. United States

Page Proof Pending Publication
A
P R E L I M I N A R Y P R I N T
Volume 607 U. S. Part 1
Pages 163–186
OFFICIAL REPORTS
OF
T H E S U P R E M E C O U R T
January 20, 2026
REBECCA A. WOMELDORF
reporter of decisions
N O T I C E : This preliminary print is subject to formal revision before
the bound volume is published. Users are requested to notify the Reporter
of Decisions, Supreme Court of the United States, Washington, D. C. 20543,
pio@supremecourt.gov, of any typographical or other formal errors.

Page Proof Pending Publication
OCTOBER
TERM, 2025
163
Syllabus
ELLINGBURG v. UNITED STATES
certiorari
to the united states court of appeals for
the eighth circuit
No. 24–482. Argued October 14, 2025—Decided January 20, 2026
The Mandatory Victims Restitution Act of 1996 requires defendants con-
victed of certain federal crimes to pay monetary restitution to victims.
Although petitioner Ellingburg committed his crime before the MVRA's
enactment, he was sentenced under the MVRA and ordered to pay
$7,567.25 in restitution. Ellingburg raised an Ex Post Facto Clause
challenge to his continued restitution obligation. The Eighth Circuit
concluded that restitution under the MVRA is not criminal punishment
subject to the Ex Post Facto Clause.
Held: Restitution under the MVRA is plainly criminal punishment for pur-
poses of the Ex Post Facto Clause. Whether a law violates the Ex Post
Facto Clause requires evaluating whether the law imposes a criminal or
penal sanction as opposed to a civil remedy. That question is one “of
statutory construction” that requires the Court to “consider the stat-
ute's text and its structure.” Smith v. Doe, 538 U. S. 84, 92 (quotation
marks omitted). When viewed as a whole, the MVRA makes abun-
dantly clear that restitution is criminal punishment. The MVRA labels
restituti on as a “pena lty” for a cr i mi na l “offense. ” 18 U. S. C.
§ 3663A(a)(1). Only a criminal defendant convicted of a qualifying
crime may be ordered to pay restitution. Restitution is imposed at
sentencing for that offense together with other criminal punishments
such as imprisonment and fnes. And at the sentencing proceeding
where restitution is imposed, the Government, not the victim, is the
party adverse to the defendant. Further, the federal MVRA restitu-
tion regime is codifed in Title 18, “Crimes and Criminal Procedure,”
and the statutory provisions authorizing restitution orders are con-
tained in Chapter 232 of that Title, entitled “Miscellaneous Sentencing
Provisions.” A district court imposing restitution must follow the pro-
cedures applicable to other criminal penalties.
The Court's precedents have understood restitution under the MVRA
to be criminal punishment. See Manrique v. United States, 581 U. S.
116, 118. And the Court's precedents on related issues further buttress
the conclusion that MVRA restitution is criminal punishment. See
United States v. Bajakajian, 524 U. S. 321, 328; United States v. One
Assortment of 89 Firearms, 465 U. S. 354, 363–366.

Page Proof Pending Publication
164 EL
LINGBURG v. UNITED STATES
Opinion of the Court
Finally, while Congress intended restitution under the MVRA to both
punish
offenders and compensate victims, victims cannot initiate or
settle the restitution process as they would if it were a civil pro-
ceeding. The text and structure of the Act demonstrate that Congress
intended restitution under the Act to impose criminal punishment.
Pp. 165–169.
113 F. 4th 839, reversed and remanded.
Kavanaugh, J., delivered the opinion for a unanimous Court. Thomas,
J., fled a concurring opinion, in which Gorsuch, J., joined, post, p. 169.
Amy Mason Saharia argued the cause for petitioner.
With her on the briefs were Lisa S. Blatt, Atticus W. De-
Prospo, Rohit P. Asirvatham, and R. Shane Roberts, Jr.
Ashley Robertson argued the cause for the United States
urging vacatur. On the briefs were Solicitor General
Sauer, Deputy Solicitor General Feigin, Acting Assistant
Attorney General Galeotti, and Andrew W. Lang.
John F. Bash, by appointment of the Court, 605 U. S. 908,
argued the cause as amicus curiae in support of the judg-
ment below. With him on the brief were Nicholas J. Ca-
luda, Alex Van Dyke, and Christopher G. Michel.*
Justice Kavanaugh delivered the opinion of the Court.
Under the Mandatory Victims Restitution Act of 1996, de-
fendants convicted of certain federal crimes must pay mone-
tary restitution to the victims. That Act, known as the
*Andrew T. Tutt, Dana Kagan McGinley, and Nicole L. Masiello fled
a brief for Restitution Scholars as amici curiae urging reversal. Briefs
of amici curiae urging vacatur were fled for the Cato Institute et al.
by Matthew P. Cavedon; for the Constitutional Accountability Center by
Elizabeth B. Wydra and Brianne J. Gorod; for the NACDL et al. by
Lauren Willard Zehmer, Abby Wright, Hassan Ahmad, and Jeffrey T.
Green; and for Wayne A. Logan by Krystal B. Swendsboe.
Briefs of amici curiae were fled for Beth A. Colgan, pro se; for Jennifer
Laurin by Mahogane Reed; and for Debra Ricketts-Holder by Allyson N.
Ho, Bradley G. Hubbard, and Paul G. Cassell.

Page Proof Pending Publication
Cite
as: 607 U. S.163 (2026)
165
Opinion of the Court
MVRA, became law on April 24, 1996. Petitioner Elling-
burg
was sentenced later in 1996 and ordered to pay restitu-
tion in the amount of $7,567.25. He has not yet satisfed
that obligation.
Ellingburg has now raised an Ex Post Facto Clause chal-
lenge to his continued restitution obligation because he com-
mitted his crime before the enactment of the MVRA. The
Ex Post Facto Clause issue turns in part on the threshold
question of whether restitution under the MVRA is criminal
punishment. Applying Circuit precedent, the U. S. Court of
Appeals for the Eighth Circuit concluded that restitution
under the MVRA is not criminal punishment. 113 F. 4th
839, 841–842 (2024). We now reverse.
In this Court, Ellingburg and the United States agree that
the Eighth Circuit erred and that MVRA restitution is crimi-
nal punishment. The Court therefore appointed John F.
Bash as amicus curiae to defend the judgment of the Eighth
Circuit. 605 U. S. 908 (2025). He has ably discharged his
responsibilities.
When determining whether a law violates the Ex Post
Facto Clause, the Court must evaluate whether the law im-
poses a criminal or penal sanction as opposed to a civil rem-
edy. Assessing whether “a statutory scheme is civil or
criminal is frst of all a question of statutory construction”
that requires the Court to “consider the statute's text and
its structure.” Smith v. Doe, 538 U. S. 84, 92 (2003) (quota-
tion marks omitted).
1
1
If the text and structure of a statute do not demonstrate that Congress
intended criminal punishment, the statute may still be deemed criminal
or penal if the “party challenging the statute provides `the clearest proof '
that `the statutory scheme [is] so punitive either in purpose or effect as to
negate [the Government's] intention' to deem it `civil.' ” Kansas v. Hen-
dricks, 521 U. S. 346, 361 (1997) (quoting United States v. Ward, 448 U. S.
242, 248–249 (1980)).

Page Proof Pending Publication
166 EL
LINGBURG v. UNITED STATES
Opinion of the Court
Here, the statutory analysis is straightforward: Restitu-
ti
on under the MVRA is plainly criminal punishment for pur-
poses of the Ex Post Facto Clause.
2
Numerous features of the MVRA lead to that conclusion.
The MVRA labels restitution as a “penalty” for a criminal
“offense.” 18 U. S. C. § 3663A(a)(1). A court may order
restitution only with respect to a criminal “defendant” and
only after that defendant's conviction of a qualifying crime.
Ibid. Restitution is imposed during “sentencing ” for the of-
fense. Ibid. At the sentencing proceeding where restitu-
tion is ordered, the Government, not the victim, is the party
adverse to the defendant.
At sentencing, restitution is imposed together with other
criminal punishments such as imprisonment and fnes. In-
deed, for misdemeanors, restitution may be “in lieu of ” those
punishments, making restitution the sole punishment for a
federal offense in certain circumstances. Ibid. In addition,
when a defendant does not make the required restitution
payments, the court may modify the terms of his supervised
release or probation and impose imprisonment if the court
determines that “alternatives to imprisonment are not ade-
quate to serve the purposes of punishment and deterrence.”
§§ 3614(b)(2), 3613A(
a)(1).
As
a further sign that MVRA restitution is criminal pun-
ishment, the federal MVRA restitution regime is codifed in
Title 18, “Crimes and Criminal Procedure.” The main stat-
utory provisions authorizing restitution orders, § 3663 and
§ 3663A, are contained in Chapter 232 of Title 18,
entitled
“Miscel
laneous Sentencing Provisions.” Another st
atutory
prov
ision, § 3556, states that a court “shall order restituti
on”
i
n “imposing a sentence” for covered offenses. That provi-
sion is housed in Chapter 227 of Title 18, entitled “Sen-
2
Amicus contends that Ellingburg 's restitution was not imposed under
the MVRA even though Ellingburg was sentenced after the MVRA took
effect in 1996. But the Eighth Circuit decided this case on the under-
standing that the MVRA was applied to Ellingburg.

Page Proof Pending Publication
Cite
as: 607 U. S.163 (2026)
167
Opinion of the Court
tences.” The statute also refers to a “sentence that imposes
an
order of restitution.” § 3664(o). When imposing restitu-
tion, a district court must follow the procedures applicable
to other criminal penalties, including the Federal Rules of
Criminal Procedure. See § 3664(c). And § 3663(c) directs
the U. S. Sentencing Commission to promulgate criminal
sentencing guidelines for restitution. § 208, 110 Stat. 1240;
§ 3663(c)(7).
When viewed as a whole, then, the MVRA makes abun-
dantly clear that restitution is criminal punishment. We are
not saying that all of the statutory features present here
are necessary to constitute criminal punishment, but they
are suffcient.
Given the statutory text and structure, it comes as no
surprise that this Court's precedents have understood restitu-
tion under the MVRA to be criminal punishment. In Man-
r ique v. Un i ted Sta tes, the Cour t stated that the MVRA
requires courts “to impose restitution as part of the sen-
tence.” 581 U. S. 116, 118 (2017). In Pasquantino v.
United States, the Court explained that restitution under the
MVRA is designed “to mete out appropriate criminal punish-
ment.” 544 U. S. 349, 365 (2005). And in Paroline v.
United States, the Court repeated that description. 572
U. S. 434, 456 (2014).
This Court's precedents on related issues further buttress
the conclusion that MVRA restitution is criminal punish-
ment. For example, forfeiture under 18 U. S. C. § 982(a)(1)—
like restitution under the MVRA—occurs at the culmination
of a criminal proceeding and requires conviction of an under-
lying crime. That forfeiture, the Court concluded, is crimi-
nal punishment. See United States v. Bajakajian, 524 U. S.
321, 328 (1998). By contrast, the Court determined that the
forfeiture of frearms under 18 U. S. C. § 924(d) ordered dur-
ing an in rem proceeding is a civil sanction. United States
v. One Assortment of 89 Firearms, 465 U. S. 354, 363–366
(1984).

Page Proof Pending Publication
168 EL
LINGBURG v. UNITED STATES
Opinion of the Court
As the Government cautions, not everything that occurs
at
criminal sentencing or even that appears in a criminal
judgment may necessarily be part of the punishment. See
Tr. of Oral Arg. 22–23. But we need not explore that point
further here because, for the reasons we have explained,
MVRA restitution is criminal punishment.
Amicus relies heavily on Smith v. Doe, 538 U. S. 84.
There, this Court considered a law requiring a defendant
convicted of certain crimes to register as a sex offender.
The Court held that the registration mandate was civil.
The legislature adopted “distinctly civil procedures” for the
imposition of registration requirements. Id., at 96 (quota-
tion marks omitted). By contrast, to reiterate what we said
above, MVRA restitution is labeled as a penalty, is codifed
in the criminal code, is predicated on a criminal conviction,
is imposed against a criminal defendant, is sometimes im-
posed in lieu of other penalties, is ordered at sentencing
where the United States is the adverse party, and can result
in resentencing when the defendant refuses to pay. So
Smith v. Doe does not control.
It is true, as amicus thoroughly explains, that the MVRA
seeks to compensate crime victims, a nonpunitive goal.
Several provisions of the MVRA refect that objective,
such as a provision requiring that restitution amounts be
based on a victim's loss. § 3664(f )(1)(A); see also, e. g.,
§ 3664( j). But those provisions show only that Congress in-
tended restitution under the MVRA to both punish and com-
pensate. And so long as the text and structure of the Act
demonstrate that Congress intended at least “to impose pun-
ishment,” that “ends the inquiry.” Smith, 538 U. S., at 92;
see Paroline, 572 U. S., at 456; Pasquantino, 544 U. S.,
at 365.
To be sure, as amicus points out, victims receive notice
of sentencing proceedings and are consulted by prosecutors.
But outreach to victims and some participation by victims in
criminal proceedings are not unusual. See, e. g., § 3771(a).
With respect to restitution, the key point is that victims do

Page Proof Pending Publication
Cite
as: 607 U. S. 163 (2026)
169
Thomas, J., concurring
not have the power to initiate or settle the restitution proc-
ess
as they would if it were a civil proceeding.
Our ruling today does not mean that a restitution statute
can never be civil. But the statutory text and structure of
the MVRA demonstrate that restitution under that Act is
criminal punishment.
For those reasons, we reverse the judgment of the U. S.
Court of Appeals for the Eighth Circuit and remand the case
for further proceedings consistent with this opinion. On re-
mand, the Court of Appeals may consider the Government's
separate arguments for affrmance of the District Court's
judgment.
It is so ordered.
Justice Thomas, with whom Justice Gorsuch joins,
concurring.
I join the Court's opinion in full because it correctly ap-
plies our precedent. I write separately to clarify the foun-
dation of that precedent. This Court's 1798 decision in Cal-
der v. Bull, 3 Dall. 386, established that the Ex Post Facto
Clauses forbid only those retroactive laws that impose “pun-
ishment” for a “crime.” Id., at 386, 389–391 (opinion of
Chase, J.). Over the 228 years since Calder, the Court has
struggled to articulate what it means for a law to impose
punishment for a crime, and thus to be subject to the Ex Post
Facto Clauses. The Court's more recent precedents have
implemented Calder through two multifactor tests that turn
largely on whether the legislature labels the law as criminal
or civil. But in 1798, “punishment” for a “crime” would have
been understood to refer to any coercive penalty for a public
wrong. Many laws that are nominally civil today would
therefore have been subject to the Ex Post Facto Clauses
under Calder. I would restore Calder's approach to the Ex
Post Facto Clauses.
I
The Constitution twice prohibits ex post facto laws. As
to the Federal Government, it provides that “No Bill of At-

Page Proof Pending Publication
170 EL
LINGBURG v. UNITED STATES
Thomas, J., concurring
tainder or ex post facto Law shall be passed.” Art. I, § 9,
cl.
3. As to the States, it provides that “No State shall . . .
pass any Bill of Attainder, ex post facto Law, or Law impair-
ing the Obligation of Contracts.” Art. I, § 10, cl. 1. In its
general sense, a law is ex post facto—meaning “after the
fact”—when it applies retroactively to conduct that occurred
before the law was enacted.
The two Ex Post Facto Clauses refected the importance
of the protection to the Founding Fathers. James Iredell
believed that the protection against ex post facto laws was
“the most essential right of all,” which was worth “ten thou-
sand declarations of rights” without it. Answers to Mr. Ma-
son's Objections to the New Constitution, in Pamphlets on
the Constitution of the United States 334, 368 (P. Ford ed.
1888) (reprint 1968). “A man may feel some pride in his
security,” Iredell wrote, “when he knows that what he does
innocently and safely to-day in accordance with the laws of
his country, cannot be tortured into guilt and danger to-
morrow.” Ibid. In the Federalist Papers, Alexander Ham-
ilton placed “the prohibition of ex post facto laws” among
the greatest “securities to liberty and republicanism.” The
Federalist No. 84, p. 571 (P. Ford ed. 1898). James Madison
argued that ex post facto laws are “contrary to the frst prin-
ciples of the social compact.” Id., No. 44, at 296.
Ex post facto laws lack legitimacy because laws must pre-
cede the actions that they govern. Laws regulating actions
after the fact “deprive citizens of notice and fair warning
and are, therefore, an affront to man's `reason and freewill.' ”
Peugh v. United States, 569 U. S. 530, 561 (2013) (Thomas,
J., dissenting) (quoting 1 W. Blackstone, Commentaries on
the Laws of England 39 (1765) (Blackstone)). As Blackstone
explained, “it is impossible that the party could foresee” that
his actions would become unlawful if they were lawful when
he took them. Id., at 46. When a man had no notice that
his actions were unlawful, “all punishment for not abstaining
must of consequence be cruel and unjust.” Ibid. An ex

Page Proof Pending Publication
Cite
as: 607 U. S. 163 (2026)
171
Thomas, J., concurring
post facto law is even “more unreasonable,” Blackstone
thought,
than the reviled practice of the Roman emperor
“Caligula, who . . . wrote his laws in a very small character,
and hung them up upon high pillars, the more effectually to
ensnare the people.” Ibid.
Many believed that no constitutional prohibition was nec-
essary because an ex post facto law would be invalid as con-
trary to natural law. At the Constitutional Convention, Oli-
ver Ellsworth argued that “[i]t cannot . . . be necessary to
prohibit” ex post facto laws because “no lawyer, no civilian
. . . would not say that ex post facto laws were void of them-
selves.” 2 M. Farrand, Records of the Federal Convention
of 1787, p. 376 (1966) (Farrand). James Wilson thought that
a constitutional prohibition on ex post facto laws would sug-
gest that the Convention denied that such laws were already
forbidden by natural law, to its embarrassment. “It will
bring refexions on the Constitution—and proclaim that we
are ignorant of the frst principles of Legislation.” Ibid.
In fact, “[a]ll the delegates who spoke explicitly or implicitly
regarded an ex post facto law as a violation of natural law,
and most of them therefore thought it unnecessary to include
such a basic natural law principle in the written constitu-
tion.” S. Sherry, The Founders' Unwritten Constitution, 54
U. Chi. L. Rev. 1127, 1157 (1987). The Convention nonethe-
less adopted the Ex Post Facto Clauses after it was urged
that “the Judges can take hold of ” them if a legislature ever
enacts such a law. 2 Farrand 376 (Williamson).
II
The lodestar of this Court's ex post facto jurisprudence is
Calder v. Bull, which established that the Ex Post Facto
Clauses apply only to laws that impose “punishment” for
“crime[s].” 3 Dall., at 389–391. Calder—and especially
Justice Chase's lead seriatim opinion—has since provided
the defnitive gloss on the Ex Post Facto Clauses. Justice
Chase's defnition of “ex post facto laws” continues to “pro-

Page Proof Pending Publication
172 EL
LINGBURG v. UNITED STATES
Thomas, J., concurring
vid[e] an authoritative account of the scope of the Ex Post
F
acto Clause.” Stogner v. California, 539 U. S. 607, 611
(2003); accord, e. g., Peugh, 569 U. S., at 532–533. This
Court's modern precedents, including today's opinion, decide
whether a law is subject to the Ex Post Facto Clauses based
on whether it satisfes Calder's requirement that it be “crimi-
nal or penal.” Weaver v. Graham, 450 U. S. 24, 29 (1981);
see also ante, at 165. But this Court's implementation of
Calder seems to have lost sight of how Calder would have
been understood when it was decided.
Calder concerned a state legislature's intervention in a
traditional private-law dispute. The case arose out of the
probate proceedings for the estate of a Connecticut man
named Normand Morison. In 1779, Morison wrote a will
leaving his property to his wife Abigail and her heirs. See
8 The Documentary History of the Supreme Court of the
United States, 1789–1800, p. 89 (M. Marcus ed. 2007). Soon
after Morison wrote his will, he and Abigail had a son. Ibid.
When the son was young, Morison died. Ibid. Abigail
started the probate process so that she could inherit Mori-
son's property as his will promised. Ibid. But the probate
judge held up the process because he wanted to check
whether the birth of their son affected the will's validity.
Ibid. At the time, Abigail did not mind the delay because
the default intestacy rules would have given the property to
her and her son anyway. Id., at 89–90.
But later events jeopardized Abigail's right to inherit Mor-
ison's property. First, Connecticut changed the default in-
testacy rules so that, absent a valid will, property like Mori-
son's would pass to his nearest blood relative, not to a
surviving spouse like Abigail. Id., at 90, and n. 9. Second,
the probate court, without telling Abigail, disapproved Mori-
son's original will. Id., at 90. And third, a few years later,
Abigail and Morison's son died. Ibid. After her son died,
Abigail and her new husband Caleb Bull returned to probate
court. Ibid. The Bulls were surprised to learn that the

Page Proof Pending Publication
Cite
as: 607 U. S. 163 (2026)
173
Thomas, J., concurring
probate court had disapproved Morison's will years earlier
w
ithout telling Abigail—apparently on the theory that he
would have wanted to change it after his son was born.
Ibid. And, since Connecticut had enacted new default intes-
tacy rules in the meantime, the absence of a valid will meant
that Morison's estate would pass on to his blood relatives
rather than to Abigail. Ibid. Morison's blood relatives, the
Calders, were suddenly the lawful heirs of his property. Id.,
at 90–91.
Abigail and Caleb Bull petitioned the Connecticut Legisla-
ture for relief. They convinced the legislature that Morison
did not intend to revoke his will after his son died, so the
legislature issued a decree in 1795 granting them a rehearing
on the validity of Morison's original will. Id., at 91. The
legislature's 1795 decree effectively vacated the original pro-
bate-court decision deeming the will void and granted a new
hearing at which the probate court could adjudicate the will's
validity. Id., at 91, 107. The probate court held a new
hearing pursuant to the decree and concluded that Morison's
original will was valid after all. Id., at 91. Having now
lost in probate court, the Calders appealed this new decision
and challenged the legislature's decree that had granted the
Bulls a new hearing. Id., at 91–92. The Calders argued,
eventually to this Court, that the legislature's decree was an
unconstitutional ex post facto law because it retroactively
voided the probate court's original decision, which caused
them to lose an inheritance that would otherwise have been
theirs. Id., at 92.
In Calder v. Bull, this Court gave two reasons why the
legislature's decree granting a new trial did not violate the
State Ex Post Facto Clause. See Art. I, § 10, cl. 1. The
frst reason was that the legislature's decree operated on a
legal decision by a court, not a past action by the Calders.
There was nothing that the parties had done in the past that
was “in any manner affected by the law or resolution of Con-
necticut.” 3 Dall., at 392 (opinion of Chase, J.). “It does not

Page Proof Pending Publication
174 EL
LINGBURG v. UNITED STATES
Thomas, J., concurring
concern, or relate to, any act done by them.” Ibid. In-
stead,
the decree effectively vacated a probate decision and
gave the court an opportunity to correct its decision, just
like appellate courts do every day. Ibid. The new hearing,
after all, was conducted under the same substantive laws as
were in force at the original hearing. In fact, the Justices
seemed to believe that the Connecticut Legislature's “resolu-
tion or law,” id., at 387, was not a legislative but a judicial
act, consistent with state legislatures' traditional judicial
power to review state-court decisions—and thus could not
be an ex post facto “law” at all. See id., at 395–396 (opinion
of Paterson, J.) (“[W]e may, in the present instance, consider
the Legislature of the State, as having acted in their custom-
ary judicial capacity”); id., at 398 (opinion of Iredell, J.) (“an
exercise of judicial, not of legislative, authority”); id., at 400
(opinion of Cushing, J.) (“If the act is a judicial act, it is not
touched by the Federal Constitution”).
The more lasting legacy of Calder, however, was its second
reason. Three of the four seriatim opinions in Calder lim-
ited the Ex Post Facto Clauses to laws imposing criminal
punishment. On their view, the Clauses did not forbid other
kinds of retroactive laws. Justice Chase, in what has proved
to be the most infuential opinion, gave a comprehensive ac-
count of the scope of the Ex Post Facto Clauses. He under-
stood the Clauses to be limited to retroactive laws imposing
criminal punishment. He explained that the Ex Post Facto
Clauses covered only those laws that “create” a “crime,” “ag-
gravate” a crime, “encrease the punishment” for a crime, or
“change the rules of evidence, for the purpose of conviction.”
Id., at 391.
Justice Chase's account implied that the Clauses did not
apply to other kinds of retroactive laws. First, on his view,
the Clauses did not touch private law. Justice Chase did
“not think” that the Ex Post Facto Clause “was inserted to
Secure the citizen in his private rights, of either property,
or contracts.” Id., at 390. Accordingly, the Clauses would

Page Proof Pending Publication
Cite
as: 607 U. S. 163 (2026)
175
Thomas, J., concurring
not prohibit a law that retroactively paused a debtor's obliga-
ti
ons to a creditor or voided a contract. Such laws, in Jus-
tice Chase's view, were governed by separate constitutional
prohibitions on impairing the obligations of contracts and
making anything but gold or silver a tender in payments
of debts. Ibid.; see Art. I, § 10, cl. 1. Second, because the
Clauses were meant as protections against the government,
they did not apply to laws that retroactively made a law
more lenient. They would not, for example, forbid a law
that retroactively “mollifes the rigor of the criminal law.”
Calder, 3 Dall., at 391 (opinion of Chase, J.).
The other two seriatim opinions in Calder that addressed
the question agreed. Justice Paterson explained that “the
meaning, annexed to the terms ex post facto laws, unques-
tionably refers to crimes, and nothing else.” Id., at 396.
Justice Iredell—an adamant proponent of the Ex Post Facto
Clauses, see supra, at 170—likewise thought that they did
not “extend to civil cases, to cases that merely affect the
private property of citizens.” Calder, 3 Dall., at 400. In-
stead, the Clauses applied only to laws that retroactively “in-
fict a punishment” for an “offence.” Ibid.
Some, myself included, have questioned whether Calder
was right to limit the Ex Post Facto Clauses to laws impos-
ing criminal punishment. See, e. g., Satterlee v. Mat-
thewson, 2 Pet. 380, 416, and n. (a) (1829) (opinion of Johnson,
J.) (arguing that the Ex Post Facto Clauses protect more); O.
Field, Ex Post Facto in the Constitution, 20 Mich. L. Rev.
315, 331 (1922) (“[T]here have been reputable authorities,
both past and present, who incline to the view that the ex
post facto provisions of the Constitution prohibited civil as
well as criminal legislation”); Eastern Enterprises v. Apfel,
524 U. S. 498, 538–539 (1998) (Thomas, J., concurring). Oth-
ers have defended Calder's limitation. See, e. g., R. Natel-
son, Statutory Retroactivity: The Founders' View, 39 Idaho
L. Rev. 489, 493–494 (2003); C. Nelson, Originalism and In-
terpretive Conventions, 70 U. Chi. L. Rev. 519, 578–582

Page Proof Pending Publication
176 EL
LINGBURG v. UNITED STATES
Thomas, J., concurring
(2003). But often lost in that debate has been the question
of
what Calder meant when it limited the Clauses to laws
imposing criminal punishment.
III
This Court's recent precedents have attempted to imple-
ment Calder's limitation of the Clauses to criminal punish-
ment. But, in doing so, they have adopted a framework that
turns largely on legislative labeling, has little basis in his-
tory, and is unnecessarily convoluted.
A
The modern precedents follow Calder's limitation of the
Ex Post Facto Clauses to laws that impose criminal punish-
ment. Based on “Justice Chase's now familiar opinion in
Calder,” this Court has repeatedly held that the Clauses
apply exclusively to laws that “retroactively alter the defni-
tion of crimes or increase the punishment for criminal acts,”
but not to other kinds of retroactive laws. Collins v. Young-
blood, 497 U. S. 37, 41–43 (1990); see also, e. g., Kansas v.
Hendricks, 521 U. S. 346, 370 (1997); Smith v. Doe, 538 U. S.
84, 92 (2003).
But in defning criminal punishment—and thus the scope
of Calder's limitation—the Court's modern framework ap-
plies two multifactor tests developed in the 20th century.
The frst multifactor test asks whether the legislature “in-
tended” the law to be viewed as criminal or penal. Smith,
538 U. S., at 92–93; see also ante, at 165, n. 1. Under this
test, a law is subject to the Ex Post Facto Clauses “[i]f the
intention of the legislature was to impose punishment.”
Smith, 538 U. S., at 92. But if “the intention was to enact a
regulatory scheme that is civil and nonpunitive,” then a law
is presumptively not subject to the Ex Post Facto Clauses.
Ibid. This test has included up to fve factors, most of which
allow a legislature to avoid the Clauses through labeling or se-
mantics. Those fve factors are whether the legislature had a

Page Proof Pending Publication
Cite
as: 607 U. S. 163 (2026)
177
Thomas, J., concurring
“ `preference' ” for “ `one label or the other,' ” id., at 93 (quot-
i
ng Hudson v. United States, 522 U. S. 93, 99 (1997)); whether
the “codifcation” was in the criminal or civil code, Smith, 538
U. S., at 94; whether legislative history or similar indicators
suggested nonpunitive goals, United States v. One Assort-
ment of 89 Firearms, 465 U. S. 354, 364 (1984); whether the
legislature provided traditional “safeguards associated with
the criminal process,” Smith, 538 U. S., at 96; and whether
the legislature gave enforcement power to an agency with
power to enforce civil laws, ibid.
The second multifactor test, which applies only if the law
survives the frst test, asks whether the law's other features
render it criminal or penal. See id., at 92. Under this sec-
ond test, a law is subject to the Ex Post Facto Clauses if it
is “so punitive either in purpose or effect as to negate [the
legislature's] intention to deem it civil.” Ibid. (internal quo-
tation marks omitted). The Court has at times said that
this second test is a function of up to seven factors: whether
the law involves an affrmative disability or restraint, im-
poses what has historically been regarded as a punishment,
requires a fnding of scienter, promotes the traditional aims
of punishment, applies to behavior which is already a crime,
lacks a rational alternative purpose, or is excessive in rela-
tion to that purpose. Kennedy v. Mendoza-Martinez, 372
U. S. 144, 168–169 (1963); see Smith, 538 U. S., at 97 (explain-
ing that Mendoza-Martinez factors “migrated” to ex post
facto jurisprudence). So long as a law survives the frst
test, “ `only the clearest proof ' ” under the second test can
make it subject to the Ex Post Facto Clauses. Smith, 538
U. S., at 92 (quoting Hudson, 522 U. S., at 100).
B
This modern framework is incongruous with the historical
purpose of Ex Post Facto Clauses. The modern framework,
because of how much it turns on legislative labeling and se-
mantics, allows a legislature to manipulate when the protec-

Page Proof Pending Publication
178 EL
LINGBURG v. UNITED STATES
Thomas, J., concurring
tion will apply. Under the frst test, for example, a legisla-
ture
might be able to impose a retroactive $10,000 fne on
some previously innocent conduct—such as drinking coffee
or going to the gym—if it labeled the law “civil,” placed it
in the civil code, expressed a nonpunitive purpose, omitted
protections for the accused, and let non-traditional prosecu-
tors (such as a health agency) enforce it. See supra, at 176–
177. Under the second test, the legislature might be able to
prevail by aggravating the fair-notice problems that moti-
vate the Clauses, such as by having its retroactive law im-
pose strict liability and then applying it only to previously
innocent conduct. See Mendoza-Martinez, 372 U. S., at 168–
169. It is unlikely that a constitutional prohibition designed
to curb a “cruel and unjust” legislature, 1 Blackstone 46, can
be so easily manipulated by the same legislature. See
United States v. Rahimi, 602 U. S. 680, 776 (2024) (Thomas,
J., dissenting) (explaining that the Constitution does not
“allow . . . majoritarian interests to determine” the scope of
“constitutional rights”).
The modern framework is also convoluted. It is diffcult
for courts, let alone ordinary citizens, to predict how this
Court will weigh the modern framework's combined twelve
factors spread over two tests. This Court's own precedents
admit that the twelve factors are “neither exhaustive nor
dispositive,” United States v. Ward, 448 U. S. 242, 249 (1980),
and need not be given equal weights, Smith, 538 U. S., at
105. Such “multifactor balancing test[s]” invite “policy-
driven, `arbitrary discretion.' ” Gamble v. United States,
587 U. S. 678, 724 (2019) (Thomas, J., concurring). We
should strive to avoid such unclear rules in any context, but
especially when interpreting a constitutional guarantee of
clear notice.
IV
When Calder said that the Ex Post Facto Clauses apply
only to laws imposing punishment for crimes, it was refer-
ring to an established category of laws. A crime meant a

Page Proof Pending Publication
Cite
as: 607 U. S. 163 (2026)
179
Thomas, J., concurring
“public wrong,” which is an injury to the sovereign in its
sovereig
n capacity. Calder therefore encompassed offenses
against the sovereign regardless of whether they were nomi-
nally criminal or civil. And punishment simply referred to
the law's coercive sanction—meaning a traditional depriva-
tion of life, liberty, or property—redressing that public
wrong. Applying this understanding would restore the
Clauses to their proper role without the problems attendant
to the more recent precedents.
A
The English common law divided all unlawful acts into
“private wrongs” and “public wrongs.” See 3 Blackstone 2.
Blackstone titled the third and fourth volumes of his Com-
mentaries, respectively, “Of Private Wrongs” and “Of Public
Wrongs.” This dichotomy was fundamental to English and
American law.
Public wrongs were injuries to the sovereign. A public
wrong was “a breach and violation of public rights and du-
ties, which affect the whole community, considered as a com-
munity.” Ibid. Because a public wrong injured the com-
munity, it was deemed an injury to the sovereign in his
sovereign capacity. 4 Blackstone 2–7. It followed that an
action to redress a public wrong was brought on behalf of
the sovereign, not the victim. J. Locke, Second Treatise of
Civil Government 7 (J. Gough ed. 1948) (Locke). In concrete
terms, actions for public wrongs were brought in England
on behalf of the King, and here on behalf of the State or the
United States—typically by public prosecutors, rather than
privately by the victims. The sovereign “is supposed by the
law to be the person injured by every infraction of the public
rights belonging to that community, and is therefore in all
cases the proper prosecutor for every public offense.” 4
Blackstone 2; see also Robertson v. United States ex rel. Wat-
son, 560 U. S. 272, 279 (2010) (Roberts, C. J., dissenting) (a
criminal prosecution is an action “on behalf of the sovereign,

Page Proof Pending Publication
180 EL
LINGBURG v. UNITED STATES
Thomas, J., concurring
seeking to vindicate a public wrong ”). Public wrongs in-
cluded
all wrongs deemed injuries to the sovereign and re-
dressable by the sovereign, from murder to mayhem to pub-
lic nuisance. 4 Blackstone 5–6.
Private wrongs, in contrast, were injuries to individuals in
their private capacity. An action inficted a private wrong
insofar as it infringed “the private or civil rights belonging
to individuals, considered as individuals.” 3 Blackstone 2.
For example, a dispute with a fellow citizen over a land claim
or a failure to pay a debt concerned a private wrong. “[I]f
I detain a feld from another man, to which the law has given
him a right, this is a civil injury, and not a crime,” Blackstone
explained, because “only the right of an individual is con-
cerned, and it is immaterial to the public, which of us is in
possession of the land.” 4 Blackstone 5. Private wrongs
were redressable by the individuals whose rights were vio-
lated, not the sovereign.
A single act could be both a private wrong and a public
wrong. Id., at 5–6. “The same acts will generally consti-
tute a private injury, as well as a public crime. A public
punishment is inficted on the part of the state, to restrain
such conduct, and the party injured is entitled to a compen-
sation for the injury he has sustained.” 2 Z. Swift, System
of the Laws of the State of Connecticut 292 (1796); accord,
ibid. (“If one man assaults and beats another, he is punished
by a fne for disturbing the peace, and compelled to compen-
sate in money the person he has abused and wounded”).
The private suit brought by the individual vindicated the
private wrong, and the public prosecution brought by the
sovereign vindicated the public wrong. This overlap carries
forward to modern law: Theft can be redressed through a
tort suit brought by the victim and a criminal prosecution
brought by the State.
When the opinions in Calder spoke of “crimes,” they
meant public wrongs. The prevailing defnition of a “crime”
around the time of Calder was “a violation of law to the

Page Proof Pending Publication
Cite
as: 607 U. S. 163 (2026)
181
Thomas, J., concurring
injury of the public, a public offense.” N. Webster, A Com-
pendi
ous Dictionary of the English Language 72 (1806)
(Webster). Blackstone used the terms “crimes” and “pub-
lic wrongs” interchangeably. See 4 Blackstone 1–5. His
fourth volume covered “public wrongs, or crimes and mis-
demeanors.” Id., at 1.
1
The “common-law conception of
crime,” this Court has long recognized, was an “offense
against the sovereignty of the government.” Heath v. Ala-
bama, 474 U. S. 82, 88 (1985). So Calder's limitation of the
Ex Post Facto Clauses to crimes would have originally been
understood to include all public wrongs.
Calder itself confrms this understanding. Justice Chase
distinguished crimes not with nominally civil laws enforced
by the sovereign, but with laws that merely “secure[d] pri-
vate rights,” like contract laws. 3 Dall., at 390. Justices
Chase and Iredell used “crime” interchangeably with “of-
fence.” Ibid.; id., at 400 (opinion of Iredell, J.). Justices
Chase and Paterson said that their defnitions tracked Black-
stone's. See id., at 391 (opinion of Chase, J.) (“Sir William
Blackstone, in his commentaries, considers an ex post facto
law precisely in the same light I have done”); id., at 396
(opinion of Paterson, J.) (similar). Justice Iredell explained
in a contemporaneous opinion that the Ex Post Facto Clauses
could be limited to crimes because “there . . . is little reason
to apprehend a legislative interference for the sake of un-
justly transferring property from one man to another.”
Minge v. Gilmour, 17 F. Cas. 440, 443, No. 9,631 (CC NC
1798). And, of course, the occasion for Calder's distinction
was a probate dispute, where only private rights are
implicated.
2
1
Crimes and misdemeanors were for these purposes “synonymous.” 4
Blackstone 5.
2
Professor Caleb Nelson has read Calder along the same lines: “By
`criminal' laws,” he has explained, “Justice Iredell and his colleagues may
simply have meant laws that defned offenses and imposed punishments
for them,” even if “the penalties that they prescribed could be collected

Page Proof Pending Publication
182 EL
LINGBURG v. UNITED STATES
Thomas, J., concurring
In turn, when Calder referred to criminal “punishment,”
it
simply meant whatever coercive sanction—or deprivation
of life, liberty, or property—the law imposed as redress for
committing those crimes. Blackstone understood punish-
ment to include the traditional “coercive penalties” that the
law imposed as redress for a public wrong. See 4 Black-
stone 7–8. Dictionaries defned “punishment” broadly as
“[a]ny infiction or pain imposed in vengeance of a crime,” S.
Johnson, A Dictionary of the English Language (6th ed.
1785), or “any thing inficted for a crime,” Webster 241. The
power of punishing an offender was the power to harm the
offender in a way that was otherwise forbidden: “[N]o one
ought to harm another in his life, health, liberty, or posses-
sions”—“unless it be to do justice on an offender.” Locke 5
(emphasis added); see also id., at 6 (describing when “one
man may lawfully do harm to another, which is that we call
punishment”).
The Ex Post Facto Clauses therefore prohibit retroactive
laws that impose coercive penalties for public wrongs. Con-
trary to the modern framework, it does not matter what the
legislature labels the law, where it places the law, which
agency it vests enforcement with, what its stated goals were,
whether it provides safeguards for the accused, whether it
requires a showing of scienter, or whether the conduct to
which it applies is already a crime. Contra, Smith, 538
U. S., at 96; Martinez-Mendoza, 372 U. S., at 168–169. What
matters is what the law does. If it punishes a public
wrong—or an injury to the sovereign in its sovereign capac-
ity—then it is subject to the Ex Post Facto Clauses. If it
changes merely private relations, such as the rules of con-
tract or property or probate, then it is not.
Whether a law is subject to the Ex Post Facto Clauses will
therefore typically depend on how it is enforced. If it is
enforced on behalf of the sovereign to redress a sovereign
through civil suits.” C. Nelson, Originalism and Interpretive Conven-
tions, 70 U. Chi. L. Rev. 519, 582, n. 255 (2003).

Page Proof Pending Publication
Cite
as: 607 U. S. 163 (2026)
183
Thomas, J., concurring
injury, then it is subject to the Clauses. See 4 Blackstone
2.
If instead it is enforced by a private person to vindicate
his own private rights, then it is not. In this case, for exam-
ple, the law was subject to the Ex Post Facto Clauses be-
cause it was enforced against Ellingburg by the United
States, not by the First Union National Bank, whose private
rights Ellingburg violated when he robbed it. And the
United States's action was redressing a sovereign injury to
the community as a whole, not a private wrong. Cf. 3 Black-
stone 257 (explaining that the King can bring “common law
actions” to redress private wrongs to himself, such as an “ac-
tion for trespass for taking away his goods”).
3
Calder's understanding secures Ex Post Facto Clause pro-
tection in a wide range of contexts involving nominally civil
laws. Those contexts include civil proceedings seeking fnes
for public offenses. See, e. g., Army Corps of Engineers v.
Hawkes Co., 578 U. S. 590, 600 (2016) (describing “civil penal-
ties of up to $37,500 for each day [the challenger] violated
the Act”). They include enforcement proceedings brought
by administrative agencies. See Axon Enterprise, Inc. v.
FTC, 598 U. S. 175, 196 (2023) (Thomas, J., concurring) (de-
scribing the 20th-century rise of such proceedings). And
they include municipal sanctions like speeding tickets. “If
one were to commit a minor traffc offense at a time the of-
fense was to be punished by a $25 fne, and the government
were to then amend the statute and impose a fne of
$1,000,000 dollars, it would be nonsensical to treat that fne
as non-punitive simply because the offense was processed
civilly.” Brief for Professor Beth Colgan as Amicus Curiae
8, n. 3. Because all of these offenses impose punishments
3
Sovereigns sometimes delegated enforcement of public wrongs to pri-
vate persons. But, unlike in actions redressing private wrongs, the pri-
vate persons in those circumstances acted “on behalf of the sovereign” and
were subject to the sovereign's control. Robertson v. United States
ex rel. Watson, 560 U. S. 272, 279 (2010) (Roberts, C. J., dissenting); see
ibid. (collecting authorities).

Page Proof Pending Publication
184 EL
LINGBURG v. UNITED STATES
Thomas, J., concurring
for public wrongs, Calder would treat them all as subject to
the
Ex Post Facto Clauses.
B
This understanding of Calder long prevailed. Courts took
the position that “the relevant line for the Clause's scope”
lay between “punishments imposed in response to public of-
fenses whether prosecuted criminally or civilly on the one
hand, and purely private disputes on the other.” Brief for
Professor Beth Colgan as Amicus Curiae 5. This Court
confrmed, contrary to the modern precedents, that “the ex
post facto effect of a law cannot be evaded by giving a civil
form to that which is essentially criminal.” Burgess v.
Salmon, 97 U. S. 381, 385 (1878). The infuential commenta-
tor Thomas Cooley likewise explained that a “law will be
ex post facto in the constitutional sense” if it “subject[s] an
individual to a pecuniary penalty” or “deprives a party of
any valuable right” retroactively, even if “it does not in
terms declare the acts to which the penalty is attached crimi-
nal.” Constitutional Limitations 321 (6th ed. 1890).
In a similar context, when a State sought nominally civil
fnes against a company for doing business without proper
forms, this Court explained that the statute “was in the
strictest sense a penal statute.” Wisconsin v. Pelican Ins.
Co., 127 U. S. 265, 299 (1888). The Court held that the nomi-
nally civil statute was penal because “[t]he cause of action
was not any private injury, but solely the offence committed
against the State by violating her law.” Ibid. Likewise,
“[t]he prosecution was in the name of the State.” Ibid.
The Court explained that “[t]he real nature of the case is not
affected by the forms provided by the law of the State”;
these forms were “immaterial” to whether the law was crim-
inal or penal in a constitutional sense. Ibid. Thus, the
Court concluded, “[i]n whatever form the State pursues her
right to punish the offence against her sovereignty,” an ac-
tion is criminal or penal in substance whenever it seeks to

Page Proof Pending Publication
Cite
as: 607 U. S. 163 (2026)
185
Thomas, J., concurring
“compe[l] the offender to pay a pecuniary fne by way of pun-
ishment
for the offence.” Ibid.
Other decisions followed the same reasoning. For exam-
ple, when the United States brought a nominally civil action
for damages based on a public offense, the defendants were
entitled to other criminal-procedure protections because the
action “was in its nature and essence, though not its form, a
penal or criminal action.” United States v. Poyllon, 27
F. Cas. 608, 611, No. 16,081 (NY 1812). And when a statute
forbade “any fne or forfeiture under any penal statute,” the
statute applied to a nominally civil action for debt on behalf
of the sovereign against someone who committed a public
offense. Adams v. Woods, 2 Cranch 336, 337, 340–341 (1805)
(emphasis deleted). Chief Justice Marshall explained that
the category of penal laws referred “not to any particular
mode of proceeding, but generally to any prosecution, trial,
or punishment for the offence.” Ibid.
This understanding of Calder harmonizes the Ex Post
Facto Clauses with their historical purpose. It means that
the Clauses ensure fair notice against government enforce-
ment actions, regardless of whether the legislature labels
them civil. As Justice Gorsuch has explained, today's
“civil” laws “regularly impose penalties far more severe than
those found in many criminal statutes.” Sessions v. Di-
maya, 584 U. S. 148, 184 (2018) (opinion concurring in part
and concurring in judgment). “Today's `civil' penalties in-
clude confscatory rather than compensatory fnes, forfeiture
provisions that allow homes to be taken, remedies that strip
persons of their professional licenses and livelihoods, and the
power to commit persons against their will indefnitely.' ”
Ibid. This understanding also helps reconcile Calder with
its critics, many of whom have focused on the injustice of
allowing the government to retroactively impose sanctions
for public offenses that it deems civil. See, e. g., Brief for
Professor Wayne A. Logan as Amicus Curiae 31–32; cf. D.
Troy, Retroactive Legislation 49–55 (1998).

Page Proof Pending Publication
186 EL
LINGBURG v. UNITED STATES
Thomas, J., concurring
This understanding of Calder also simplifes the law.
Rather
than making the Ex Post Facto Clauses' application
depend on twelve factors with indeterminate weights, this
understanding makes it depend on a simple legal inquiry
with a long pedigree in our legal tradition—whether the law
imposes a coercive penalty to redress a public wrong. See,
e. g., Pelican Ins. Co., 127 U. S., at 299; Robertson, 560 U. S.,
at 278–279 (Roberts, C. J., dissenting). In most cases, that
inquiry will just come down to who enforces the law, the
sovereign or the injured private party.
***
In Collins v. Youngblood, this Court explained that any
“departure from Calder's explanation of the original under-
standing of the Ex Post Facto Clause[s] was . . . unjustifed.”
497 U. S., at 49. The Court understood itself to be bound by
what Calder meant rather than by its intervening precedents
that misunderstood Calder. 497 U. S., at 49–50. Today,
this Court's precedents concerning the scope of laws impos-
ing criminal punishment have departed from Calder's under-
standing of that category. In a future case, the Court should
consider returning to Calder's understanding.

Page Proof Pending Publication
Reporter’s Note
The attached opinion has been revised to refect the usual publication
and citation style of the United States Reports. The revised pagination
makes available the offcial United States Reports citation in advance of
publication. The syllabus has been prepared by the Reporter of Decisions
for the convenience of the reader and constitutes no part of the opinion of
the Court. A list of counsel who argued or fled briefs in this case, and
who were members of the bar of this Court at the time this case was
argued, has been inserted following the syllabus. Other revisions may
include adjustments to formatting, captions, citation form, and any errant
punctuation. The following additional edits were made:
p. 163, paragraph 1, line 4: “MRVA” is changed to “MVRA”