McGirt v. Oklahoma

(Slip Opinion) OCTOBER TERM, 2019 1

Syllabus

NOTE: Where it is feasible, a syllabus (headnote) will be released, as is
being done in connection with this case, at the time the opinion is issued.
The syllabus constitutes no part of the opinion of the Court but has been
prepared by the Reporter of Decisions for the convenience of the reader.
See United States v. Detroit Timber & Lumber Co., 200 U. S. 321, 337.

SUPREME COURT OF THE UNITED STATES

Syllabus

McGIRT v. OKLAHOMA

CERTIORARI TO THE COURT OF CRIMINAL APPEALS OF
OKLAHOMA

No. 18–9526. Argued May 11, 2020—Decided July 9, 2020
The Major Crimes Act (MCA) provides that, within “the Indian country,”
“[a]ny Indian who commits” certain enumerated offenses “shall be sub-
ject to the same law and penalties as all other persons committing any
of [those] offenses, within the exclusive jurisdiction of the United
States.” 18 U. S. C. §1153(a). “Indian country” includes “all land
within the limits of any Indian reservation under the jurisdiction of
the United States Government.” §1151. Petitioner Jimcy McGirt was
convicted by an Oklahoma state court of three serious sexual offenses.
He unsuccessfully argued in state postconviction proceedings that the
State lacked jurisdiction to prosecute him because he is an enrolled
member of the Seminole Nation and his crimes took place on the Creek
Reservation. He seeks a new trial, which, he contends, must take place
in federal court.
Held: For MCA purposes, land reserved for the Creek Nation since the
19th century remains “Indian country.” Pp. 3–42.
(a) Congress established a reservation for the Creek Nation. An
1833 Treaty fixed borders for a “permanent home to the whole Creek
Nation of Indians,” 7 Stat. 418, and promised that the United States
would “grant a patent, in fee simple, to the Creek nation of Indians for
the [assigned] land” to continue “so long as they shall exist as a nation,
and continue to occupy the country hereby assigned to them,” id., at
419. The patent formally issued in 1852.
Though the early treaties did not refer to the Creek lands as a “res-
ervation,” similar language in treaties from the same era has been held
sufficient to create a reservation, see, e.g., Menominee Tribe v. United
States, 391 U. S. 404, 405, and later Acts of Congress—referring to the
“Creek reservation”—leave no room for doubt, see, e.g., 17 Stat. 626.
In addition, an 1856 Treaty promised that “no portion” of Creek lands
2 MCGIRT v. OKLAHOMA

Syllabus

“would ever be embraced or included within, or annexed to, any Terri-
tory or State,” 11 Stat. 700, and that the Creeks would have the “un-
restricted right of self-government,” with “full jurisdiction” over en-
rolled Tribe members and their property, id., at 704. Pp. 3–6.
(b) Congress has since broken more than a few promises to the Tribe.
Nevertheless, the Creek Reservation persists today. Pp. 6–28.
(1) Once a federal reservation is established, only Congress can
diminish or disestablish it. Doing so requires a clear expression of con-
gressional intent. Pp. 6–8.
(2) Oklahoma claims that Congress ended the Creek Reservation
during the so-called “allotment era”—a period when Congress sought
to pressure many tribes to abandon their communal lifestyles and par-
cel their lands into smaller lots owned by individual tribal members.
Missing from the allotment-era agreement with the Creek, see 31 Stat.
862–864, however, is any statute evincing anything like the “present
and total surrender of all tribal interests” in the affected lands. And
this Court has already rejected the argument that allotments automat-
ically ended reservations. Pp. 8–13.
(3) Oklahoma points to other ways Congress intruded on the
Creeks’ promised right to self-governance during the allotment era, in-
cluding abolishing the Creeks’ tribal courts, 30 Stat. 504–505, and re-
quiring Presidential approval for certain tribal ordinances, 31 Stat.
872. But these laws fall short of eliminating all tribal interest in the
contested lands. Pp. 13–17.
(4) Oklahoma ultimately claims that historical practice and de-
mographics are enough by themselves to prove disestablishment. This
Court has consulted contemporaneous usages, customs, and practices
to the extent they shed light on the meaning of ambiguous statutory
terms, but Oklahoma points to no ambiguous language in any of the
relevant statutes that could plausibly be read as an act of cession.
Such extratextual considerations are of “ ‘limited interpretive value,’ ”
Nebraska v. Parker, 577 U. S. 481, ___, and the “least compelling” form
of evidence, South Dakota v. Yankton Sioux Tribe, 522 U. S. 329, 356.
In the end, Oklahoma resorts to the State’s long historical practice of
prosecuting Indians in state court for serious crimes on the contested
lands, various statements made during the allotment era, and the
speedy and persistent movement of white settlers into the area. But
these supply little help with the law’s meaning and much potential for
mischief. Pp. 17–28.
(c) In the alternative, Oklahoma contends that Congress never es-
tablished a reservation but instead created a “dependent Indian com-
munity.” To hold that the Creek never had a reservation would require
willful blindness to the statutory language and a belief that the land
 Cite as: 591 U. S. ____ (2020) 3

Syllabus

patent the Creek received somehow made their tribal sovereignty eas-
ier to divest. Congress established a reservation, not a dependent In-
dian community, for the Creek Nation. Pp. 28–31.
(d) Even assuming that the Creek land is a reservation, Oklahoma
argues that the MCA has never applied in eastern Oklahoma. It
claims that the Oklahoma Enabling Act, which transferred all non-
federal cases pending in the territorial courts to Oklahoma’s state
courts, made the State’s courts the successors to the federal territorial
courts’ sweeping authority to try Indians for crimes committed on res-
ervations. That argument, however, rests on state prosecutorial prac-
tices that defy the MCA, rather than on the law’s plain terms. Pp. 32–
36.
(e) Finally, Oklahoma warns of the potential consequences that will
follow a ruling against it, such as unsettling an untold number of con-
victions and frustrating the State’s ability to prosecute crimes in the
future. This Court is aware of the potential for cost and conflict around
jurisdictional boundaries. But Oklahoma and its tribes have proven
time and again that they can work successfully together as partners,
and Congress remains free to supplement its statutory directions
about the lands in question at any time. Pp. 36–42.
Reversed.

GORSUCH, J., delivered the opinion of the Court, in which GINSBURG,
BREYER, SOTOMAYOR, and KAGAN, JJ., joined. ROBERTS, C. J., filed a dis-
senting opinion, in which ALITO and KAVANAUGH, JJ., joined, and in
which THOMAS, J., joined, except as to footnote 9. THOMAS, J., filed a dis-
senting opinion.
 Cite as: 591 U. S. ____ (2020) 1

Opinion of the Court

NOTICE: This opinion is subject to formal revision before publication in the
preliminary print of the United States Reports. Readers are requested to
notify the Reporter of Decisions, Supreme Court of the United States, Wash-
ington, D. C. 20543, of any typographical or other formal errors, in order that
corrections may be made before the preliminary print goes to press.

SUPREME COURT OF THE UNITED STATES
_________________

No. 18–9526
_________________

JIMCY MCGIRT, PETITIONER v. OKLAHOMA
ON WRIT OF CERTIORARI TO THE COURT OF CRIMINAL
APPEALS OF OKLAHOMA
[July 9, 2020]

JUSTICE GORSUCH delivered the opinion of the Court.
On the far end of the Trail of Tears was a promise. Forced
to leave their ancestral lands in Georgia and Alabama, the
Creek Nation received assurances that their new lands in
the West would be secure forever. In exchange for ceding
“all their land, East of the Mississippi river,” the U. S. gov-
ernment agreed by treaty that “[t]he Creek country west of
the Mississippi shall be solemnly guarantied to the Creek
Indians.” Treaty With the Creeks, Arts. I, XIV, Mar. 24,
1832, 7 Stat. 366, 368 (1832 Treaty). Both parties settled
on boundary lines for a new and “permanent home to the
whole Creek nation,” located in what is now Oklahoma.
Treaty With the Creeks, preamble, Feb. 14, 1833, 7 Stat.
418 (1833 Treaty). The government further promised that
“[no] State or Territory [shall] ever have a right to pass laws
for the government of such Indians, but they shall be al-
lowed to govern themselves.” 1832 Treaty, Art. XIV, 7 Stat.
368.
Today we are asked whether the land these treaties
promised remains an Indian reservation for purposes of fed-
eral criminal law. Because Congress has not said other-
wise, we hold the government to its word.
2 MCGIRT v. OKLAHOMA

Opinion of the Court

I
At one level, the question before us concerns Jimcy
McGirt. Years ago, an Oklahoma state court convicted him
of three serious sexual offenses. Since then, he has argued
in postconviction proceedings that the State lacked jurisdic-
tion to prosecute him because he is an enrolled member of
the Seminole Nation of Oklahoma and his crimes took place
on the Creek Reservation. A new trial for his conduct, he
has contended, must take place in federal court. The Okla-
homa state courts hearing Mr. McGirt’s arguments rejected
them, so he now brings them here.
Mr. McGirt’s appeal rests on the federal Major Crimes
Act (MCA). The statute provides that, within “the Indian
country,” “[a]ny Indian who commits” certain enumerated
offenses “against the person or property of another Indian
or any other person” “shall be subject to the same law and
penalties as all other persons committing any of the above
offenses, within the exclusive jurisdiction of the United
States.” 18 U. S. C. §1153(a). By subjecting Indians to fed-
eral trials for crimes committed on tribal lands, Congress
may have breached its promises to tribes like the Creek
that they would be free to govern themselves. But this par-
ticular incursion has its limits—applying only to certain
enumerated crimes and allowing only the federal govern-
ment to try Indians. State courts generally have no juris-
diction to try Indians for conduct committed in “Indian
country.” Negonsott v. Samuels, 507 U. S. 99, 102–103
(1993).
The key question Mr. McGirt faces concerns that last
qualification: Did he commit his crimes in Indian country?
A neighboring provision of the MCA defines the term to in-
clude, among other things, “all land within the limits of any
Indian reservation under the jurisdiction of the United
States Government, notwithstanding the issuance of any
patent, and, including rights-of-way running through the
reservation.” §1151(a). Mr. McGirt submits he can satisfy
 Cite as: 591 U. S. ____ (2020) 3

Opinion of the Court

this condition because he committed his crimes on land re-
served for the Creek since the 19th century.
The Creek Nation has joined Mr. McGirt as amicus cu-
riae. Not because the Tribe is interested in shielding Mr.
McGirt from responsibility for his crimes. Instead, the
Creek Nation participates because Mr. McGirt’s personal
interests wind up implicating the Tribe’s. No one disputes
that Mr. McGirt’s crimes were committed on lands de-
scribed as the Creek Reservation in an 1866 treaty and fed-
eral statute. But, in seeking to defend the state-court judg-
ment below, Oklahoma has put aside whatever procedural
defenses it might have and asked us to confirm that the
land once given to the Creeks is no longer a reservation to-
day.
At another level, then, Mr. McGirt’s case winds up as a
contest between State and Tribe. The scope of their dispute
is limited; nothing we might say today could unsettle Okla-
homa’s authority to try non-Indians for crimes against non-
Indians on the lands in question. See United States v.
McBratney, 104 U. S. 621, 624 (1882). Still, the stakes are
not insignificant. If Mr. McGirt and the Tribe are right, the
State has no right to prosecute Indians for crimes commit-
ted in a portion of Northeastern Oklahoma that includes
most of the city of Tulsa. Responsibility to try these matters
would fall instead to the federal government and Tribe. Re-
cently, the question has taken on more salience too. While
Oklahoma state courts have rejected any suggestion that
the lands in question remain a reservation, the Tenth Cir-
cuit has reached the opposite conclusion. Murphy v. Royal,
875 F. 3d 896, 907–909, 966 (2017). We granted certiorari
to settle the question. 589 U. S. ___ (2019).
II
Start with what should be obvious: Congress established
a reservation for the Creeks. In a series of treaties, Con-
4 MCGIRT v. OKLAHOMA

Opinion of the Court

gress not only “solemnly guarantied” the land but also “es-
tablish[ed] boundary lines which will secure a country and
permanent home to the whole Creek Nation of Indians.”
1832 Treaty, Art. XIV, 7 Stat. 368; 1833 Treaty, preamble,
7 Stat. 418. The government’s promises weren’t made gra-
tuitously. Rather, the 1832 Treaty acknowledged that
“[t]he United States are desirous that the Creeks should re-
move to the country west of the Mississippi” and, in service
of that goal, required the Creeks to cede all lands in the
East. Arts. I, XII, 7 Stat. 366, 367. Nor were the govern-
ment’s promises meant to be delusory. Congress twice as-
sured the Creeks that “[the] Treaty shall be obligatory on
the contracting parties, as soon as the same shall be ratified
by the United States.” 1832 Treaty, Art. XV, id., at 368; see
1833 Treaty, Art. IX, 7 Stat. 420 (“agreement shall be bind-
ing and obligatory” upon ratification). Both treaties were
duly ratified and enacted as law.
Because the Tribe’s move west was ostensibly voluntary,
Congress held out another assurance as well. In the statute
that precipitated these negotiations, Congress authorized
the President “to assure the tribe . . . that the United States
will forever secure and guaranty to them . . . the country so
exchanged with them.” Indian Removal Act of 1830, §3, 4
Stat. 412. “[A]nd if they prefer it,” the bill continued, “the
United States will cause a patent or grant to be made and
executed to them for the same; Provided always, that such
lands shall revert to the United States, if the Indians be-
come extinct, or abandon the same.” Ibid. If agreeable to
all sides, a tribe would not only enjoy the government’s sol-
emn treaty promises; it would hold legal title to its lands.
It was an offer the Creek accepted. The 1833 Treaty fixed
borders for what was to be a “permanent home to the whole
Creek nation of Indians.” 1833 Treaty, preamble, 7 Stat.
418. It also established that the “United States will grant
a patent, in fee simple, to the Creek nation of Indians for
the land assigned said nation by this treaty.” Art. III, id.,
 Cite as: 591 U. S. ____ (2020) 5

Opinion of the Court

at 419. That grant came with the caveat that “the right
thus guaranteed by the United States shall be continued to
said tribe of Indians, so long as they shall exist as a nation,
and continue to occupy the country hereby assigned to
them.” Ibid. The promised patent formally issued in 1852.
See Woodward v. De Graffenried, 238 U. S. 284, 293–294
(1915).
These early treaties did not refer to the Creek lands as a
“reservation”—perhaps because that word had not yet ac-
quired such distinctive significance in federal Indian law.
But we have found similar language in treaties from the
same era sufficient to create a reservation. See Menominee
Tribe v. United States, 391 U. S. 404, 405 (1968) (grant of
land “ ‘for a home, to be held as Indian lands are held,’ ” es-
tablished a reservation). And later Acts of Congress left no
room for doubt. In 1866, the United States entered yet an-
other treaty with the Creek Nation. This agreement re-
duced the size of the land set aside for the Creek, compen-
sating the Tribe at a price of 30 cents an acre. Treaty
Between the United States and the Creek Nation of Indi-
ans, Art. III, June 14, 1866, 14 Stat. 786. But Congress ex-
plicitly restated its commitment that the remaining land
would “be forever set apart as a home for said Creek Na-
tion,” which it now referred to as “the reduced Creek reser-
vation.” Arts. III, IX, id., at 786, 788.1 Throughout the late
——————
1 The dissent by THE CHIEF JUSTICE (hereinafter the dissent) suggests

that the Creek’s intervening alliance with the Confederacy “ ‘unsettled’ ”
and “ ‘forfeit[ed]’ ” the longstanding promises of the United States. Post,
at 3. But the Treaty of 1866 put an end to any Civil War hostility, prom-
ising mutual amnesty, “perpetual peace and friendship,” and guarantee-
ing the Tribe the “quiet possession of their country.” Art. I, 14 Stat. 786.
Though this treaty expressly reduced the size of the Creek Reservation,
the Creek were compensated for the lost territory, and otherwise “re-
tained” their unceded portion. Art. III, ibid. Contrary to the dissent’s
implication, nothing in the Treaty of 1866 purported to repeal prior
treaty promises. Cf. Art. XII, id., at 790 (the United States expressly “re-
affirms and reassumes all obligations of treaty stipulations with the
6 MCGIRT v. OKLAHOMA

Opinion of the Court

19th century, many other federal laws also expressly re-
ferred to the Creek Reservation. See, e.g., Treaty Between
United States and Cherokee Nation of Indians, Art. IV, July
19, 1866, 14 Stat. 800 (“Creek reservation”); Act of Mar. 3,
1873, ch. 322, 17 Stat. 626; (multiple references to the
“Creek reservation” and “Creek India[n] Reservation”); 11
Cong. Rec. 2351 (1881) (discussing “the dividing line be-
tween the Creek reservation and their ceded lands”); Act of
Feb. 13, 1891, 26 Stat. 750 (describing a cession by refer-
encing the “West boundary line of the Creek Reservation”).
There is a final set of assurances that bear mention, too.
In the Treaty of 1856, Congress promised that “no portion”
of the Creek Reservation “shall ever be embraced or in-
cluded within, or annexed to, any Territory or State.” Art.
IV, 11 Stat. 700. And within their lands, with exceptions,
the Creeks were to be “secured in the unrestricted right of
self-government,” with “full jurisdiction” over enrolled
Tribe members and their property. Art. XV, id., at 704. So
the Creek were promised not only a “permanent home” that
would be “forever set apart”; they were also assured a right
to self-government on lands that would lie outside both the
legal jurisdiction and geographic boundaries of any State.
Under any definition, this was a reservation.
III
A
While there can be no question that Congress established
a reservation for the Creek Nation, it’s equally clear that
Congress has since broken more than a few of its promises
to the Tribe. Not least, the land described in the parties’
treaties, once undivided and held by the Tribe, is now frac-
tured into pieces. While these pieces were initially distrib-
uted to Tribe members, many were sold and now belong to
persons unaffiliated with the Nation. So in what sense, if

——————
Creek nation entered into before” the Civil War).
 Cite as: 591 U. S. ____ (2020) 7

Opinion of the Court

any, can we say that the Creek Reservation persists today?
To determine whether a tribe continues to hold a reser-
vation, there is only one place we may look: the Acts of Con-
gress. This Court long ago held that the Legislature wields
significant constitutional authority when it comes to tribal
relations, possessing even the authority to breach its own
promises and treaties. Lone Wolf v. Hitchcock, 187 U. S.
553, 566–568 (1903). But that power, this Court has cau-
tioned, belongs to Congress alone. Nor will this Court
lightly infer such a breach once Congress has established a
reservation. Solem v. Bartlett, 465 U. S. 463, 470 (1984).
Under our Constitution, States have no authority to re-
duce federal reservations lying within their borders. Just
imagine if they did. A State could encroach on the tribal
boundaries or legal rights Congress provided, and, with
enough time and patience, nullify the promises made in the
name of the United States. That would be at odds with the
Constitution, which entrusts Congress with the authority
to regulate commerce with Native Americans, and directs
that federal treaties and statutes are the “supreme Law of
the Land.” Art. I, §8; Art. VI, cl. 2. It would also leave tribal
rights in the hands of the very neighbors who might be least
inclined to respect them.
Likewise, courts have no proper role in the adjustment of
reservation borders. Mustering the broad social consensus
required to pass new legislation is a deliberately hard busi-
ness under our Constitution. Faced with this daunting
task, Congress sometimes might wish an inconvenient res-
ervation would simply disappear. Short of that, legislators
might seek to pass laws that tiptoe to the edge of disestab-
lishment and hope that judges—facing no possibility of elec-
toral consequences themselves—will deliver the final push.
But wishes don’t make for laws, and saving the political
branches the embarrassment of disestablishing a reserva-
tion is not one of our constitutionally assigned prerogatives.
“[O]nly Congress can divest a reservation of its land and
8 MCGIRT v. OKLAHOMA

Opinion of the Court

diminish its boundaries.” Solem, 465 U. S., at 470. So it’s
no matter how many other promises to a tribe the federal
government has already broken. If Congress wishes to
break the promise of a reservation, it must say so.
History shows that Congress knows how to withdraw a
reservation when it can muster the will. Sometimes, legis-
lation has provided an “[e]xplicit reference to cession” or an
“unconditional commitment . . . to compensate the Indian
tribe for its opened land.” Ibid. Other times, Congress has
directed that tribal lands shall be “ ‘restored to the public
domain.’ ” Hagen v. Utah, 510 U. S. 399, 412 (1994) (em-
phasis deleted). Likewise, Congress might speak of a res-
ervation as being “ ‘discontinued,’ ” “ ‘abolished,’ ” or “ ‘va-
cated.’ ” Mattz v. Arnett, 412 U. S. 481, 504, n. 22 (1973).
Disestablishment has “never required any particular form
of words,” Hagen, 510 U. S., at 411. But it does require that
Congress clearly express its intent to do so, “[c]ommon[ly
with an] ‘[e]xplicit reference to cession or other language
evidencing the present and total surrender of all tribal in-
terests.’ ” Nebraska v. Parker, 577 U. S. 481, ___–___ (2016)
(slip op., at 6).
B
In an effort to show Congress has done just that with the
Creek Reservation, Oklahoma points to events during the
so-called “allotment era.” Starting in the 1880s, Congress
sought to pressure many tribes to abandon their communal
lifestyles and parcel their lands into smaller lots owned by
individual tribe members. See 1 F. Cohen, Handbook of
Federal Indian Law §1.04 (2012) (Cohen), discussing Gen-
eral Allotment Act of 1887, ch. 119, 24 Stat. 388. Some al-
lotment advocates hoped that the policy would create a
class of assimilated, landowning, agrarian Native Ameri-
cans. See Cohen §1.04; F. Hoxie, A Final Promise: The
Campaign To Assimilate 18–19 (2001). Others may have
hoped that, with lands in individual hands and (eventually)
 Cite as: 591 U. S. ____ (2020) 9

Opinion of the Court

freely alienable, white settlers would have more space of
their own. See id., at 14–15; cf. General Allotment Act of
1887, §5, 24 Stat. 389–390.
The Creek were hardly exempt from the pressures of the
allotment era. In 1893, Congress charged the Dawes Com-
mission with negotiating changes to the Creek Reservation.
Congress identified two goals: Either persuade the Creek
to cede territory to the United States, as it had before, or
agree to allot its lands to Tribe members. Act of Mar. 3,
1893, ch. 209, §16, 27 Stat. 645–646. A year later, the Com-
mission reported back that the Tribe “would not, under any
circumstances, agree to cede any portion of their lands.” S.
Misc. Doc. No. 24, 53d Cong., 3d Sess., 7 (1894). At that
time, before this Court’s decision in Lone Wolf, Congress
may not have been entirely sure of its power to terminate
an established reservation unilaterally. Perhaps for that
reason, perhaps for others, the Commission and Congress
took this report seriously and turned their attention to al-
lotment rather than cession.2
The Commission’s work culminated in an allotment
agreement with the Tribe in 1901. Creek Allotment Agree-
ment, ch. 676, 31 Stat. 861. With exceptions for certain pre-
existing town sites and other special matters, the Agree-
ment established procedures for allotting 160-acre parcels
to individual Tribe members who could not sell, transfer, or
otherwise encumber their allotments for a number of years.
§§3, 7, id., at 862–864 (5 years for any portion, 21 years for
the designated “homestead” portion). Tribe members were
given deeds for their parcels that “convey[ed] to [them] all
right, title, and interest of the Creek Nation.” §23, id., at

——————
2 The dissent stresses, repeatedly, that the Dawes Commission was

charged with seeking to extinguish the reservation. Post, at 18, 24. Yet,
the dissent fails to mention the Commission’s various reports acknowl-
edging that those efforts were unsuccessful precisely because the Creek
refused to cede their lands.
10 MCGIRT v. OKLAHOMA

Opinion of the Court

867–868. In 1908, Congress relaxed these alienation re-
strictions in some ways, and even allowed the Secretary of
the Interior to waive them. Act of May 27, 1908, ch. 199,
§1, 35 Stat. 312. One way or the other, individual Tribe
members were eventually free to sell their land to Indians
and non-Indians alike.
Missing in all this, however, is a statute evincing any-
thing like the “present and total surrender of all tribal in-
terests” in the affected lands. Without doubt, in 1832 the
Creek “cede[d]” their original homelands east of the Missis-
sippi for a reservation promised in what is now Oklahoma.
1832 Treaty, Art. I, 7 Stat. 366. And in 1866, they “cede[d]
and convey[ed]” a portion of that reservation to the United
States. Treaty With the Creek, Art. III, 14 Stat. 786. But
because there exists no equivalent law terminating what
remained, the Creek Reservation survived allotment.
In saying this we say nothing new. For years, States have
sought to suggest that allotments automatically ended res-
ervations, and for years courts have rejected the argument.
Remember, Congress has defined “Indian country” to in-
clude “all land within the limits of any Indian reservation
. . . notwithstanding the issuance of any patent, and, includ-
ing any rights-of-way running through the reservation.” 18
U. S. C. §1151(a). So the relevant statute expressly contem-
plates private land ownership within reservation bounda-
ries. Nor under the statute’s terms does it matter whether
these individual parcels have passed hands to non-Indians.
To the contrary, this Court has explained repeatedly that
Congress does not disestablish a reservation simply by al-
lowing the transfer of individual plots, whether to Native
Americans or others. See Mattz, 412 U. S., at 497 (“[A]llot-
ment under the . . . Act is completely consistent with con-
tinued reservation status”); Seymour v. Superintendent of
Wash. State Penitentiary, 368 U. S. 351, 356–358 (1962)
(holding that allotment act “did no more than open the way
for non-Indian settlers to own land on the reservation”);
 Cite as: 591 U. S. ____ (2020) 11

Opinion of the Court

Parker, 577 U. S., at ___ (slip op., at 7) (“[T]he 1882 Act falls
into another category of surplus land Acts: those that
merely opened reservation land to settlement. . . . Such
schemes allow non-Indian settlers to own land on the res-
ervation” (internal quotation marks omitted)).
It isn’t so hard to see why. The federal government issued
its own land patents to many homesteaders throughout the
West. These patents transferred legal title and are the ba-
sis for much of the private land ownership in a number of
States today. But no one thinks any of this diminished the
United States’s claim to sovereignty over any land. To ac-
complish that would require an act of cession, the transfer
of a sovereign claim from one nation to another. 3 E. Wash-
burn, American Law of Real Property *521–*524. And
there is no reason why Congress cannot reserve land for
tribes in much the same way, allowing them to continue to
exercise governmental functions over land even if they no
longer own it communally. Indeed, such an arrangement
seems to be contemplated by §1151(a)’s plain terms. Cf.
Seymour, 368 U. S., at 357–358.3
Oklahoma reminds us that allotment was often the first
step in a plan ultimately aimed at disestablishment. As
this Court explained in Mattz, Congress’s expressed policy
at the time “was to continue the reservation system and the
trust status of Indian lands, but to allot tracts to individual
Indians for agriculture and grazing.” 412 U. S., at 496.
Then, “[w]hen all the lands had been allotted and the trust
expired, the reservation could be abolished.” Ibid. This
plan was set in motion nationally in the General Allotment
——————
3 The dissent not only fails to acknowledge these features of the statute

and our precedents. It proceeds in defiance of them, suggesting that by
moving to eliminate communal title and relaxing restrictions on aliena-
tion, “Congress destroyed the foundation of [the Creek Nation’s] sover-
eignty.” Post, at 18–19. But this Court long ago rejected the notion that
the purchase of lands by non-Indians is inconsistent with reservation
status. See Seymour, 368 U. S., at 357–358.
12 MCGIRT v. OKLAHOMA

Opinion of the Court

Act of 1887, and for the Creek specifically in 1901. No
doubt, this is why Congress at the turn of the 20th century
“believed to a man” that “the reservation system would
cease” “within a generation at most.” Solem, 465 U. S., at
468. Still, just as wishes are not laws, future plans aren’t
either. Congress may have passed allotment laws to create
the conditions for disestablishment. But to equate allot-
ment with disestablishment would confuse the first step of
a march with arrival at its destination.4
Ignoring this distinction would run roughshod over many
other statutes as well. In some cases, Congress chose not
to wait for allotment to run its course before disestablishing
a reservation. When it deemed that approach appropriate,
Congress included additional language expressly ending
reservation status. So, for example, in 1904, Congress al-
lotted reservations belonging to the Ponca and Otoe Tribes,
reservations also lying within modern-day Oklahoma, and
then provided “further, That the reservation lines of the
said . . . reservations . . . are hereby abolished.” Act of Apr.
21, 1904, §8, 33 Stat. 217–218 (emphasis deleted); see also
DeCoteau v. District County Court for Tenth Judicial Dist.,
420 U. S. 425, 439–440, n. 22 (1975) (collecting other exam-
ples). Tellingly, however, nothing like that can be found in
the nearly contemporary 1901 Creek Allotment Agreement
or the 1908 Act. That doesn’t make these laws special. Ra-
ther, in using the language that they did, these allotment
laws tracked others of the period, parceling out individual

——————
4 The dissent seemingly conflates these steps in other ways, too, by im-

plying that the passage of an allotment Act itself extinguished title. Post,
at 18–19. The reality proved more complicated. Allotment of the Creek
lands did not occur overnight, but dragged on for years, well past Okla-
homa’s statehood, until Congress finally prohibited any further allot-
ments more than 15 years later. Act of Mar. 2, 1917, 39 Stat. 986.
 Cite as: 591 U. S. ____ (2020) 13

Opinion of the Court

tracts, while saving the ultimate fate of the land’s reserva-
tion status for another day.5
C
If allotment by itself won’t work, Oklahoma seeks to
prove disestablishment by pointing to other ways Congress
intruded on the Creek’s promised right to self-governance
during the allotment era. It turns out there were many.
For example, just a few years before the 1901 Creek Allot-
ment Agreement, and perhaps in an effort to pressure the
Tribe to the negotiating table, Congress abolished the
Creeks’ tribal courts and transferred all pending civil and
criminal cases to the U. S. Courts of the Indian Territory.
Curtis Act of 1898, §28, 30 Stat. 504–505. Separately, the
Creek Allotment Agreement provided that tribal ordi-
nances “affecting the lands of the Tribe, or of individuals
after allotment, or the moneys or other property of the
Tribe, or of the citizens thereof ” would not be valid until
approved by the President of the United States. §42, 31
Stat. 872.
Plainly, these laws represented serious blows to the

——————
5 The dissent doesn’t purport to find any of the hallmarks of diminish-

ment in the Creek Allotment Agreement. Instead, the dissent tries to
excuse their absence by saying that it would have made “little sense” to
find such language in an Act transferring the Tribe’s lands to private
owners. Post, at 14. But the dissent’s account is impossible to reconcile
with history and precedent. As we have noted, plenty of allotment agree-
ments during this era included precisely the language of cession and
compensation that the dissent says it would make “little sense” to find
there. And this Court has confirmed time and again that allotment
agreements without such language do not necessarily disestablish or di-
minish the reservation at issue. See Mattz v. Arnett, 412 U. S. 481, 497
(1973); Seymour v. Superintendent of Wash. State Penitentiary, 368 U. S.
351, 358 (1962). The dissent’s only answer is to suggest that allotment
combined with other statutes limiting the Creek Nation’s governing au-
thority amounted to disestablishment—in other words that it’s the argu-
ments in the next section that really do the work.
14 MCGIRT v. OKLAHOMA

Opinion of the Court

Creek. But, just as plainly, they left the Tribe with signifi-
cant sovereign functions over the lands in question. For ex-
ample, the Creek Nation retained the power to collect taxes,
operate schools, legislate through tribal ordinances, and,
soon, oversee the federally mandated allotment process.
§§39, 40, 42, id., at 871–872; Buster v. Wright, 135 F. 947,
949–950, 953–954 (CA8 1905). And, in its own way, the
congressional incursion on tribal legislative processes only
served to prove the power: Congress would have had no
need to subject tribal legislation to Presidential review if
the Tribe lacked any authority to legislate. Grave though
they were, these congressional intrusions on pre-existing
treaty rights fell short of eliminating all tribal interests in
the land.
Much more ominously, the 1901 allotment agreement
ended by announcing that the Creek tribal government
“shall not continue” past 1906, although the agreement
quickly qualified that statement, adding the proviso “sub-
ject to such further legislation as Congress may deem
proper.” §46, 31 Stat. 872. Thus, while suggesting that the
tribal government might end in 1906, Congress also neces-
sarily understood it had not ended in 1901. All of which
was consistent with the Legislature’s general practice of
taking allotment as a first, not final, step toward disestab-
lishment and dissolution.
When 1906 finally arrived, Congress adopted the Five
Civilized Tribes Act. But instead of dissolving the tribal
government as some may have expected, Congress
“deem[ed] proper” a different course, simply cutting away
further at the Tribe’s autonomy. Congress empowered the
President to remove and replace the principal chief of the
Creek, prohibited the tribal council from meeting more than
30 days a year, and directed the Secretary of the Interior to
assume control of tribal schools. §§6, 10, 28, 34 Stat. 139–
140, 148. The Act also provided for the handling of the
 Cite as: 591 U. S. ____ (2020) 15

Opinion of the Court

Tribe’s funds, land, and legal liabilities in the event of dis-
solution. §§11, 27, id., at 141, 148. Despite these additional
incursions on tribal authority, however, Congress expressly
recognized the Creek’s “tribal existence and present tribal
governmen[t]” and “continued [them] in full force and effect
for all purposes authorized by law.” §28, id., at 148.
In the years that followed, Congress continued to adjust
its arrangements with the Tribe. For example, in 1908, the
Legislature required Creek officials to turn over all “tribal
properties” to the Secretary of the Interior. Act of May 27,
1908, §13, 35 Stat. 316. The next year, Congress sought the
Creek National Council’s release of certain money claims
against the U. S. government. Act of Mar. 3, 1909, ch. 263,
35 Stat. 781, 805. And, further still, Congress offered the
Creek Nation a one-time opportunity to file suit in the fed-
eral Court of Claims for “any and all legal and equitable
claims arising under or growing out of any treaty or agree-
ment between the United States and the Creek Indian Na-
tion.” Act of May 24, 1924, ch. 181, 43 Stat. 139; see, e.g.,
United States v. Creek Nation, 295 U. S. 103 (1935). But
Congress never withdrew its recognition of the tribal gov-
ernment, and none of its adjustments would have made any
sense if Congress thought it had already completed that job.
Indeed, with time, Congress changed course completely.
Beginning in the 1920s, the federal outlook toward Native
Americans shifted “away from assimilation policies and to-
ward more tolerance and respect for traditional aspects of
Indian culture.” 1 Cohen §1.05. Few in 1900 might have
foreseen such a profound “reversal of attitude” was in the
making or expected that “new protections for Indian
rights,” including renewed “support for federally defined
tribalism,” lurked around the corner. Ibid.; see also M.
Scherer, Imperfect Victories: The Legal Tenacity of the
Omaha Tribe, 1945–1995, pp. 2–4 (1999). But that is ex-
actly what happened. Pursuant to this new national policy,
16 MCGIRT v. OKLAHOMA

Opinion of the Court

in 1936, Congress authorized the Creek to adopt a constitu-
tion and bylaws, see Act of June 26, 1936, §3, 49 Stat. 1967,
enabling the Creek government to resume many of its pre-
viously suspended functions. Muscogee (Creek) Nation v.
Hodel, 851 F. 2d 1439, 1442–1447 (CADC 1988).6
The Creek Nation has done exactly that. In the interven-
ing years, it has ratified a new constitution and established
three separate branches of government. Ibid.; see Mus-
cogee Creek Nation (MCN) Const., Arts. V, VI, and VII. To-
day the Nation is led by a democratically elected Principal
Chief, Second Chief, and National Council; operates a police
force and three hospitals; commands an annual budget of
more than $350 million; and employs over 2,000 people.
Brief for Muscogee (Creek) Nation as Amicus Curiae 36–39.
In 1982, the Nation passed an ordinance reestablishing the
criminal and civil jurisdiction of its courts. See Hodel, 851
F. 2d, at 1442, 1446–1447 (confirming Tribe’s authority to
do so). The territorial jurisdiction of these courts extends
to any Indian country within the Tribe’s territory as defined
by the Treaty of 1866. MCN Stat. 27, §1–102(A). And the
State of Oklahoma has afforded full faith and credit to its
judgments since at least 1994. See Barrett v. Barrett, 878
——————
6 The dissent calls it “fantasy” to suggest that Congress evinced “any

unease about extinguishing the Creek domain” because Congress “did
what it set out to do: transform a reservation into a State.” Post, at 22–
23. The dissent stresses, too, that the Creek were afforded U. S. citizen-
ship and the right to vote. Post, at 20. But the only thing implausible
here is the suggestion that “creat[ing] a new State” or enfranchising Na-
tive Americans implies an “intent to terminate” any and all reservations
within a State’s boundaries. Post, at 15. This Court confronted—and
rejected—that sort of argument long ago in United States v. Sandoval,
231 U. S. 28, 47–48 (1913). The dissent treats that case as a one-off:
special because “the tribe in Sandoval, the Pueblo Indians of New Mex-
ico, retained a rare communal title to their lands.” Post, at 21, n. 4. But
Sandoval is not only a case about the Pueblos; it is a foundational prec-
edent recognizing that Congress can welcome Native Americans to par-
ticipate in a broader political community without sacrificing their tribal
sovereignty.
 Cite as: 591 U. S. ____ (2020) 17

Opinion of the Court

P. 2d 1051, 1054 (Okla. 1994); Full Faith and Credit of
Tribal Courts, Okla. State Cts. Network (Apr. 18, 2019),
https://www.oscn.net/applications/oscn/DeliverDocument.
asp?CiteID=458214.
Maybe some of these changes happened for altruistic rea-
sons, maybe some for other reasons. It seems, for example,
that at least certain Members of Congress hesitated about
disestablishment in 1906 because they feared any reversion
of the Creek lands to the public domain would trigger a stat-
utory commitment to hand over portions of these lands to
already powerful railroad interests. See, e.g., 40 Cong. Rec.
2976 (1906) (Sen. McCumber); Id., at 3053 (Sen. Aldrich).
Many of those who advanced the reorganization efforts of
the 1930s may have done so more out of frustration with
efforts to assimilate Native Americans than any disaffec-
tion with assimilation as the ultimate goal. See 1 Cohen
§1.05; Scherer, Imperfect Victories, at 2–4. But whatever
the confluence of reasons, in all this history there simply
arrived no moment when any Act of Congress dissolved the
Creek Tribe or disestablished its reservation. In the end,
Congress moved in the opposite direction.7
D
Ultimately, Oklahoma is left to pursue a very different
sort of argument. Now, the State points to historical prac-
tices and demographics, both around the time of and long
after the enactment of all the relevant legislation. These
facts, the State submits, are enough by themselves to prove
disestablishment. Oklahoma even classifies and catego-

——————
7 The dissent ultimately concedes what Oklahoma will not: that no

“individual congressional action or piece of evidence, standing alone, dis-
established the Creek reservation.” Post, at 9–10. Instead we’re told we
must consider “all of the relevant Acts of Congress together, viewed in
light of contemporaneous and subsequent contextual evidence.” Ibid. So,
once again, the dissent seems to suggest that it’s the arguments in the
next section that will get us across the line to disestablishment.
18 MCGIRT v. OKLAHOMA

Opinion of the Court

rizes how we should approach the question of disestablish-
ment into three “steps.” It reads Solem as requiring us to
examine the laws passed by Congress at the first step, con-
temporary events at the second, and even later events and
demographics at the third. On the State’s account, we have
so far finished only the first step; two more await.
This is mistaken. When interpreting Congress’s work in
this arena, no less than any other, our charge is usually to
ascertain and follow the original meaning of the law before
us. New Prime Inc. v. Oliveira, 586 U. S. ___, ___ (2019)
(slip op., at 6). That is the only “step” proper for a court of
law. To be sure, if during the course of our work an ambig-
uous statutory term or phrase emerges, we will sometimes
consult contemporaneous usages, customs, and practices to
the extent they shed light on the meaning of the language
in question at the time of enactment. Ibid. But Oklahoma
does not point to any ambiguous language in any of the rel-
evant statutes that could plausibly be read as an Act of dis-
establishment. Nor may a court favor contemporaneous or
later practices instead of the laws Congress passed. As So-
lem explained, “[o]nce a block of land is set aside for an In-
dian reservation and no matter what happens to the title of
individual plots within the area, the entire block retains its
reservation status until Congress explicitly indicates other-
wise.” 465 U. S., at 470 (citing United States v. Celestine,
215 U. S. 278, 285 (1909)).
Still, Oklahoma reminds us that other language in Solem
isn’t so constrained. In particular, the State highlights a
passage suggesting that “[w]here non-Indian settlers
flooded into the opened portion of a reservation and the
area has long since lost its Indian character, we have
acknowledged that de facto, if not de jure, diminishment
may have occurred.” 465 U. S., at 471. While acknowledg-
ing that resort to subsequent demographics was “an unor-
thodox and potentially unreliable method of statutory in-
terpretation,” the Court seemed nonetheless taken by its
 Cite as: 591 U. S. ____ (2020) 19

Opinion of the Court

“obvious practical advantages.” Id., at 472, n. 13, 471.
Out of context, statements like these might suggest his-
torical practices or current demographics can suffice to dis-
establish or diminish reservations in the way Oklahoma en-
visions. But, in the end, Solem itself found these kinds of
arguments provided “no help” in resolving the dispute be-
fore it. Id., at 478. Notably, too, Solem suggested that
whatever utility historical practice or demographics might
have was “demonstrated” by this Court’s earlier decision in
Rosebud Sioux Tribe v. Kneip, 430 U. S. 584 (1977). See
Solem, 465 U. S., at 470, n. 10. And Rosebud Sioux hardly
endorsed the use of such sources to find disestablishment.
Instead, based on the statute at issue there, the Court came
“to the firm conclusion that congressional intent” was to di-
minish the reservation in question. 430 U. S., at 603. At
that point, the Tribe sought to cast doubt on the clear im-
port of the text by citing subsequent historical events—and
the Court rejected the Tribe’s argument exactly because this
kind of evidence could not overcome congressional intent as
expressed in a statute. Id., at 604–605.
This Court has already sought to clarify that extratextual
considerations hardly supply the blank check Oklahoma
supposes. In Parker, for example, we explained that “[e]vi-
dence of the subsequent treatment of the disputed land . . .
has ‘limited interpretive value.’ ” 577 U. S., at ___ (slip op.,
at 11) (quoting South Dakota v. Yankton Sioux Tribe, 522
U. S. 329, 355 (1998)).8 Yankton Sioux called it the “least

——————
8 The dissent suggests Parker meant to say only that evidence of sub-

sequent treatment had limited interpretative value “in that case.” Post,
at 12. But the dissent includes just a snippet of the relevant passage.
Read in full, there is little room to doubt Parker invoked a general rule:
“This subsequent demographic history cannot overcome our conclusion
that Congress did not intend to diminish the reservation in 1882. And it
is not our rule to ‘rewrite’ the 1882 Act in light of this subsequent demo-
graphic history. DeCoteau, 420 U. S., at 447. After all, evidence of the
changing demographics of disputed land is ‘the least compelling’ evi-
20 MCGIRT v. OKLAHOMA

Opinion of the Court

compelling” form of evidence. Id., at 356. Both cases em-
phasized that what value such evidence has can only be in-
terpretative—evidence that, at best, might be used to the
extent it sheds light on what the terms found in a statute
meant at the time of the law’s adoption, not as an alterna-
tive means of proving disestablishment or diminishment.
To avoid further confusion, we restate the point. There
is no need to consult extratextual sources when the mean-
ing of a statute’s terms is clear. Nor may extratextual
sources overcome those terms. The only role such materials
can properly play is to help “clear up . . . not create” ambi-
guity about a statute’s original meaning. Milner v. Depart-
ment of Navy, 562 U. S. 562, 574 (2011). And, as we have
said time and again, once a reservation is established, it re-
tains that status “until Congress explicitly indicates other-
wise.” Solem, 465 U. S., at 470 (citing Celestine, 215 U. S.,
at 285); see also Yankton Sioux, 522 U. S., at 343 (“[O]nly
Congress can alter the terms of an Indian treaty by dimin-
ishing a reservation, and its intent to do so must be clear
and plain”) (citation and internal quotation marks omitted).
The dissent charges that we have failed to take account
of the “compelling reasons” for considering extratextual ev-
idence as a matter of course. Post, at 11–12. But Oklahoma
and the dissent have cited no case in which this Court has
found a reservation disestablished without first concluding
that a statute required that result. Perhaps they wish this
case to be the first. To follow Oklahoma and the dissent
down that path, though, would only serve to allow States
and courts to finish work Congress has left undone, usurp
——————
dence in our diminishment analysis, for ‘[e]very surplus land Act neces-
sarily resulted in a surge of non-Indian settlement and degraded the “In-
dian character” of the reservation, yet we have repeatedly stated that not
every surplus land Act diminished the affected reservation.’ Yankton
Sioux, 522 U. S., at 356. . . . Evidence of the subsequent treatment of the
disputed land by Government officials likewise has ‘limited interpretive
value.’ Id., at 355.” 577 U. S., at ___ (slip op., at 11).
 Cite as: 591 U. S. ____ (2020) 21

Opinion of the Court

the legislative function in the process, and treat Native
American claims of statutory right as less valuable than
others. None of that can be reconciled with our normal in-
terpretive rules, let alone our rule that disestablishment
may not be lightly inferred and treaty rights are to be con-
strued in favor, not against, tribal rights. Solem, 465 U. S.,
at 472.9
To see the perils of substituting stories for statutes, we
need look no further than the stories we are offered in the
case before us. Put aside that the Tribe could tell more than
a few stories of its own: Take just the evidence on which
Oklahoma and the dissent wish to rest their case. First,
they point to Oklahoma’s long historical prosecutorial prac-
tice of asserting jurisdiction over Indians in state court,
even for serious crimes on the contested lands. If the Creek
lands really were part of a reservation, the argument goes,
all of these cases should have been tried in federal court
pursuant to the MCA. Yet, until the Tenth Circuit’s Mur-
phy decision a few years ago, no court embraced that possi-
bility. See Murphy, 875 F. 3d 896. Second, they offer state-
ments from various sources to show that “everyone” in the
late 19th and early 20th century thought the reservation
system—and the Creek Nation—would be disbanded soon.
Third, they stress that non-Indians swiftly moved on to the
reservation in the early part of the last century, that Tribe

——————
9 In an effort to support its very different course, the dissent stitches

together quotes from Rosebud Sioux Tribe v. Knelp, 430 U. S. 584 (1977),
and South Dakota v. Yankton Sioux Tribe, 522 U. S. 329 (1998). Post, at
10–11. But far from supporting the dissent, both cases emphasize that
“[t]he focus of our inquiry is congressional intent,” Rosebud, 430 U. S., at
588, n. 4; see also Yankton Sioux, 522 U. S., at 343, and merely
acknowledge that extratextual sources may help resolve ambiguity about
Congress’s directions. The dissent’s appeal to Solem fares no better. As
we have seen, the extratextual sources in Solem only confirmed what the
relevant statute already suggested—that the reservation in question was
not diminished or disestablished. 465 U. S., at 475–476.
22 MCGIRT v. OKLAHOMA

Opinion of the Court

members today constitute a small fraction of those now re-
siding on the land, and that the area now includes a “vi-
brant city with expanding aerospace, healthcare, technol-
ogy, manufacturing, and transportation sectors.” Brief for
Petitioner in Carpenter v. Murphy, O. T. 2018, No. 17–1107,
p. 15. All this history, we are told, supplies “compelling”
evidence about the lands in question.
Maybe so, but even taken on its own terms none of this
evidence tells the story we are promised. Start with the
State’s argument about its longstanding practice of assert-
ing jurisdiction over Native Americans. Oklahoma pro-
ceeds on the implicit premise that its historical practices
are unlikely to have defied the mandates of the federal
MCA. That premise, though, appears more than a little
shaky. In conjunction with the MCA, §1151(a) not only
sends to federal court certain major crimes committed by
Indians on reservations. Two doors down, in §1151(c), the
statute does the same for major crimes committed by Indi-
ans on “Indian allotments, the Indian titles of which have
not been extinguished.” Despite this direction, however,
Oklahoma state courts erroneously entertained prosecu-
tions for major crimes by Indians on Indian allotments for
decades, until state courts finally disavowed the practice in
1989. See State v. Klindt, 782 P. 2d 401, 404 (Okla. Crim.
App. 1989) (overruling Ex parte Nowabbi, 60 Okla. Crim.
III, 61 P. 2d 1139 (1936)); see also United States v. Sands,
968 F. 2d 1058, 1062–1063 (CA10 1992). And if the State’s
prosecution practices disregarded §1151(c) for so long, it’s
unclear why we should take those same practices as a reli-
able guide to the meaning and application of §1151(a).
Things only get worse from there. Why did Oklahoma
historically think it could try Native Americans for any
crime committed on restricted allotments or anywhere else?
Part of the explanation, Oklahoma tells us, is that it
thought the eastern half of the State was always categori-
cally exempt from the terms of the federal MCA. So
 Cite as: 591 U. S. ____ (2020) 23

Opinion of the Court

whether a crime was committed on a restricted allotment,
a reservation, or land that wasn’t Indian country at all, to
Oklahoma it just didn’t matter. In the State’s view, when
Congress adopted the Oklahoma Enabling Act that paved
the way for its admission to the Union, it carved out a spe-
cial exception to the MCA for the eastern half of the State
where the Creek lands can be found. By Oklahoma’s own
admission, then, for decades its historical practices in the
area in question didn’t even try to conform to the MCA, all
of which makes the State’s past prosecutions a meaningless
guide for determining what counted as Indian country. As
it turns out, too, Oklahoma’s claim to a special exemption
was itself mistaken, yet one more error in historical prac-
tice that even the dissent does not attempt to defend. See
Part V, infra.10
To be fair, Oklahoma is far from the only State that has
overstepped its authority in Indian country. Perhaps often
in good faith, perhaps sometimes not, others made similar
mistakes in the past. But all that only underscores further
the danger of relying on state practices to determine the
meaning of the federal MCA. See, e.g., Negonsett, 507 U. S.,
at 106–107 (“[I]n practice, Kansas had exercised jurisdic-
tion over all offenses committed on Indian reservations in-
volving Indians” (quoting memorandum from Secretary of
the Interior, H. R. Rep. No. 1999, 76th Cong., 3d Sess., 4
(1940)); Scherer, Imperfect Victories, at 18 (describing “na-
tionwide jurisdictional confusion” as a result of the MCA);

——————
10 The dissent tries to avoid this inconvenient history by distinguishing
fee allotments from reservations, noting that the two categories are le-
gally distinct and geographically incommensurate. Post, at 27. But this
misses the point: The reason that Oklahoma thought it could prosecute
Indians for crimes on restricted allotments applied with equal force to
reservations. And it hardly “stretches the imagination” to think that
reason was wrong, post, at 28, when the dissent itself does not dispute
our rejection of it in Part V.
24 MCGIRT v. OKLAHOMA

Opinion of the Court

Cohen §6.04(4)(a) (“Before 1942 the state of New York reg-
ularly exercised or claimed the right to exercise jurisdiction
over the New York reservations, but a federal court decision
in that year raised questions about the validity of state ju-
risdiction”); Brief for United States as Amicus Curiae in
Carpenter v. Murphy, O. T. 2018, No. 17–1107, pp. 7a–8a
(Letter from Secretary of the Interior, Mar. 27, 1963) (not-
ing that many States have asserted criminal jurisdiction
over Indians without an apparent basis in a federal law).11
Oklahoma next points to various statements during the
allotment era which, it says, show that even the Creek un-
derstood their reservation was under threat. And there’s
no doubt about that. By 1893, the leadership of the Creek
Nation saw what the federal government had in mind:
“They [the federal government] do not deny any of our
rights under treaty, but say they will go to the people them-
selves and confer with them and urge upon them the neces-
sity of a change in their present condition, and upon their
refusal will force a change upon them.” P. Porter & A.
McKellop, Printed Statement of Creek Delegates, reprinted
in Creek Delegation Documents 8–9 (Feb. 9, 1893). Not a
decade later, and as a result of these forced changes, the
leadership recognized that “ ‘[i]t would be difficult, if not im-
possible to successfully operate the Creek government
now.’ ” App. to Brief for Respondent 8a (Message to Creek
——————
11 Unable to answer Oklahoma’s admitted error about the very federal

criminal statute before us, the dissent travels far afield, pointing to the
fact an Oklahoma court heard a civil case in 1915 about an inheritance—
involving members of a different Tribe—as “evidence” Congress dises-
tablished the Creek Reservation. See post, at 21 (citing Palmer v. Cully,
52 Okla. 454, 455–465, 153 P. 154, 155–157 (1915) (per curiam)). But
even assuming that Oklahoma courts exercised civil jurisdiction over
Creek members, too, the dissent never explains why this jurisdiction im-
plies the Creek Reservation must have been disestablished. After all,
everyone agrees that the Creeks were prohibited from having their own
courts at the time. So it should be no surprise that some Creek might
have resorted to state courts in hope of resolving their disputes.
 Cite as: 591 U. S. ____ (2020) 25

Opinion of the Court

National Council (May 7, 1901), reprinted in The Indian
Journal (May 10, 1901)). Surely, too, the future looked even
bleaker: “ ‘The remnant of a government now accorded to us
can be expected to be maintained only until all settlements
of our landed and other interests growing out of treaty stip-
ulations with the government of the United States shall
have been settled.’ ” Ibid.
But note the nature of these statements. The Creek Na-
tion recognized that the federal government will seek to get
popular support or otherwise would force change. Like-
wise, the Tribe’s government would continue for only so
long. These were prophesies, and hardly groundbreaking
ones at that. After all, the 1901 Creek Allotment Agree-
ment explicitly said that the tribal government “shall not
continue” past 1906. §46, 31 Stat. 872. So what might
statements like these tell us that isn’t already evident from
the statutes themselves? Oklahoma doesn’t suggest they
shed light on the meaning of some disputed and ambiguous
statutory direction. More nearly, the State seeks to render
the Creek’s fears self-fulfilling.12
We are also asked to consider commentary from those
outside the Tribe. In particular, the dissent reports that
the federal government “operated” on the “understanding”
that the reservation was disestablished. Post, at 32. In
support of its claim, the dissent highlights a 1941 statement
from Felix Cohen. Then serving as an official at the Interior
Department, Cohen opined that “ ‘all offenses by or against
Indians’ in the former Indian Territory ‘are subject to State
——————
12 The dissent finds the statements of the Creek leadership so proba-

tive that it cites them not just as evidence about the meaning of treaties
the Tribe signed but even as evidence about the meaning of general pur-
pose laws the Creek had no hand in. See post, at 26 (citing Chief Porter’s
views on the legal effects of the Oklahoma Enabling Act). That is quite
a stretch from using tribal statements as “historical evidence of ‘the man-
ner in which [treaties were] negotiated’ with the . . . Tribe.” Parker, 577
U. S., at ___ (slip op., at 9) (quoting Solem v. Bartlett, 465 U. S. 463, 471
(1984)).
26 MCGIRT v. OKLAHOMA

Opinion of the Court

laws.’ ” Ibid. (quoting App. to Supp. Reply Brief for Peti-
tioner in Carpenter v. Murphy, O. T. 2018, No. 17–1107, p.
1a (Memorandum for Commissioner of Indian Affairs (July
11, 1941)). But that statement is incorrect. As we have just
seen, Oklahoma’s courts acknowledge that the State lacks
jurisdiction over Indian crimes on Indian allotments. See
Klindt, 782 P. 2d, at 403–404. And the dissent does not dis-
pute that Oklahoma is without authority under the MCA to
try Indians for crimes committed on restricted allotments
and any reservation. All of which highlights the pitfalls of
elevating commentary over the law.13
Finally, Oklahoma points to the speedy and persistent
movement of white settlers onto Creek lands throughout
the late 19th and early 20th centuries. But this history
proves no more helpful in discerning statutory meaning.
Maybe, as Oklahoma supposes, it suggests that some white
settlers in good faith thought the Creek lands no longer con-
stituted a reservation. But maybe, too, some didn’t care and
——————
13 Part of the reason for Cohen’s error might be explained by a portion

of the memorandum the dissent leaves unquoted. Cohen concluded that
Oklahoma was free to try Indians anywhere in the State because, among
other things, the Oklahoma Enabling Act “transfer[red] . . . jurisdiction
from the Federal courts to the State courts upon the establishment of the
State of Oklahoma.” App. to Supp. Reply Brief for Petitioner in Carpen-
ter v. Murphy, O. T. 2018, No. 17–1107, p. 1a (Memorandum for Commis-
sioner of Indian Affairs (July 11, 1941)). Yet, as we explore below, the
Oklahoma Enabling Act did not send cases covered by the federal MCA
to state court. See Part V, infra. Other, contemporaneous Interior De-
partment memoranda acknowledged that Oklahoma state courts had
simply “assumed jurisdiction” over cases arising on restricted allotments
without any clear authority in the Oklahoma Enabling Act or the MCA,
and much the same appears to have occurred here. App. to Supp. Reply
Brief for Respondent in Carpenter v. Murphy, O. T. 2018, No. 17–1107,
p. 1a (Memorandum from N. Gray, Dept. of Interior, for Mr. Flanery
(Aug. 12, 1942)). So rather than Oklahoma and the United States having
a “shared understanding” that Congress had disestablished the Creek
Reservation, post, at 27, it seems more accurate to say that for many
years much uncertainty remained about whether the MCA applied in
eastern Oklahoma.
 Cite as: 591 U. S. ____ (2020) 27

Opinion of the Court

others never paused to think about the question. Certain
historians have argued, for example, that the loss of Creek
land ownership was accelerated by the discovery of oil in
the region during the period at issue here. A number of the
federal officials charged with implementing the laws of
Congress were apparently openly conflicted, holding shares
or board positions in the very oil companies who sought to
deprive Indians of their lands. A. Debo, And Still the Wa-
ters Run 86–87, 117–118 (1940). And for a time Okla-
homa’s courts appear to have entertained sham competency
and guardianship proceedings that divested Tribe members
of oil rich allotments. Id., at 104–106, 233–234; Brief for
Historians et al. as Amici Curiae 26–30. Whatever else
might be said about the history and demographics placed
before us, they hardly tell a story of unalloyed respect for
tribal interests.14
In the end, only one message rings true. Even the care-
fully selected history Oklahoma and the dissent recite is not
nearly as tidy as they suggest. It supplies us with little help

——————
14 The dissent asks us to examine a hodge-podge of other, but no more

compelling, material. For example, the dissent points to later statutes
that do no more than confirm there are former reservations in the State
of Oklahoma. Post, at 30–31. It cites legislative history to show that
Congress had the Creek Nation—or, at least, its neighbors—in mind
when it added these in 1988. Post, at 31, n. 7. The dissent cites a Senate
Report from 1989 and post-1980 statements made by representatives of
other tribes. Post, at 30, 32–33. It highlights three occasions on which
this Court referred to something like a “former Creek Nation,” though it
neglects to add that in each the Court was referring to the loss of the
Nation’s communal fee title, not its sovereignty. Grayson v. Harris, 267
U. S. 352, 357 (1925); Woodward v. DeGraffenreid, 238 U. S. 284, 289–
290 (1915); Washington v. Miller, 235 U. S. 422, 423–425 (1914). The
dissent points as well to a single instance in which the Creek Nation dis-
claimed reservation boundaries for purposes of litigation in a lower court,
post, at 32, but ignores that the Creek Nation has repeatedly filed briefs
in this Court to the contrary. This is thin gruel to set against treaty
promises enshrined in statutes.
28 MCGIRT v. OKLAHOMA

Opinion of the Court

in discerning the law’s meaning and much potential for mis-
chief. If anything, the persistent if unspoken message here
seems to be that we should be taken by the “practical ad-
vantages” of ignoring the written law. How much easier it
would be, after all, to let the State proceed as it has always
assumed it might. But just imagine what it would mean to
indulge that path. A State exercises jurisdiction over Na-
tive Americans with such persistence that the practice
seems normal. Indian landowners lose their titles by fraud
or otherwise in sufficient volume that no one remembers
whose land it once was. All this continues for long enough
that a reservation that was once beyond doubt becomes
questionable, and then even farfetched. Sprinkle in a few
predictions here, some contestable commentary there, and
the job is done, a reservation is disestablished. None of
these moves would be permitted in any other area of statu-
tory interpretation, and there is no reason why they should
be permitted here. That would be the rule of the strong, not
the rule of law.
IV
Unable to show that Congress disestablished the Creek
Reservation, Oklahoma next tries to turn the tables in a
completely different way. Now, it contends, Congress never
established a reservation in the first place. Over all the
years, from the federal government’s first guarantees of
land and self-government in 1832 and through the litany of
promises that followed, the Tribe never received a reserva-
tion. Instead, what the Tribe has had all this time qualifies
only as a “dependent Indian community.”
Even if we were to accept Oklahoma’s bold feat of reclas-
sification, however, it’s hardly clear the State would win
this case. “Reservation[s]” and “Indian allotments, the In-
dian titles to which have not been extinguished,” qualify as
Indian country under subsections (a) and (c) of §1151. But
“dependent Indian communities” also qualify as Indian
 Cite as: 591 U. S. ____ (2020) 29

Opinion of the Court

country under subsection (b). So Oklahoma lacks jurisdic-
tion to prosecute Mr. McGirt whether the Creek lands hap-
pen to fall in one category or another.
About this, Oklahoma is at least candid. It admits the
entire point of its reclassification exercise is to avoid So-
lem’s rule that only Congress may disestablish a reserva-
tion. And to achieve that, the State has to persuade us not
only that the Creek lands constitute a “dependent Indian
community” rather than a reservation. It also has to con-
vince us that we should announce a rule that dependent In-
dian community status can be lost more easily than reser-
vation status, maybe even by the happenstance of shifting
demographics.
To answer this argument, it’s enough to address its first
essential premise. Holding that the Creek never had a res-
ervation would require us to stand willfully blind before a
host of federal statutes. Perhaps that is why the Solicitor
General, who supports Oklahoma’s disestablishment argu-
ment, refuses to endorse this alternative effort. It also may
be why Oklahoma introduced this argument for affirmance
only for the first time in this Court. And it may be why the
dissent makes no attempt to defend Oklahoma here. What
are we to make of the federal government’s repeated treaty
promises that the land would be “solemnly guarantied to
the Creek Indians,” that it would be a “permanent home,”
“forever set apart,” in which the Creek would be “secured in
the unrestricted right of self-government”? What about
Congress’s repeated references to a “Creek reservation” in
its statutes? No one doubts that this kind of language nor-
mally suffices to establish a federal reservation. So what
could possibly make this case different?
Oklahoma’s answer only gets more surprising. The rea-
son that the Creek’s lands are not a reservation, we’re told,
is that the Creek Nation originally held fee title. Recall that
the Indian Removal Act authorized the President not only
to “solemnly . . . assure the tribe . . . that the United States
30 MCGIRT v. OKLAHOMA

Opinion of the Court

will forever secure and guaranty to them . . . the country so
exchanged with them,” but also, “if they prefer it, . . . the
United States will cause a patent or grant to be made and
executed to them for the same.” 4 Stat. 412. Recall that the
Creek insisted on this additional protection when negotiat-
ing the Treaty of 1833, and in fact received a land patent
pursuant to that treaty some 19 years later. In the eyes of
Oklahoma, the Tribe’s choice on this score was a fateful one.
By asking for (and receiving) fee title to their lands, the
Creek inadvertently made their tribal sovereignty easier to
divest rather than harder.
The core of Oklahoma’s argument is that a reservation
must be land “reserved from sale.” Celestine, 215 U. S., at
285. Often, that condition is satisfied when the federal gov-
ernment promises to hold aside a particular piece of feder-
ally owned land in trust for the benefit of the Tribe. And,
admittedly, the Creek’s arrangement was different, be-
cause the Tribe held “fee simple title, not the usual Indian
right of occupancy.” United States v. Creek Nation, 295
U. S. 103, 109 (1935). Still, as we explained in Part II, the
land was reserved from sale in the very real sense that the
government could not “give the tribal lands to others, or to
appropriate them to its own purposes,” without engaging in
“ ‘an act of confiscation.’ ” Id., at 110.
It’s hard to see, too, how any difference between these two
arrangements might work to the detriment of the Tribe.
Just as we have never insisted on any particular form of
words when it comes to disestablishing a reservation, we
have never done so when it comes to establishing one. See
Minnesota v. Hitchcock, 185 U. S. 373, 390 (1902) (“[I]n or-
der to create a reservation it is not necessary that there
should be a formal cession or a formal act setting apart a
particular tract. It is enough that from what has been there
results a certain defined tract appropriated to certain pur-
poses”). As long as 120 years ago, the federal court for the
Indian Territory recognized all this and rightly rejected the
 Cite as: 591 U. S. ____ (2020) 31

Opinion of the Court

notion that fee title is somehow inherently incompatible
with reservation status. Maxey v. Wright, 54 S. W. 807, 810
(Indian Terr. 1900).
By now, Oklahoma’s next move will seem familiar. Seek-
ing to sow doubt around express treaty promises, it cites
some stray language from a statute that does not control
here, a piece of congressional testimony there, and the scat-
tered opinions of agency officials everywhere in between.
See, e.g., Act of July 31, 1882, ch. 360, 22 Stat. 179 (refer-
ring to Creek land as “Indian country” as opposed to an “In-
dian reservation”); S. Doc. No. 143, 59th Cong., 1st. Sess.,
33 (1906) (Chief of Choctaw Nation—which had an arrange-
ment similar to the Creek’s—testified that both Tribes “ob-
ject to being classified with the reservation Indians”); Dept.
of Interior, Census Office, Report on Indians Taxed and In-
dians Not Taxed in the U. S. 284 (1894) (Creeks and neigh-
boring Tribes were “not on the ordinary Indian reservation,
but on lands patented to them by the United States”). Ok-
lahoma stresses that this Court even once called the Creek
lands a “dependent Indian community,” though it used that
phrase in passing and only to show that the Tribe’s “prop-
erty and affairs were subject to the control and manage-
ment of that government”—a point that would also be true
if the lands were a reservation. Creek Nation, 295 U. S., at
109. Unsurprisingly given the Creek Nation’s nearly 200-
year occupancy of these lands, both sides have turned up a
few clues suggesting the label “reservation” either did or did
not apply. One thing everyone can agree on is this history
is long and messy.
But the most authoritative evidence of the Creek’s rela-
tionship to the land lies not in these scattered references; it
lies in the treaties and statutes that promised the land to
the Tribe in the first place. And, if not for the Tribe’s fee
title to its land, no one would question that these treaties
and statutes created a reservation. So the State’s argument
inescapably boils down to the untenable suggestion that,
32 MCGIRT v. OKLAHOMA

Opinion of the Court

when the federal government agreed to offer more protec-
tion for tribal lands, it really provided less. All this time,
fee title was nothing more than another trap for the wary.
V
That leaves Oklahoma to attempt yet another argument
in the alternative. We alluded to it earlier in Part III. Now,
the State accepts for argument’s sake that the Creek land
is a reservation and thus “Indian country” for purposes of
the Major Crimes Act. It accepts, too, that this would nor-
mally mean serious crimes by Indians on the Creek Reser-
vation would have to be tried in federal court. But, the
State tells us, none of that matters; everything the parties
have briefed and argued so far is beside the point. It’s all
irrelevant because it turns out the MCA just doesn’t apply
to the eastern half of Oklahoma, and it never has. That
federal law may apply to other States, even to the western
half of Oklahoma itself. But eastern Oklahoma is and has
always been exempt. So whether or not the Creek have a
reservation, the State’s historic practices have always been
correct and it remains free to try individuals like Mr.
McGirt in its own courts.
Notably, the dissent again declines to join Oklahoma in
its latest twist. And, it turns out, for good reason. In sup-
port of its argument, Oklahoma points to statutory artifacts
from its territorial history. The State of Oklahoma was
formed from two territories: the Oklahoma Territory in the
west and Indian Territory in the east. Originally, it seems
criminal prosecutions in the Indian Territory were split be-
tween tribal and federal courts. See Act of May 2, 1890,
§30, 26 Stat. 94. But, in 1897, Congress abolished that
scheme, granting the U. S. Courts of the Indian Territory
“exclusive jurisdiction” to try “all criminal causes for the
punishment of any offense.” Act of June 7, 1897, 30 Stat.
83. These federal territorial courts applied federal law and
 Cite as: 591 U. S. ____ (2020) 33

Opinion of the Court

state law borrowed from Arkansas “to all persons . . . irre-
spective of race.” Ibid. A year later, Congress abolished
tribal courts and transferred all pending criminal cases to
U. S. courts of the Indian Territory. Curtis Act of 1898, §28,
30 Stat. 504–505. And, Oklahoma says, sending Indians to
federal court and all others to state court would be incon-
sistent with this established and enlightened policy of ap-
plying the same law in the same courts to everyone.
Here again, however, arguments along these and similar
lines have been “frequently raised” but rarely “accepted.”
United States v. Sands, 968 F. 2d 1058, 1061 (CA10 1992)
(Kelly, J.). “The policy of leaving Indians free from state
jurisdiction and control is deeply rooted in this Nation’s his-
tory.” Rice v. Olson, 324 U. S. 786, 789 (1945). Chief Jus-
tice Marshall, for example, held that Indian Tribes were
“distinct political communities, having territorial bounda-
ries, within which their authority is exclusive . . . which is
not only acknowledged, but guarantied by the United
States,” a power dependent on and subject to no state au-
thority. Worcester v. Georgia, 6 Pet. 515, 557 (1832); see
also McClanahan v. Arizona Tax Comm’n, 411 U. S. 164,
168–169 (1973). And in many treaties, like those now be-
fore us, the federal government promised Indian Tribes the
right to continue to govern themselves. For all these rea-
sons, this Court has long “require[d] a clear expression of
the intention of Congress” before the state or federal gov-
ernment may try Indians for conduct on their lands. Ex
parte Crow Dog, 109 U. S. 556, 572 (1883).
Oklahoma cannot come close to satisfying this standard.
In fact, the only law that speaks expressly here speaks
against the State. When Oklahoma won statehood in 1907,
the MCA applied immediately according to its plain terms.
That statute, as phrased at the time, provided exclusive fed-
eral jurisdiction over qualifying crimes by Indians in “any
Indian reservation” located within “the boundaries of any
34 MCGIRT v. OKLAHOMA

Opinion of the Court

State.” Act of Mar. 3, 1885, ch. 341, §9, 23 Stat. 385 (em-
phasis added); see also 18 U. S. C. §1151 (defining “Indian
country” even more broadly). By contrast, every one of the
statutes the State directs us to merely discusses the assign-
ment of cases among courts in the Indian Territory. They
say nothing about the division of responsibilities between
federal and state authorities after Oklahoma entered the
Union. And however enlightened the State may think it
was for territorial law to apply to all persons irrespective of
race, some Tribe members may see things differently, given
that the same policy entailed the forcible closure of tribal
courts in defiance of treaty terms.
Left to hunt for some statute that might have rendered
the MCA inapplicable in Oklahoma after statehood, the
best the State can find is the Oklahoma Enabling Act. Con-
gress adopted that law in preparation for Oklahoma’s ad-
mission in 1907. Among its many provisions sorting out the
details associated with Oklahoma’s transition to statehood,
the Enabling Act transferred all nonfederal cases pending
in territorial courts to Oklahoma’s new state courts. Act of
June 16, 1906, §20, 34 Stat. 277; see also Act of Mar. 4,
1907, §3, 34 Stat. 1287 (clarifying treatment of cases to
which United States was a party). The State says this
transfer made its courts the inheritors of the federal terri-
torial courts’ sweeping authority to try Indians for crimes
committed on reservations.
But, at best, this tells only half the story. The Enabling
Act not only sent all nonfederal cases pending in territorial
courts to state court. It also transferred pending cases that
arose “under the Constitution, laws, or treaties of the
United States” to federal district courts. §16, 34 Stat. 277.
Pending criminal cases were thus transferred to federal
court if the prosecution would have belonged there had the
Territory been a State at the time of the crime. §1, 34 Stat.
1287 (amending the Enabling Act). Nor did the statute
make any distinction between cases arising in the former
 Cite as: 591 U. S. ____ (2020) 35

Opinion of the Court

eastern (Indian) and western (Oklahoma) territories. So,
simply put, the Enabling Act sent state-law cases to state
court and federal-law cases to federal court. And serious
crimes by Indians in Indian country were matters that
arose under the federal MCA and thus properly belonged in
federal court from day one, wherever they arose within the
new State.
Maybe that’s right, Oklahoma acknowledges, but that’s
not what happened. Instead, for many years the State con-
tinued to try Indians for crimes committed anywhere
within its borders. But what can that tell us? The State
identifies not a single ambiguous statutory term in the
MCA that its actions might illuminate. And, as we have
seen, its own courts have acknowledged that the State’s his-
toric practices deviated in meaningful ways from the MCA’s
terms. See supra, at 22–23. So, once more, it seems Okla-
homa asks us to defer to its usual practices instead of fed-
eral law, something we will not and may never do.
That takes Oklahoma down to its last straw when it
comes to the MCA. If Oklahoma lacks the jurisdiction to
try Native Americans it has historically claimed, that
means at the time of its entry into the Union no one had the
power to try minor Indian-on-Indian crimes committed in
Indian country. This much follows, Oklahoma reminds us,
because the MCA provides federal jurisdiction only for ma-
jor crimes, and no tribal forum existed to try lesser cases
after Congress abolished the tribal courts in 1898. Curtis
Act, §28, 30 Stat. 504–505. Whatever one thinks about the
plausibility of other discontinuities between federal law
and state practice, the State says, it is unthinkable that
Congress would have allowed such a significant “jurisdic-
tional gap” to open at the moment Oklahoma achieved
statehood.
But what the State considers unthinkable turns out to be
easily imagined. Jurisdictional gaps are hardly foreign to
this area of the law. See, e.g., Duro v. Reina, 495 U. S. 676,
36 MCGIRT v. OKLAHOMA

Opinion of the Court

704–706 (1990) (Brennan, J., dissenting). Many tribal
courts across the country were absent or ineffective during
the early part of the last century, yielding just the sort of
gaps Oklahoma would have us believe impossible. Indeed,
this might be why so many States joined Oklahoma in pros-
ecuting Indians without proper jurisdiction. The judicial
mind abhors a vacuum, and the temptation for state prose-
cutors to step into the void was surely strong. See supra, at
23–24.
With time, too, Congress has filled many of the gaps Ok-
lahoma worries about. One way Congress has done so is by
reauthorizing tribal courts to hear minor crimes in Indian
country. Congress chose exactly this course for the Creeks
and others in 1936. Act of June 26, 1936, §3, 49 Stat. 1967;
see also Hodel, 851 F. 2d, at 1442–1446. Another option
Congress has employed is to allow affected Indian tribes to
consent to state criminal jurisdiction. 25 U. S. C. §§1321(a),
1326. Finally, Congress has sometimes expressly expanded
state criminal jurisdiction in targeted bills addressing spe-
cific States. See, e.g., 18 U. S. C. §3243 (creating jurisdic-
tion for Kansas); Act of May 31, 1946, ch. 279, 60 Stat. 229
(same for a reservation in North Dakota); Act of June 30,
1948, ch. 759, 62 Stat. 1161 (same for certain reservations
in Iowa); 18 U. S. C. §1162 (creating jurisdiction for six ad-
ditional States). But Oklahoma doesn’t claim to have com-
plied with the requirements to assume jurisdiction volun-
tarily over Creek lands. Nor has Congress ever passed a
law conferring jurisdiction on Oklahoma. As a result, the
MCA applies to Oklahoma according to its usual terms:
Only the federal government, not the State, may prosecute
Indians for major crimes committed in Indian country.
VI
In the end, Oklahoma abandons any pretense of law and
speaks openly about the potentially “transform[ative]” ef-
fects of a loss today. Brief for Respondent 43. Here, at
 Cite as: 591 U. S. ____ (2020) 37

Opinion of the Court

least, the State is finally rejoined by the dissent. If we
dared to recognize that the Creek Reservation was never
disestablished, Oklahoma and dissent warn, our holding
might be used by other tribes to vindicate similar treaty
promises. Ultimately, Oklahoma fears that perhaps as
much as half its land and roughly 1.8 million of its residents
could wind up within Indian country.
It’s hard to know what to make of this self-defeating ar-
gument. Each tribe’s treaties must be considered on their
own terms, and the only question before us concerns the
Creek. Of course, the Creek Reservation alone is hardly in-
significant, taking in most of Tulsa and certain neighboring
communities in Northeastern Oklahoma. But neither is it
unheard of for significant non-Indian populations to live
successfully in or near reservations today. See, e.g., Brief
for National Congress of American Indians Fund as Amicus
Curiae 26–28 (describing success of Tacoma, Washington,
and Mount Pleasant, Michigan); see also Parker, 577 U. S.,
at ___–___ (slip op., at 10–12) (holding Pender, Nebraska,
to be within Indian country despite tribe’s absence from the
disputed territory for more than 120 years). Oklahoma re-
plies that its situation is different because the affected pop-
ulation here is large and many of its residents will be sur-
prised to find out they have been living in Indian country
this whole time. But we imagine some members of the 1832
Creek Tribe would be just as surprised to find them there.
What are the consequences the State and dissent worry
might follow from an adverse ruling anyway? Primarily,
they argue that recognizing the continued existence of the
Creek Reservation could unsettle an untold number of con-
victions and frustrate the State’s ability to prosecute crimes
in the future. But the MCA applies only to certain crimes
committed in Indian country by Indian defendants. A
neighboring statute provides that federal law applies to a
broader range of crimes by or against Indians in Indian
country. See 18 U. S. C. §1152. States are otherwise free
38 MCGIRT v. OKLAHOMA

Opinion of the Court

to apply their criminal laws in cases of non-Indian victims
and defendants, including within Indian country. See
McBratney, 104 U. S., at 624. And Oklahoma tells us that
somewhere between 10% and 15% of its citizens identify as
Native American. Given all this, even Oklahoma admits
that the vast majority of its prosecutions will be unaffected
whatever we decide today.
Still, Oklahoma and the dissent fear, “[t]housands” of Na-
tive Americans like Mr. McGirt “wait in the wings” to chal-
lenge the jurisdictional basis of their state-court convic-
tions. Brief for Respondent 3. But this number is
admittedly speculative, because many defendants may
choose to finish their state sentences rather than risk
reprosecution in federal court where sentences can be
graver. Other defendants who do try to challenge their
state convictions may face significant procedural obstacles,
thanks to well-known state and federal limitations on post-
conviction review in criminal proceedings.15
In any event, the magnitude of a legal wrong is no reason
to perpetuate it. When Congress adopted the MCA, it broke
many treaty promises that had once allowed tribes like the
Creek to try their own members. But, in return, Congress
allowed only the federal government, not the States, to try

——————
15 For example, Oklahoma appears to apply a general rule that “issues

that were not raised previously on direct appeal, but which could have
been raised, are waived for further review.” Logan v. State, 2013 OK CR
2, ¶ 1, 293 P. 3d 969, 973. Indeed, JUSTICE THOMAS contends that this
state-law limitation on collateral review prevents us from considering
even the case now before us. Post, at 2 (dissenting opinion). But while
that state-law rule may often bar our way, it doesn’t in this case. After
noting a potential state-law obstacle, the Oklahoma Court of Criminal
Appeals (OCCA) proceeded to address the merits of Mr. McGirt’s federal
MCA claim anyway. Because the OCCA’s opinion “fairly appears to rest
primarily on federal law or to be interwoven with federal law” and lacks
any “plain statement” that it was relying on a state-law ground, we have
jurisdiction to consider the federal-law question presented to us. See
Michigan v. Long, 463 U. S. 1032, 1040–1041, 1044 (1983).
 Cite as: 591 U. S. ____ (2020) 39

Opinion of the Court

tribal members for major crimes. All our decision today
does is vindicate that replacement promise. And if the
threat of unsettling convictions cannot save a precedent of
this Court, see Ramos v. Louisiana, 590 U. S. ___, ___–___
(2020) (plurality opinion) (slip op., at 23–26), it certainly
cannot force us to ignore a statutory promise when no prec-
edent stands before us at all.
What’s more, a decision for either party today risks up-
setting some convictions. Accepting the State’s argument
that the MCA never applied in Oklahoma would preserve
the state-court convictions of people like Mr. McGirt, but
simultaneously call into question every federal conviction
obtained for crimes committed on trust lands and restricted
Indian allotments since Oklahoma recognized its jurisdic-
tional error more than 30 years ago. See supra, at 22. It’s
a consequence of their own arguments that Oklahoma and
the dissent choose to ignore, but one which cannot help but
illustrate the difficulty of trying to guess how a ruling one
way or the other might affect past cases rather than simply
proceeding to apply the law as written.
Looking to the future, Oklahoma warns of the burdens
federal and tribal courts will experience with a wider juris-
diction and increased caseload. But, again, for every juris-
dictional reaction there seems to be an opposite reaction:
recognizing that cases like Mr. McGirt’s belong in federal
court simultaneously takes them out of state court. So
while the federal prosecutors might be initially under-
staffed and Oklahoma prosecutors initially overstaffed, it
doesn’t take a lot of imagination to see how things could
work out in the end.
Finally, the State worries that our decision will have sig-
nificant consequences for civil and regulatory law. The only
question before us, however, concerns the statutory defini-
tion of “Indian country” as it applies in federal criminal law
under the MCA, and often nothing requires other civil stat-
40 MCGIRT v. OKLAHOMA

Opinion of the Court

utes or regulations to rely on definitions found in the crim-
inal law. Of course, many federal civil laws and regulations
do currently borrow from §1151 when defining the scope of
Indian country. But it is far from obvious why this collat-
eral drafting choice should be allowed to skew our interpre-
tation of the MCA, or deny its promised benefits of a federal
criminal forum to tribal members.
It isn’t even clear what the real upshot of this borrowing
into civil law may be. Oklahoma reports that recognizing
the existence of the Creek Reservation for purposes of the
MCA might potentially trigger a variety of federal civil stat-
utes and rules, including ones making the region eligible for
assistance with homeland security, 6 U. S. C. §§601, 606,
historical preservation, 54 U. S. C. §302704, schools, 20
U. S. C. §1443, highways, 23 U. S. C. §120, roads, §202, pri-
mary care clinics, 25 U. S. C. §1616e–1, housing assistance,
§4131, nutritional programs, 7 U. S. C. §§2012, 2013, disa-
bility programs, 20 U. S. C. §1411, and more. But what are
we to make of this? Some may find developments like these
unwelcome, but from what we are told others may celebrate
them.
The dissent isn’t so sanguine—it assures us, without fur-
ther elaboration, that the consequences will be “drastic pre-
cisely because they depart from . . . more than a century [of]
settled understanding.” Post, at 37. The prediction is a fa-
miliar one. Thirty years ago the Solicitor General warned
that “[l]aw enforcement would be rendered very difficult”
and there would be “grave uncertainty regarding the appli-
cation” of state law if courts departed from decades of “long-
held understanding” and recognized that the federal MCA
applies to restricted allotments in Oklahoma. Brief for
United States as Amicus Curiae in Oklahoma v. Brooks,
O.T. 1988, No. 88–1147, pp. 2, 9, 18, 19. Yet, during the
intervening decades none of these predictions panned out,
and that fact stands as a note of caution against too readily
crediting identical warnings today.
 Cite as: 591 U. S. ____ (2020) 41

Opinion of the Court

More importantly, dire warnings are just that, and not a
license for us to disregard the law. By suggesting that our
interpretation of Acts of Congress adopted a century ago
should be inflected based on the costs of enforcing them to-
day, the dissent tips its hand. Yet again, the point of look-
ing at subsequent developments seems not to be determin-
ing the meaning of the laws Congress wrote in 1901 or 1906,
but emphasizing the costs of taking them at their word.
Still, we do not disregard the dissent’s concern for reli-
ance interests. It only seems to us that the concern is mis-
placed. Many other legal doctrines—procedural bars, res
judicata, statutes of repose, and laches, to name a few—are
designed to protect those who have reasonably labored un-
der a mistaken understanding of the law. And it is precisely
because those doctrines exist that we are “fre[e] to say what
we know to be true . . . today, while leaving questions about
. . . reliance interest[s] for later proceedings crafted to ac-
count for them.” Ramos, 590 U. S., at ___ (plurality opin-
ion) (slip op., at 24).
In reaching our conclusion about what the law demands
of us today, we do not pretend to foretell the future and we
proceed well aware of the potential for cost and conflict
around jurisdictional boundaries, especially ones that have
gone unappreciated for so long. But it is unclear why pes-
simism should rule the day. With the passage of time, Ok-
lahoma and its Tribes have proven they can work success-
fully together as partners. Already, the State has
negotiated hundreds of intergovernmental agreements
with tribes, including many with the Creek. See Okla.
Stat., Tit. 74, §1221 (2019 Cum. Supp.); Oklahoma Secre-
tary of State, Tribal Compacts and Agreements,
www.sos.ok.gov/tribal.aspx. These agreements relate to
taxation, law enforcement, vehicle registration, hunting
and fishing, and countless other fine regulatory questions.
See Brief for Tom Cole et al. as Amici Curiae 13–19. No one
before us claims that the spirit of good faith, “comity and
42 MCGIRT v. OKLAHOMA

Opinion of the Court

cooperative sovereignty” behind these agreements, id., at
20, will be imperiled by an adverse decision for the State
today any more than it might be by a favorable one.16 And,
of course, should agreement prove elusive, Congress re-
mains free to supplement its statutory directions about the
lands in question at any time. It has no shortage of tools at
its disposal.
*
The federal government promised the Creek a reserva-
tion in perpetuity. Over time, Congress has diminished
that reservation. It has sometimes restricted and other
times expanded the Tribe’s authority. But Congress has
never withdrawn the promised reservation. As a result,
many of the arguments before us today follow a sadly famil-
iar pattern. Yes, promises were made, but the price of keep-
ing them has become too great, so now we should just cast
a blind eye. We reject that thinking. If Congress wishes to
withdraw its promises, it must say so. Unlawful acts, per-
formed long enough and with sufficient vigor, are never
enough to amend the law. To hold otherwise would be to
elevate the most brazen and longstanding injustices over
the law, both rewarding wrong and failing those in the
right.
The judgment of the Court of Criminal Appeals of Okla-
homa is
Reversed.

——————
16 This sense of cooperation and a shared future is on display in this

very case. The Creek Nation is supported by an array of leaders of other
Tribes and the State of Oklahoma, many of whom had a role in negotiat-
ing exactly these agreements. See Brief for Tom Cole et al. as Amici
Curiae 1 (“Amici are a former Governor, State Attorney General, cabinet
members, and legislators of the State of Oklahoma, and two federally
recognized Indian tribes, the Chickasaw Nation and Choctaw Nation of
Oklahoma”) (brief authored by Robert H. Henry, also a former State At-
torney General and Chief Judge of the Tenth Circuit).
 Cite as: 591 U. S. ____ (2020) 1

ROBERTS, C. J., dissenting

SUPREME COURT OF THE UNITED STATES
_________________

No. 18–9526
_________________

JIMCY MCGIRT, PETITIONER v. OKLAHOMA
ON WRIT OF CERTIORARI TO THE COURT OF CRIMINAL
APPEALS OF OKLAHOMA
[July 9, 2020]

CHIEF JUSTICE ROBERTS, with whom JUSTICE ALITO and
JUSTICE KAVANAUGH join, and with whom JUSTICE
THOMAS joins except as to footnote 9, dissenting.
In 1997, the State of Oklahoma convicted petitioner
Jimcy McGirt of molesting, raping, and forcibly sodomizing
a four-year-old girl, his wife’s granddaughter. McGirt was
sentenced to 1,000 years plus life in prison. Today, the
Court holds that Oklahoma lacked jurisdiction to prosecute
McGirt—on the improbable ground that, unbeknownst to
anyone for the past century, a huge swathe of Oklahoma is
actually a Creek Indian reservation, on which the State
may not prosecute serious crimes committed by Indians like
McGirt. Not only does the Court discover a Creek reserva-
tion that spans three million acres and includes most of the
city of Tulsa, but the Court’s reasoning portends that there
are four more such reservations in Oklahoma. The redis-
covered reservations encompass the entire eastern half of
the State—19 million acres that are home to 1.8 million peo-
ple, only 10%–15% of whom are Indians.
Across this vast area, the State’s ability to prosecute se-
rious crimes will be hobbled and decades of past convictions
could well be thrown out. On top of that, the Court has pro-
foundly destabilized the governance of eastern Oklahoma.
The decision today creates significant uncertainty for the
State’s continuing authority over any area that touches In-
dian affairs, ranging from zoning and taxation to family and
2 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

environmental law.
None of this is warranted. What has gone unquestioned
for a century remains true today: A huge portion of Okla-
homa is not a Creek Indian reservation. Congress disestab-
lished any reservation in a series of statutes leading up to
Oklahoma statehood at the turn of the 19th century. The
Court reaches the opposite conclusion only by disregarding
the “well settled” approach required by our precedents. Ne-
braska v. Parker, 577 U. S. 481, ___ (2016) (slip op., at 5).
Under those precedents, we determine whether Congress
intended to disestablish a reservation by examining the rel-
evant Acts of Congress and “all the [surrounding] circum-
stances,” including the “contemporaneous and subsequent
understanding of the status of the reservation.” Id., at ___
(slip op., at 6) (internal quotation marks omitted). Yet the
Court declines to consider such understandings here, pre-
ferring to examine only individual statutes in isolation.
Applying the broader inquiry our precedents require, a
reservation did not exist when McGirt committed his
crimes, so Oklahoma had jurisdiction to prosecute him. I
respectfully dissent.
I
The Creek Nation once occupied what is now Alabama
and Georgia. In 1832, the Creek were compelled to cede
these lands to the United States in exchange for land in pre-
sent day Oklahoma. The expanse set aside for the Creek
and the other Indian nations that composed the “Five Civi-
lized Tribes”—the Cherokees, Chickasaws, Choctaws, and
Seminoles—became known as Indian Territory. See F. Co-
hen, Handbook of Federal Indian Law §4.07(1)(a), pp. 289–
290 (N. Newton ed. 2012) (Cohen). Each of the Five Tribes
formed a tripartite system of government. See Marlin v.
Lewallen, 276 U. S. 58, 60 (1928). They “enact[ed] and ex-
ecut[ed] their own laws,” “punish[ed] their own criminals,”
and “rais[ed] and expend[ed] their own revenues.” Atlantic
 Cite as: 591 U. S. ____ (2020) 3

ROBERTS, C. J., dissenting

& Pacific R. Co. v. Mingus, 165 U. S. 413, 436 (1897).
The Five Tribes also enjoyed unique property rights.
While many tribes held only a “right of occupancy” on lands
owned by the United States, United States v. Creek Nation,
295 U. S. 103, 109 (1935), each of the Five Tribes possessed
title to its lands in communal fee simple, meaning the lands
were “considered the property of the whole.” E.g., Treaty
with the Creeks, Arts. III and IV, Feb. 14, 1833, 7 Stat. 419;
see Marlin, 276 U. S., at 60. Congress promised the Tribes
that their lands would never be “included within, or an-
nexed to, any Territory or State,” see, e.g., Treaty with
Creeks and Seminoles, Art. IV, Aug. 7, 1856, 11 Stat. 700
(1856 Treaty), and that their new homes would be “forever
secure,” Indian Removal Act, §3, 4 Stat. 412; see also Treaty
with the Creeks, Arts. I and XIV, Mar. 24, 1832, 7 Stat. 368.
Forever, it turns out, did not last very long, because the
Civil War disrupted both relationships and borders. The
Five Tribes, whose members collectively held at least 8,000
slaves, signed treaties of alliance with the Confederacy and
contributed forces to fight alongside Rebel troops. See Gib-
son, Native Americans and the Civil War, 9 Am. Indian Q.
4, 385, 388–389, 393 (1985); Doran, Negro Slaves of the Five
Civilized Tribes, 68 Annals Assn. Am. Geographers 335,
346–347, and Table 3 (1978); Cohen §4.07(1)(a), at 289. Af-
ter the war, the United States and the Tribes formed new
treaties, which required each Tribe to free its slaves and
allow them to become tribal citizens. E.g., Treaty with the
Creek Indians, Art. II, June 14, 1866, 14 Stat. 786 (1866
Treaty); see Cohen §4.07(1)(a), at 289, and n. 9. The trea-
ties also stated that the Tribes had “ignored their allegiance
to the United States” and “unsettled the [existing] treaty
relations,” thereby rendering themselves “liable to forfeit”
all “benefits and advantages enjoyed by them”—including
their lands. E.g., 1866 Treaty, Preamble, 14 Stat. 785. Due
to “said liabilities,” the treaties departed from prior prom-
ises and required each Tribe to give up the “west half ” of its
4 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

“entire domain.” E.g., Preamble and Art. III, id., at 785–
786. These western lands became the Oklahoma Territory.
As before, the new treaties promised that the reduced In-
dian Territory would be “forever set apart as a home” for
the Tribes. E.g., Art. III, id., at 786.1
Again, however, it was not to last. In the wake of the war,
a renewed “determination to thrust the nation westward”
gripped the country. Cohen §1.04, at 71. Spurred by new
railroads and protected by the repurposed Union Army, set-
tlers rapidly transformed vast stretches of territorial wil-
derness into farmland and ranches. See id., at 71–74. The
Indian Territory was no exception. By 1900, over 300,000
settlers had poured in, outnumbering members of the Five
Tribes by over 3 to 1. See H. R. Rep. No. 1762, 56th Cong.,
1st Sess., 1 (1900). There to stay, the settlers founded
“[f]lourishing towns” along the railway lines that crossed
the territory. S. Rep. No. 377, 53d Cong., 2d Sess., 6 (1894).
Coexistence proved complicated. The new towns had no
municipal governments or the things that come with
them—laws, taxes, police, and the like. See H. R. Doc.
No. 5, 54th Cong., 1st Sess., 89 (1895). No one had mean-
ingful access to private property ownership, as the unique
communal titles of the Five Tribes precluded ownership by
Indians and non-Indians alike. Despite the millions of dol-
lars that had been invested in the towns and farmlands,
residents had no durable claims to their improvements.
Ibid. Members of the Tribes were little better off, as the

——————
1 I assume that the Creek Nation’s territory constituted a “reservation”

at this time. See ante, at 5–6. The State contends that no reservation
existed in the first place because the territory instead constituted a “de-
pendent Indian communit[y].” Brief for Respondent 8 (quoting 18
U. S. C. §1151(b)). The United States disagrees and states that defining
the territory as a dependent Indian community could disrupt the appli-
cation of various federal statutes. Tr. of Oral Arg. 79–80. I do not ad-
dress this debate because, regardless, I conclude that any reservation
was disestablished.
 Cite as: 591 U. S. ____ (2020) 5

ROBERTS, C. J., dissenting

Tribes failed to hold the communal lands for the “equal ben-
efit” of all members. Woodward v. De Graffenried, 238 U. S.
284, 297 (1915). Instead, a few “enterprising citizens” of the
Tribes “appropriate[d] to their exclusive use almost the en-
tire property of the Territory that could be rendered profit-
able.” Id., at 297, 299, n. 1 (internal quotation marks omit-
ted). As a result, “the poorer class of Indians [were] unable
to secure enough lands for houses and farms,” and “the
great body of the tribe derive[d] no more benefit from their
title than the neighbors in Kansas, Arkansas, or Missouri.”
Id., at 299–301, n. 1 (emphasis deleted; internal quotation
marks omitted).
Attuned to these new realities, Congress decided that it
could not maintain an Indian Territory predicated on “ex-
clusion of the Indians from the whites.” S. Rep. No. 377, at
6. Congress therefore set about transforming the Indian
Territory into a State.
Congress began by establishing a uniform body of law ap-
plicable to all occupants of the territory, regardless of race.
To apply these laws, Congress established the U. S. Courts
for the Indian Territory. Next Congress systematically dis-
mantled the tribal governments. It abolished tribal courts,
hollowed out tribal lawmaking power, and stripped tribal
taxing authority. Congress also eliminated the foundation
of tribal sovereignty, extinguishing the Creek Nation’s title
to the lands. Finally, Congress made the tribe members
citizens of the United States and incorporated them in the
drafting and ratification of the constitution for their new
State, Oklahoma.
In taking these transformative steps, Congress made no
secret of its intentions. It created a commission tasked with
extinguishing the Five Tribes’ territory and, in one report
after another, explained that it was creating a homogenous
population led by a common government. That contempo-
raneous understanding was shared by the tribal leadership
6 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

and the State of Oklahoma. The tribal leadership acknowl-
edged that its only remaining power was to parcel out the
last of its land, and the State assumed jurisdiction over
criminal cases that, if a reservation had continued to exist,
would have belonged in federal court.
A century of practice confirms that the Five Tribes’ prior
domains were extinguished. The State has maintained un-
questioned jurisdiction for more than 100 years. Tribe
members make up less than 10%–15% of the population of
their former domain, and until a few years ago the Creek
Nation itself acknowledged that it no longer possessed the
reservation the Court discovers today. This on-the-ground
reality is enshrined throughout the U. S. Code, which re-
peatedly terms the Five Tribes’ prior holdings the “former”
Indian reservations in Oklahoma. As the Tribes, the State,
and Congress have recognized from the outset, those “res-
ervations were destroyed” when “Oklahoma entered the
Union.” S. Rep. No. 101–216, pt. 2, p. 47 (1989).
II
Much of this important context is missing from the
Court’s opinion, for the Court restricts itself to viewing each
of the statutes enacted by Congress in a vacuum. That ap-
proach is wholly inconsistent with our precedents on reser-
vation disestablishment, which require a highly contextual
inquiry. Our “touchstone” is congressional “purpose” or “in-
tent.” South Dakota v. Yankton Sioux Tribe, 522 U. S. 329,
343 (1998). To “decipher Congress’ intention” in this spe-
cialized area, we are instructed to consider three categories
of evidence: the relevant Acts passed by Congress; the con-
temporaneous understanding of those Acts and the histori-
cal context surrounding their passage; and the subsequent
understanding of the status of the reservation and the pat-
tern of settlement there. Solem v. Bartlett, 465 U. S. 463,
470–472 (1984). The Court resists calling these “steps,” be-
 Cite as: 591 U. S. ____ (2020) 7

ROBERTS, C. J., dissenting

cause “the only ‘step’ proper for a court of law” is interpret-
ing the laws enacted by Congress. Ante, at 17–18. Any la-
bel is fine with us. What matters is that these are catego-
ries of evidence that our precedents “direct[ ] us” to examine
in determining whether the laws enacted by Congress dis-
established a reservation. Hagen v. Utah, 510 U. S. 399,
410–411 (1994). Because those precedents are not followed
by the Court today, it is necessary to describe several at
length.2
In Solem v. Bartlett, 465 U. S. 463 (1984), a unanimous
Court summarized the appropriate methodology. “Con-
gress [must] clearly evince an intent to change boundaries
before diminishment will be found.” Id., at 470 (internal
quotation marks and alterations omitted). This inquiry
first considers the “statutory language used to open the In-
dian lands,” which is the “most probative evidence of con-
gressional intent.” Ibid. “Explicit reference to cession or
other language evidencing the present and total surrender
of all tribal interests strongly suggests that Congress meant
to divest from the reservation all unallotted opened lands.”
Ibid. But “explicit language of cession and unconditional
compensation are not prerequisites” for a finding of dises-
tablishment. Id., at 471.
Second, we consider “events surrounding the passage of

——————
2 Our precedents have generally considered whether Congress dises-

tablished or diminished a reservation by enacting “surplus land Acts”
that opened land to non-Indian settlement. Here Congress did much
more than that, as I will explain. Even so, there is broad agreement
among the parties, the United States, the Creek Nation, and even the
Court that our precedents on surplus land Acts provide the governing
framework for this case, so I proceed on the same course. See Brief for
Petitioner 1; Brief for Respondent 29, 35, 40; Brief for United States as
Amicus Curiae 4–5; Brief for Muscogee (Creek) Nation as Amicus Curiae
1–2; ante, at 7–8, 18–19.
8 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

[an] Act—particularly the manner in which the transaction
was negotiated with the tribes involved and the tenor of leg-
islative Reports presented to Congress.” Ibid. When such
materials “unequivocally reveal a widely held, contempora-
neous understanding that the affected reservation would
shrink as a result of the proposed legislation,” we will “infer
that Congress shared the understanding that its action
would diminish the reservation,” even in the face of “statu-
tory language that would otherwise suggest reservation
boundaries remained unchanged.” Ibid.
Third, to a “lesser extent,” we examine “events that oc-
curred after the passage of [an] Act to decipher Congress’
intentions.” Ibid. “Congress’ own treatment of the affected
areas, particularly in the years immediately following the
opening, has some evidentiary value, as does the manner in
which the Bureau of Indian Affairs and local judicial au-
thorities dealt with [the areas].” Ibid. In addition, “we have
recognized that who actually moved onto opened reserva-
tion lands is also relevant.” Ibid. “Where non-Indian set-
tlers flooded into the opened portion of a reservation and
the area has long since lost its Indian character, we have
acknowledged that de facto, if not de jure, diminishment
may have occurred.” Ibid. This “subsequent demographic
history” provides an “additional clue as to what Congress
expected would happen.” Id., at 471–472.
Fifteen years later, another unanimous Court described
the same methodology more pithily in South Dakota v.
Yankton Sioux Tribe, 522 U. S. 329 (1998). First, the Court
reiterated that the “most probative evidence of diminish-
ment is, of course, the statutory language.” Id., at 344 (in-
ternal quotation marks omitted). The Court continued that
it would also consider, second, “the historical context sur-
rounding the passage of the . . . Acts,” and third, “the sub-
sequent treatment of the area in question and the pattern
of settlement there.” Ibid. (quoting Hagen, 510 U. S., at
411).
 Cite as: 591 U. S. ____ (2020) 9

ROBERTS, C. J., dissenting

The Court today treats these precedents as aging relics
in need of “clarif[ication].” Ante, at 19. But these prece-
dents have been clear enough for some time. Just a few
Terms ago, the same inquiry was described as “well settled”
by the unanimous Court in Nebraska v. Parker, 577 U. S.
481, ___ (2016) (slip op., at 5). First, the Court explained,
“we start with the statutory text.” Ibid. “Under our prece-
dents,” the Court continued, “we also ‘examine all the cir-
cumstances surrounding the opening of a reservation.’ ” Id.,
at ___ (slip op., at 6) (quoting Hagen, 510 U. S., at 412).
Thus, second and third, we “look to any unequivocal evi-
dence of the contemporaneous and subsequent understand-
ing of the status of the reservation by members and non-
members, as well as the United States and the State.” 577
U. S., at ___ (slip op., at 6) (internal quotation marks omit-
ted). These inquiries include, respectively, the “history sur-
rounding the passage of the [relevant] Act” as well as the
subsequent “demographic history” and “treatment” of the
lands at issue. Id., at ___, ___ (slip op., at 8, 10).
Today the Court does not even discuss the governing ap-
proach reiterated throughout these precedents. The Court
briefly recites the general rule that disestablishment re-
quires clear congressional “intent,” ante, at 8, but the Court
then declines to examine the categories of evidence that our
precedents demand we consider. Instead, the Court argues
at length that allotment alone is not enough to disestablish
a reservation. Ante, at 8–12. Then the Court argues that
the “many” “serious blows” dealt by Congress to tribal gov-
ernance, and the creation of the new State of Oklahoma, are
each insufficient for disestablishment. Ante, at 13–16.
Then the Court emphasizes that “historical practices or cur-
rent demographics” do not “by themselves” “suffice” to dis-
establish a reservation. Ante, at 17–18.
This is a school of red herrings. No one here contends
that any individual congressional action or piece of evi-
10 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

dence, standing alone, disestablished the Creek reserva-
tion. Rather, Oklahoma contends that all of the relevant
Acts of Congress together, viewed in light of contemporane-
ous and subsequent contextual evidence, demonstrate Con-
gress’s intent to disestablish the reservation. “[O]ur tradi-
tional approach . . . requires us” to determine Congress’s
intent by “examin[ing] all the circumstances surrounding
the opening of a reservation.” Hagen, 510 U. S., at 412 (em-
phasis added). Yet the Court refuses to confront the cumu-
lative import of all of Congress’s actions here.
The Court instead announces a new approach sharply re-
stricting consideration of contemporaneous and subsequent
evidence of congressional intent. The Court states that
such “extratextual sources” may be considered in “only” one
narrow circumstance: to help “ ‘clear up’ ” ambiguity in a
particular “statutory term or phrase.” Ante, at 17–18, 20
(quoting Milner v. Department of Navy, 562 U. S. 562, 574
(2011), and citing New Prime Inc. v. Oliveira, 586 U. S. ___,
___ (2019) (slip op., at 6)).
But, if that is the right approach, what have we been do-
ing all these years? Every single one of our disestablish-
ment cases has considered extratextual sources, and in do-
ing so, none has required the identification of ambiguity in
a particular term. That is because, while it is well estab-
lished that Congress’s “intent” must be “clear,” ante, at 20
(quoting Yankton Sioux Tribe, 522 U. S., at 343), in this
area we have expressly held that the appropriate inquiry
does not focus on the statutory text alone.
Today the Court suggests that only the text can satisfy
the longstanding requirement that Congress “explicitly in-
dicate[ ]” its intent. Ante, at 20 (quoting Solem, 465 U. S.,
at 470). The Court reiterates that a reservation persists
unless Congress “said otherwise,” ante, at 1; if Congress
wishes to disestablish a reservation, “it must say so,” with
the right “language.” Ante, at 8, 18; see ante, at 42 (same).
 Cite as: 591 U. S. ____ (2020) 11

ROBERTS, C. J., dissenting

Our precedents disagree. They explain that disestablish-
ment can occur “[e]ven in the absence of a clear expression
of congressional purpose in the text of [the] Act.” Yankton
Sioux Tribe, 522 U. S., at 351. The “notion” that “express
language in an Act is the only method by which congres-
sional action may result in disestablishment” is “quite in-
consistent” with our precedents. Rosebud Sioux Tribe v.
Kneip, 430 U. S. 584, 586, 588, n. 4 (1977); see Solem, 465
U. S., at 471 (intent may be discerned from a “widely held,
contemporaneous understanding,” “notwithstanding the
presence of statutory language that would otherwise sug-
gest reservation boundaries remained unchanged”); see
also DeCoteau v. District County Court for Tenth Judicial
Dist., 420 U. S. 425, 444 (1975); Mattz v. Arnett, 412 U. S.
481, 505 (1973).
These are not “stiche[d] together quotes” but rather plain
language reflecting a consistent theme running through our
precedents. Ante, at 20, n. 9. They make clear that the
Court errs in focusing on whether “a statute” alone “re-
quired” disestablishment, ante, at 20; under these prece-
dents, we cannot determine what Congress “required” with-
out first considering evidence in addition to the relevant
statutes. Oddly, the Court claims these precedents actually
support its new approach because they “emphasize that
‘[t]he focus of our inquiry is congressional intent.’ ” Ante, at
20–21, n. 9 (quoting Rosebud Sioux Tribe, 430 U. S., at 588,
n. 4, and citing Yankton Sioux Tribe, 522 U. S., at 343). But
in this context that intent is determined by examining a
broad array of evidence—“all the circumstances.” Parker,
577 U. S., at ___ (slip op., at 6) (quoting Hagen, 510 U. S.,
at 412). Unless the Court is prepared to overrule these
precedents, it should follow them.
The Court appears skeptical of these precedents, but does
not address the compelling reasons they give for consider-
ing extratextual evidence. At the turn of the century, the
possibility that a reservation might persist in the absence
12 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

of “tribal ownership” of the underlying lands was “unfamil-
iar,” and the prevailing “assumption” was that “Indian res-
ervations were a thing of the past.” Solem, 465 U. S., at
468. Congress believed “to a man” that “within a short
time” the “Indian tribes would enter traditional American
society and the reservation system would cease to exist.”
Ibid. As a result, Congress—while intending disestablish-
ment—did not always “detail” precise changes to reserva-
tion boundaries. Ibid. Recognizing this distinctive back-
drop, our precedents determine Congress’s intent by
considering a broader variety of evidence than we might for
more run-of-the-mill questions of statutory interpretation.
See id., at 468–469; Parker, 577 U. S., at ___ (slip op., at 6);
Yankton Sioux Tribe, 522 U. S., at 343. See also Cohen
§2.02(1), at 113 (“The theory and practice of interpretation
in federal Indian law differs from that of other fields of
law.”).
The Court next claims that Parker “clarif[ied]” that evi-
dence of the subsequent treatment of the disputed land by
government officials “ ‘has limited interpretive value.’ ”
Ante, at 19 (quoting Parker, 577 U. S., at ___ (slip op., at
11)). But Parker held that the subsequent evidence in that
case “ha[d] ‘limited interpretive value,’ ” as in the case that
Parker relied on. 577 U. S., at ___–___ (slip op., at 11–12)
(quoting Yankton Sioux Tribe, 522 U. S., at 355). The ade-
quacy of evidence in a particular case says nothing about
whether our precedents require us to consider such evi-
dence in others.3

——————
3 The Court rejects this reading of Parker based on a quotation that

ends with what sounds like a general principle that “[e]vidence of the
subsequent treatment of the disputed land by Government officials like-
wise has ‘limited interpretive value.’ ” Ante, at 19, n. 8 (quoting Parker,
577 U. S., at ___ (slip op., at 11)). But that sentence was actually the
topic sentence of a new paragraph that addressed the particular evidence
of subsequent treatment of the particular land by the particular govern-
ment officials in that case. Id., at ___–___ (slip op., at 11–12). It is clear
 Cite as: 591 U. S. ____ (2020) 13

ROBERTS, C. J., dissenting

The Court finally resorts to torching strawmen. No one
relying on our precedents contends that “practical ad-
vantages” require “ignoring the written law.” Ante, at 27.
No one claims a State has “authority to reduce federal res-
ervations.” Ante, at 7. No one says the role of courts is to
“sav[e] the political branches” from “embarrassment.” Ibid.
No one argues that courts can “adjust[ ]” reservation bor-
ders. Ibid. Such notions have nothing to do with our prec-
edents. What our precedents do provide is the settled ap-
proach for determining whether Congress disestablished a
reservation, and the Court starkly departs from that ap-
proach here.
III
Applied properly, our precedents demonstrate that Con-
gress disestablished any reservation possessed by the
Creek Nation through a relentless series of statutes leading
up to Oklahoma statehood.
A
The statutory texts are the “most probative evidence” of
congressional intent. Parker, 577 U. S., at ___ (slip op., at
5) (quoting Hagen, 510 U. S., at 411). The Court appropri-
ately examines the Original Creek Agreement of 1901 and
a subsequent statute for language of disestablishment, such
as “cession,” “abolish[ing]” the reservation, “restor[ing]”
land to the “public domain,” or an “unconditional commit-
ment” to “compensate” the Tribe. Ante, at 8–12 (internal
quotation marks omitted). But that is only the beginning

——————
that Parker merely concluded that the evidence cited by the parties pro-
vided a “mixed record of subsequent treatment” that did not move the
needle either way. Ibid. (internal quotation marks omitted). Parker did
not silently overturn our precedents requiring us to consider—and accord
“weight” to—subsequent evidence that plainly favors, or undermines,
disestablishment. Rosebud Sioux Tribe v. Kneip, 430 U. S. 584, 604
(1977); see supra, at 6–9.
14 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

of the analysis; there is no “magic words” requirement for
disestablishment, and each individual statute may not be
considered in isolation. See supra, at 10–11; Hagen, 510
U. S., at 411, 415–416 (when two statutes “buil[d]” on one
another in this area, “[both] statutes—as well as those that
came in between—must therefore be read together”); see
also Rosebud Sioux Tribe, 430 U. S., at 592 (recognizing
that a statute “cannot, and should not, be read as if it were
the first time Congress had addressed itself to” disestab-
lishment when prior statutes also indicate congressional in-
tent). In this area, “we are not free to say to Congress: ‘We
see what you are driving at, but you have not said it, and
therefore we shall go on as before.’ ” Id., at 597 (quoting
Johnson v. United States, 163 F. 30, 32 (CA1 1908) (Holmes,
J.)). Rather, we recognize that the language Congress uses
to accomplish its objective is adapted to the circumstances
it confronts.
For example, “cession” is generally what a tribe does
when it conveys land to a fellow sovereign, such as the
United States or another tribe. See Mitchel v. United
States, 9 Pet. 711, 734 (1835); e.g., 1856 Treaty, Art. I, 11
Stat. 699. But here, given that Congress sought direct al-
lotment to tribe members in order to enable private owner-
ship by both Indians and the 300,000 settlers in the terri-
tory, it would have made little sense to “cede” the lands to
the United States or “restore” the lands to the “public do-
main,” as Congress did on other occasions. So too with a
“commitment” to “compensate” the Tribe. Rather than buy-
ing land from the Creek, Congress provided for allotment to
tribe members who could then “sell their land to Indians
and non-Indians alike.” Ante, at 10; see Hagen, 510 U. S.,
at 412 (a “definite payment” is not required for disestablish-
ment). That other allotment statutes have contained vari-
ous “hallmarks” of disestablishment tells us little about
Congress’s intent here. Contra, ante, at 12–13, and n. 5.
“[W]e have never required any particular form of words” to
 Cite as: 591 U. S. ____ (2020) 15

ROBERTS, C. J., dissenting

disestablish a reservation. Hagen, 510 U. S., at 411. There
are good reasons the statutes here do not include the lan-
guage the Court looks for, and those reasons have nothing
to do with a failure to disestablish the reservation. Respect
for Congress’s work requires us to look at what it actually
did, not search in vain for what it might have done or did
on other occasions.
What Congress actually did here was enact a series of
statutes beginning in 1890 and culminating with Oklahoma
statehood that (1) established a uniform legal system for
Indians and non-Indians alike; (2) dismantled the Creek
government; (3) extinguished the Creek Nation’s title to the
lands at issue; and (4) incorporated the Creek members into
a new political community—the State of Oklahoma. These
statutes evince Congress’s intent to terminate the reserva-
tion and create a new State in its place.
First, Congress supplanted the Creek legal system with
a legal code and court system that applied equally to Indi-
ans and non-Indians. In 1890, Congress subjected the In-
dian Territory to specified federal criminal laws. Act of May
2, 1890, §31, 26 Stat. 96. For offenses not covered by federal
law, Congress did what it often did when establishing a new
territorial government. It provided that the criminal laws
from a neighboring State, here Arkansas, would apply. §33,
id., at 96–97. Seven years later, Congress provided that the
laws of the United States and Arkansas “shall apply to all
persons” in Indian Territory, “irrespective of race.” Act of
June 7, 1897 (1897 Act), 30 Stat. 83 (emphasis added). In
the same Act, Congress conferred on the U. S. Courts for
the Indian Territory “exclusive jurisdiction” over “all civil
causes in law and equity” and “all criminal causes” for the
punishment of offenses committed by “any person” in the
Indian Territory. Ibid.
The following year, the 1898 Curtis Act “abolished” all
tribal courts, prohibited all officers of such courts from ex-
16 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

ercising “any authority” to perform “any act” previously au-
thorized by “any law,” and transferred “all civil and crimi-
nal causes then pending” to the U. S. Courts for the Indian
Territory. Act of June 27, 1898 (Curtis Act), §28, id., at
504–505. In the same Act, Congress completed the shift to
a uniform legal order by banning the enforcement of tribal
law in the newly exclusive jurisdiction of the U. S. Courts.
See §26, id., at 504 (“[T]he laws of the various tribes or na-
tions of Indians shall not be enforced at law or in equity by
the courts of the United States in the Indian Territory.”).
Congress reiterated yet again in 1904 that Arkansas law
“continued” to “embrace all persons and estates” in the ter-
ritory—“whether Indian, freedmen, or otherwise.” Act of
Apr. 28, 1904, ch. 1824, §2, 33 Stat. 573 (emphasis added).
In this way, Congress replaced tribal law with local law in
matters at the core of tribal governance, such as inheritance
and marital disputes. See, e.g., George v. Robb, 4 Ind. T. 61,
64 S. W. 615, 615–616 (1901); Colbert v. Fulton, 74 Okla.
293, 157 P. 1151, 1152 (1916).
In addition, the Curtis Act established municipalities to
govern both Indians and non-Indians. It authorized “any
city or town” with at least 200 residents to incorporate. §14,
30 Stat. 499. The Act gave incorporated towns “all the pow-
ers” and “all the rights” of municipalities under Arkansas
law. Ibid. “All male inhabitants,” including Indians, were
deemed qualified to vote in town elections. Ibid. And “all
inhabitants”—“without regard to race”—were made subject
to “all” town laws and were declared to possess “equal
rights, privileges, and protection.” Id., at 499–500 (empha-
sis added). These changes reorganized the approximately
150 towns in the territory—including Tulsa, Muskogee, and
23 others within the Creek Nation’s former territory—that
were home to tens of thousands of people and nearly one
third of the territory’s population at the time, laying the
foundation for the state governance that was to come. See
H. R. Doc. No. 5, 57th Cong., 2d Sess., pt. 2, pp. 299–300,
 Cite as: 591 U. S. ____ (2020) 17

ROBERTS, C. J., dissenting

Table 1 (1903); Depts. of Commerce and Labor, Bureau of
Census, Population of Oklahoma and Indian Territory
1907, pp. 8, 30–33.
Second, Congress systematically dismantled the govern-
mental authority of the Creek Nation, targeting all three
branches. As noted, Congress dissolved the Tribe’s judicial
system. Congress also specified in the Original Creek
Agreement that the Creek government would “not con-
tinue” past March 1906, essentially preserving it only as
long as Congress thought necessary for the Tribe to wind
up its affairs. §46, 31 Stat. 872. In the meantime, Congress
radically curtailed tribal legislative authority, providing
that no statute passed by the council of the Creek Nation
affecting the Nation’s lands, money, or property would be
valid unless approved by the President of the United States.
§42, id., at 872. When 1906 came around, the Five Tribes
Act provided for the “final disposition of the affairs of the
Five Civilized Tribes.” Act of Apr. 26, 1906, ch. 1876, 34
Stat. 137. Along with “abolish[ing]” all tribal taxes, the Act
directed the Secretary of the Interior to assume control over
the collection of the Nation’s remaining revenues and to dis-
tribute them among tribe members on a per capita basis.
§§11, 17, id., at 141, 143–144. Thus, by the time Oklahoma
became the 46th State in 1907, there was little left of the
Creek Nation’s authority: No tribal courts. No tribal law.
No tribal fisc. And any lingering authority was further re-
duced in 1908, when Congress amended the Five Tribes Act
to require tribal officers and members to surrender all re-
maining tribal property, money, and records. Act of May
27, 1908, §13, 35 Stat. 316.
The Court stresses that the Five Tribes Act separately
stated that the Creek government was “continued” in “full
force and effect for all purposes authorized by law.” Ante,
at 15 (quoting §28, 34 Stat. 148). By that point, however,
such “authorized” purposes were nearly nonexistent, and
the Act’s statement is readily explained by the need to
18 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

maintain a tribal body to wrap up the distribution of Creek
lands. Indeed, the Court does not cite any examples of the
Creek Nation exercising significant government authority
in the wake of the statutes discussed above. Instead, the
Court alludes to subsequent changes in the 1920s to the
general “federal outlook towards Native Americans,” and it
observes that in the 1930s Congress authorized the Creek
Nation to reconstitute its tribal courts and adopt a consti-
tution and bylaws. Ante, at 15. That, however, simply high-
lights the drastic extent to which Congress erased the Na-
tion’s authority at the turn of the century.
Third, Congress destroyed the foundation of sovereignty
by stripping the Creek Nation of its territory. The commu-
nal title held by the Creek Nation, which “did not recognize
private property in land,” “presented a serious obstacle to
the creation of [a] State.” Choate v. Trapp, 224 U. S. 665,
667 (1912). Well aware of this impediment, Congress es-
tablished the Dawes Commission and directed it to negoti-
ate with the Five Tribes for “the extinguishment of the na-
tional or tribal title to any lands” within the Indian
Territory. Act of Mar. 3, 1893, §16, 27 Stat. 645. That ex-
tinguishment could be accomplished through “cession” of
the tribal lands to the United States, “allotment” of the
lands among the Indians, or any other agreed upon method.
Ibid. The Commission initially sought cession, but ulti-
mately sought to extinguish the title through allotment.
See ante, at 9.
In the Original Creek Agreement of 1901, Congress did
just that. The agreement provided that “[a]ll lands belong-
ing to the Creek tribe,” except town sites and lands reserved
for schools and public buildings, “shall be allotted among
the citizens of the tribe.” §§2, 3, 31 Stat. 862 (emphasis
added). Town sites, rather than being allotted, were made
available for purchase by the non-Indians residing there.
§§11–16, id., at 866–867. Unclaimed lots were to be sold at
public auction, with the proceeds divvied up among the
 Cite as: 591 U. S. ____ (2020) 19

ROBERTS, C. J., dissenting

Creeks. §§11, 14, id., at 866. The agreement required that
the deeds for the allotments and town site purchases convey
“all right, title, and interest of the Creek Nation and of all
other [Creek] citizens,” and that the deeds be executed by
the leader of the Creek Nation (the “principal chief ”). §23,
id., at 867–868. The conveyances were then approved by
the Secretary of the Interior, who in turn “relinquish[ed] to
the grantee . . . all the right, title, and interest of the United
States” in the land. Id., at 868. In this way, Congress pro-
vided for the complete termination of the Creek Nation’s in-
terest in the lands, as well as the interests of individual
Creek members apart from their personal allotments. In-
deed, the language Congress used in the Original Creek
Agreement resembles what the Court regards as model dis-
establishment language. See ante, at 8, 10 (looking for lan-
guage evincing “the present and total surrender of all tribal
interests in the affected lands” (internal quotation marks
omitted)). And, making even more clear its intent to place
Indian-held land under the same laws as all other property,
Congress subsequently eliminated restrictions on the alien-
ation of allotments, freeing tribe members “to sell their land
to Indians and non-Indians alike.” Ante, at 10.
In addition, while the Original Creek Agreement did not
allot lands reserved for schools and tribal buildings, the
Creek Nation’s interest in those lands was subsequently
terminated by the Five Tribes Act. That Act directed the
Secretary of the Interior to take possession of—and sell
off—“all” tribal buildings and underlying lands, whether
used for “governmental” or “other tribal purposes.” §15, 34
Stat. 143. The Secretary was also ordered to assume con-
trol of all tribal schools and the underlying property until
the federal or state governments established a public school
system. See §10, id., at 140–141.
These statutes evince a clear intent to leave the Creek
Nation with no communally held land and no meaningful
governing authority to exercise over the newly distributed
20 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

parcels. Contrary to the Court’s portrayal, this is not a sce-
nario in which Congress allowed a tribe to “continue to ex-
ercise governmental functions over land” that it “no longer
own[ed] communally.” Ante, at 11. From top to bottom,
these statutes, which divested the Tribes and the United
States of their interests while displacing tribal governance,
“strongly suggest[ ] that Congress meant to divest” the
lands of reservation status. Solem, 465 U. S., at 470.
Finally, having stripped the Creek Nation of its laws, its
powers of self-governance, and its land, Congress incorpo-
rated the Nation’s members into a new political community.
Congress made “every Indian” in the Oklahoma territory a
citizen of the United States in 1901—decades before confer-
ring citizenship on all native born Indians elsewhere in the
country. Act of Mar. 3, 1901, ch. 868, 31 Stat. 1447. In the
Oklahoma Enabling Act of 1906—the gateway to state-
hood—Congress confirmed that members of the Five Tribes
would participate in equal measure alongside non-Indians
in the choice regarding statehood. The Act gave Indians the
right to vote on delegates to a constitutional convention and
ultimately on the state constitution that the delegates pro-
posed. §§2, 4, 34 Stat. 268, 271. Fifteen members of the
Five Tribes were elected as convention delegates, many of
them served on significant committees, and a member of
the Chickasaw Nation even served as president of the con-
vention. See Brief for Seventeen Oklahoma District Attor-
neys et al. as Amici Curiae 9–13.
The Enabling Act also ensured that Indians and non-In-
dians would be subject to uniform laws and courts. It re-
placed Arkansas law, which had applied to all persons “ir-
respective of race,” 1897 Act, 30 Stat. 83, with the laws of
the adjacent Oklahoma Territory until the new state legis-
lature provided otherwise. Enabling Act §§2, 13, 21, 34
Stat. 268–269, 275, 277–278; see Jefferson v. Fink, 247
U. S. 288, 294 (1918). All of the pending cases in the terri-
torial courts arising under federal law were transferred to
 Cite as: 591 U. S. ____ (2020) 21

ROBERTS, C. J., dissenting

the newly created U. S. District Courts of Oklahoma. See
§16, 34 Stat. 276. Pending cases not involving federal law,
including those that involved Indians on Indian land and
had arisen under Arkansas law, were transferred to the
new Oklahoma state courts. §§16, 17, 20, id., at 276–277.
To dispel any potential confusion about the distribution of
criminal cases, Congress amended the Enabling Act the fol-
lowing year, clarifying that all cases for crimes that would
have fallen under federal jurisdiction had they been com-
mitted in a State would be transferred to the U. S. District
Courts. Act of Mar. 4, 1907, §1, id., at 1286–1287. All other
pending criminal cases would be “prosecuted to a final de-
termination in the State courts of Oklahoma.” §3, id., at
1287. As for civil cases, the new state courts were immedi-
ately empowered to resolve even disputes that previously
lay at the core of tribal self-governance. E.g., Palmer v.
Cully, 52 Okla. 454, 463–469, 153 P. 154, 157–158 (1915)
(per curiam) (marital dispute).4
In sum, in statute after statute, Congress made abun-
dantly clear its intent to disestablish the Creek territory.
The Court, for purposes of the disestablishment question
before us, defines the Creek territory as “lands that would
lie outside both the legal jurisdiction and geographic bound-
aries of any State” and on which a tribe was “assured a right
to self-government.” Ante, at 6. That territory was elimi-
nated. By establishing uniform laws for Indians and non-

——————
4 The Court, citing United States v. Sandoval, 231 U. S. 28, 47–48

(1913), argues that including a tribe within a new State is not necessarily
incompatible with the continuing existence of a reservation. Ante, at 15–
16, n. 6. But the tribe in Sandoval, the Pueblo Indians of New Mexico,
retained a rare communal title to their lands—which Congress explicitly
extinguished here. 231 U. S., at 47. More fundamentally, the Court’s
argument suffers from the same flaw that runs through its entire ap-
proach, which maintains that each of Congress’s actions alone would not
be enough for disestablishment but never confronts the import of all of
them.
22 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

Indians alike in the new State of Oklahoma, Congress
brought Creek members and the land on which they resided
under state jurisdiction. By stripping the Creek Nation of
its courts, lawmaking authority, and taxing power, Con-
gress dismantled the tribal government. By extinguishing
the Nation’s title, Congress erased the geographic bounda-
ries that once defined Creek territory. And, by conferring
citizenship on tribe members and giving them a vote in the
formation of the State, Congress incorporated them into a
new political community. “Under any definition,” that was
disestablishment. Ibid.
In the face of all this, the Court claims that recognizing
Congress’s intent would permit disestablishment in the ab-
sence of “a statute requir[ing] that result.” Ante, at 20.
Hardly. The numerous statutes discussed above demon-
strate Congress’s plain intent to terminate the reservation.
The Court resists the cumulative force of these statutes by
attacking each in isolation, first asking whether allotment
alone disestablished the reservation, then whether restrict-
ing tribal governance was sufficient, and so on. But the
Court does not consider the full picture of what Congress
accomplished. Far from justifying its blinkered approach,
the Court repeatedly tells the reader to wait until the “next
section” of the opinion—where the Court will again nitpick
discrete aspects of Congress’s disestablishment effort while
ignoring the full picture our precedents require us to honor.
Ante, at 12–13, n. 5, 17, n. 7; see supra, at 11, 14.
The Court also hypothesizes that Congress may have
taken significant steps toward disestablishment but ulti-
mately could not “complete[ ]” it; perhaps Congress just
couldn’t “muster the will” to finish the job. Ante, at 8, 15.
The Court suggests that Congress sought to “tiptoe to the
edge of disestablishment,” fearing the “embarrassment of
disestablishing a reservation” but hoping that judges would
“deliver the final push.” Ante, at 7. This is fantasy. The
congressional Acts detailed above do not evince any unease
 Cite as: 591 U. S. ____ (2020) 23

ROBERTS, C. J., dissenting

about extinguishing the Creek domain, or any shortage of
“will.” Quite the opposite. Through an open and concerted
effort, Congress did what it set out to do: transform a reser-
vation into a State. “Mustering the broad social consensus
required to pass new legislation is a deliberately hard busi-
ness,” as the Court reminds us. Ibid. Congress did that
hard work here, enacting not one but a steady progression
of major statutes. The Court today does not give effect to
the cumulative significance of Congress’s actions, because
Congress did not use explicit words of the sort the Court
insists upon. But Congress had no reason to suppose that
such words would be required of it, and this Court has held
that they were not. See Hagen, 510 U. S., at 411–412;
Yankton Sioux Tribe, 522 U. S., at 351; Solem, 465 U. S., at
471.
B
Under our precedents, we next consider the contempora-
neous understanding of the statutes enacted by Congress
and the subsequent treatment of the lands at issue. The
Court, however, declines to consider such evidence because,
in the Court’s view, the statutes clearly do not disestablish
any reservation, and there is no “ambiguity” to “clear up.”
Ante, at 20 (internal quotation marks omitted). That is not
the approach demanded by our precedent, supra, at 10–13,
and, in any event, the Court’s argument fails on its own
terms here. I find it hard to see how anyone can come away
from the statutory texts detailed above with certainty that
Congress had no intent to disestablish the territorial reser-
vation. At the very least, the statutes leave some ambigu-
ity, and thus “extratextual sources” ought to be consulted.
Ante, at 20.
Turning to such sources, our precedents direct us to “ex-
amine all the circumstances” surrounding Congress’s ac-
tions. Parker, 577 U. S., at ___ (slip op., at 6) (quoting Ha-
gen, 510 U. S., at 412). This includes evidence of the
24 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

“contemporaneous understanding” of the status of the res-
ervation and the “history surrounding the passage” of the
relevant Acts. Parker, 577 U. S., at ___ (slip op., at 8) (in-
ternal quotation marks omitted); see Yankton Sioux Tribe,
522 U. S., at 351–354; Solem, 465 U. S., at 471. The avail-
able evidence overwhelmingly confirms that Congress elim-
inated any Creek reservation. That was the purpose iden-
tified by Congress, the Dawes Commission, and the Creek
Nation itself. And that was the understanding demon-
strated by the actions of Oklahoma, the United States, and
the Creek.
According to reports published by Congress leading up to
Oklahoma statehood, the Five Tribes had failed to hold the
lands for the equal benefit of all Indians, and the tribal gov-
ernments were ill equipped to handle the largescale settle-
ment of non-Indians in the territories. See supra, at 4–5;
Woodward, 238 U. S., at 296–297. The Senate Select Com-
mittee on the Five Tribes explained that it was “impera-
tive[ ]” to “establish[ ] a government over [non-Indians] and
Indians” in the territory “in accordance with the principles
of our constitution and laws.” S. Rep. No. 377, at 12–13.
On the eve of the Original Creek Agreement, the House
Committee on Indian Affairs emphasized that “[t]he inde-
pendent self-government of the Five Tribes ha[d] practi-
cally ceased,” “[t]he policy of the Government to abolish
classes in Indian Territory and make a homogeneous popu-
lation [wa]s being rapidly carried out,” and all Indians
“should at once be put upon a level and equal footing with
the great population with whom they [were] intermingled.”
H. R. Rep. No. 1188, 56th Cong., 1st Sess., 1 (1900).
The Dawes Commission understood Congress’s intent in
the same way. The Commission explained that the “object
of Congress from the beginning has been the dissolution of
the tribal governments, the extinguishment of the commu-
nal or tribal title to the land, the vesting of possession and
title in severalty among the citizens of the Tribes, and the
 Cite as: 591 U. S. ____ (2020) 25

ROBERTS, C. J., dissenting

assimilation of the peoples and institutions of this Territory
to our prevailing American standard.” H. R. Doc. No. 5,
58th Cong., 2d Sess., pt. 2, p. 5 (1903). Accordingly, the
Commission’s aim—“in all [its] endeavors”—was a “uni-
formity of political institutions to lay the foundation for an
ultimate common government.” H. R. Doc. No. 5, 56th
Cong., 2d Sess., 163 (1900).
The Creek shared the same understanding. In 1893, the
year Congress formed the Dawes Commission, the Creek
delegation to Washington recognized that Congress’s “un-
wavering aim” was to “ ‘wipe out the line of political distinc-
tion between an Indian citizen and other citizens of the Re-
public’ ” so that the Tribe could be “ ‘absorbed and become a
part of the United States.’ ” P. Porter & A. McKellop,
Printed Statement of Creek Delegates, reprinted in Creek
Delegation Documents 8–9 (Feb. 9, 1893) (quoting Senate
Committee Report); see also S. Doc. No. 111, 54th Cong., 2d
Sess., 5, 8 (1897) (resolution of the Creek Nation “recog-
niz[ing]” that Congress proposed to “disintegrat[e] the land
of our people” and “transform[ ]” “our domestic dependent
states” “into a State of the Union”).
Particularly probative is the understanding of Pleasant
Porter, the principal Chief of the Creek Nation. He de-
scribed Congress’s decisions to the Creek people and legis-
lature in messages published in territorial newspapers dur-
ing the run-up to statehood. Following the extinguishment
of the Nation’s title, dissolution of tribal courts, and curtail-
ment of lawmaking authority, he told his people that “[i]t
would be difficult, if not impossible to successfully operate
the Creek government now.” App. to Brief for Respondent
8a (Message to Creek National Council (May 7, 1901), re-
printed in The Indian Journal (May 10, 1901)). The “rem-
nant of a government” had been reduced to a land office for
finalizing the distribution of allotments and would be
“maintained only until” the Tribe’s “landed and other inter-
ests . . . have been settled.” App. to Brief for Respondent
26 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

8a. He reiterated this understanding following the Five
Tribes Act of 1906, which stated that the tribal government
would “continue[ ] in full force and effect for all purposes
authorized by law.” §28, 34 Stat. 148. While the Court be-
lieves that meant Congress decided against disestablishing
the reservation, see ante, at 14–15, Chief Porter saw things
differently. From his vantage point as the contemporane-
ous leader of the government at issue, Congress had tempo-
rarily continued the tribal government but left it with only
“limited and circumscribed” authority: The council could
“pass[ ] resolutions respecting our wishes” regarding the
property “now in the process of distribution,” but the council
no longer had any authority to “mak[e] laws for our govern-
ment.” App. to Brief for Respondent 14a (Message to Creek
National Council (Oct. 18, 1906), reprinted in The New
State Tribune (Oct. 18, 1906)). Apart from distributing the
Nation’s property, Chief Porter maintained that “all powers
over the governing even of our landed property will cease”
once the new state government was established. App. to
Brief for Respondent 15a; see also S. Rep. No. 5013, 59th
Cong., 2d Sess., pt. 1, p. 885 (1907) (Choctaw governor
mourning that his “only” remaining authority was “to sign
deeds”).
The Creek remained of that view after Oklahoma was of-
ficially made a State through the Enabling Act. At that
point, the new principal Chief confirmed that it was “utterly
impossible” to resume “our old tribal government.” App. to
Brief for Respondent 16a–17a (Address by Moty Tiger to
Creek National Council (Oct. 8, 1908), reprinted in The In-
dian Journal (Oct. 9, 1908)). And any “appeal to the gov-
ernment at Washington to alter its purpose to wipe out all
tribal government among the five civilized tribes” would “be
to no purpose.” App. to Brief for Respondent 16a. “[C]on-
tributions” for such efforts would be “just that much money
thrown away,” and “all attorneys at Washington or else-
 Cite as: 591 U. S. ____ (2020) 27

ROBERTS, C. J., dissenting

where who encourage and receive any part of such contri-
butions do it knowing that they can give no return or service
for same and that they take such money fraudulently and
dishonestly.” Id., at 17a.5
In addition to their words, the contemporaneous actions
of Oklahoma, the Creek, and the United States in criminal
matters confirm their shared understanding that Congress
did not intend a reservation to persist. Had the land been
a reservation, the federal government—not the new State—
would have had jurisdiction over serious crimes committed
by Indians under the Major Crimes Act of 1885. See §9, 23
Stat. 385. Yet, at statehood, Oklahoma immediately began
prosecuting serious crimes committed by Indians in the
new state courts, and the federal government immediately
ceased prosecuting such crimes in federal court. At argu-
ment, McGirt’s counsel acknowledged that he could not cite
a single example of federal prosecutions for such crimes.
Tr. of Oral Arg. 17–18. Rather, the record demonstrates
that case after case was transferred to state court or filed
there outright by Oklahoma after 1907—without objection
by anyone. See, e.g., Bigfeather v. State, 7 Okla. Crim. 364,
123 P. 1026 (1912) (manslaughter); Rollen v. State, 7 Okla.
Crim. 673, 125 P. 1087 (1912) (assault with intent to kill);
Jones v. State, 3 Okla. Crim. 593, 107 P. 738 (1910) (mur-
der); see also Brief for Petitioner in Carpenter v. Murphy,
O. T. 2018, No. 17–1107, pp. 40–41 (collecting more cases).

——————
5 The Court discounts the views of the principal chiefs as mere predic-

tions about what Congress “would” do, ante, at 25, but the Court ignores
statements made after statehood, describing what Congress did do. The
Court also asserts that the chiefs’ views cannot serve as “evidence” of the
“meaning” of laws enacted by Congress. Ante, at 25, n. 12. That is in-
consistent with our precedent, which specifically instructs us to deter-
mine Congress’s intent by considering the “understanding of the status
of the reservation by members” of the affected tribe. Parker, 577 U. S.,
at ___ (slip op., at 6). The contemporaneous understanding of the leaders
of the tribe is highly probative.
28 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

These prosecutions were lawful, the Oklahoma Supreme
Court recognized at the time, because Congress had not in-
tended to “except out of [Oklahoma] an Indian reservation”
upon its admission as a State. Higgins v. Brown, 20 Okla.
355, 419, 94 P. 703, 730 (1908).
Instead of explaining how everyone at the time somehow
missed that a reservation still existed, the Court resorts to
misdirection. It observes that Oklahoma state courts have
held that they erroneously entertained prosecutions for
crimes committed by Indians on the small number of re-
maining restricted allotments and tribal trust lands from
the 1930s until 1989. But this Court has not addressed that
issue, and regardless, it would not tell us whether the State
properly prosecuted major crimes committed by Indians on
the lands at issue here—the unrestricted fee lands that
make up more than 95% of the Creek Nation’s former terri-
tory. Perhaps most telling is that the State’s jurisdiction
over crimes on Indian allotments was hotly contested from
an early date, whereas nobody raised objections based on a
surviving reservation. See, e.g., Ex parte Nowabbi, 60 Okla.
Crim. 111, 61 P. 2d 1139 (1936), overruled by State v.
Klindt, 782 P. 2d 401, 404 (Okla. Crim. App. 1989); see also
ante, at 21 (“no court” suggested the “possibility” that “the
Creek lands really were part of a reservation” until 2017).6
Lacking any other arguments, the Court suspects uni-
form lawlessness: The State must have “overstepped its au-
thority” in prosecuting thousands of cases for over a cen-
tury. Ante, at 23. Perhaps, the Court suggests, the State

——————
6 The Court claims that the Oklahoma courts’ reasons for treating re-

stricted allotments as Indian country must apply with “equal force” to
the unrestricted fee lands at issue here, but the Court ultimately admits
the two types of land are “legally distinct.” Ante, at 23, n. 10. And any
misstep with regard to the small number of restricted allotments hardly
means the Oklahoma courts made the far more extraordinary mistake of
failing to notice that the Five Tribes’ reservations—encompassing 19 mil-
lion acres—continued to exist.
 Cite as: 591 U. S. ____ (2020) 29

ROBERTS, C. J., dissenting

lacked “good faith.” Ibid. In the Court’s telling, the federal
government acquiesced in this extraordinary alleged power
grab, abdicating its responsibilities over the purported res-
ervation. And, all the while, the state and federal courts
turned a blind eye.
But we normally presume that government officials exer-
cise their duties in accordance with the law. Certainly the
presumption may be strained from time to time in this area,
but not so much as to justify the Court’s speculations, which
posit that government officials at every level either con-
spired to violate the law or uniformly misunderstood the
fundamental structure of their society and government.
Whatever the imperfections of our forebears, neither option
seems tenable. And it is downright inconceivable that this
could occur without prompting objections—from anyone, in-
cluding from the Five Tribes themselves. Indians fre-
quently asserted their rights during this period. The cases
above, for example, involve criminal appeals brought by In-
dians, and Indians raised numerous objections to land graft
in the former Territory. See Brief for Historians et al. as
Amici Curiae 28–31. Yet, according to the extensive record
compiled over several years for this case and a similar case,
Sharp v. Murphy, post, p. ___ (per curiam), Indians and
their counsel did not raise a single objection to state prose-
cutions on the theory that the lands at issue were still a
reservation. It stretches the imagination to suggest they
just missed it.
C
Finally, consider “the subsequent treatment of the area
in question and the pattern of settlement there.” Yankton
Sioux Tribe, 522 U. S., at 344. This evidence includes the
“subsequent understanding of the status of the reservation
by members and nonmembers as well as the United States
and the [relevant] State,” and the “subsequent demographic
history” of the area. Parker, 577 U. S., at ___, ___ (slip op.,
30 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

at 6, 10); see Solem, 465 U. S., at 471. Each of the indicia
from our precedents—subsequent treatment by Congress,
the State’s unquestioned exercise of jurisdiction, and demo-
graphic evidence—confirms that the Creek reservation did
not survive statehood.
First, “Congress’ own treatment of the affected areas”
strongly supports disestablishment. Id., at 471. After
statehood, Congress enacted several statutes progressively
eliminating restrictions on the alienation and taxation of
Creek allotments, and Congress subjected even restricted
lands to state jurisdiction. Since Congress had already de-
stroyed nearly all tribal authority, these statutes rendered
Creek parcels little different from other plots of land in the
State. See Act of May 27, 1908, 35 Stat. 312; Act of June
14, 1918, 40 Stat. 606; Act of Apr. 10, 1926, 44 Stat. 239.
This is not a scenario where Congress merely opened land
for “purchase . . . by non-Indians” while allowing the Tribe
to “continue to exercise governmental functions over [the]
land,” ante, at 11, and n. 3; rather, Congress eliminated
both restrictions on the lands here and the Creek Nation’s
authority over them. Such developments would be surpris-
ing if Congress intended for all of the former Indian Terri-
tory to be reservation land insulated from state jurisdiction
in significant ways. The simpler and more likely explana-
tion is that they reflect Congress’s understanding through
the years that “all Indian reservations as such have ceased
to exist” in Oklahoma, S. Rep. No. 1232, 74th Cong., 1st
Sess., 6 (1935), and that “Indian reservations [in the Indian
Territory] were destroyed” when “Oklahoma entered the
union,” S. Rep. No. 101–216, p. 47 (1989).
That understanding is now woven throughout the U. S.
Code, which applies numerous statutes to the land here by
extending them to the “former reservation[s]” “in Okla-
homa”—underscoring that no reservation exists today. 25
U. S. C. §2719(a)(2)(A)(i) (emphasis added) (Indian Gaming
 Cite as: 591 U. S. ____ (2020) 31

ROBERTS, C. J., dissenting

Regulatory Act); see Brief for United States as Amicus Cu-
riae 23; 23 U. S. C. §202(b)(1)(B)(v) (road grants; “former
Indian reservations in the State of Oklahoma”); 25 U. S. C.
§1452(d) (Indian Financing Act; “former Indian reserva-
tions in Oklahoma”); §2020(d) (education grants; “former
Indian reservations in Oklahoma”); §3103(12) (National In-
dian Forest Resources Management Act; “former Indian
reservations in Oklahoma”); 29 U. S. C. §741(d) (American
Indian Vocational Rehabilitation Services Act; “former In-
dian reservations in Oklahoma”); 33 U. S. C. §1377(c)(3)(B)
(waste treatment grants; “former Indian reservations in
Oklahoma”); 42 U. S. C. §5318(n)(2) (urban development
grants; “former Indian reservations in Oklahoma”).7
Second, consider the State’s “exercis[e] [of] unquestioned
jurisdiction over the disputed area since the passage of ” the
Enabling Act, which deserves “weight” as “an indication of
the intended purpose of the Act.” Rosebud Sioux Tribe, 430
U. S., at 599, n. 20, 604. As discussed above, for 113 years,
Oklahoma has asserted jurisdiction over the former Indian

——————
7 The Court suggests that these statutes only show that there are some

“former reservations” in Oklahoma, not that the Five Tribes’ former do-
mains are necessarily among them. Ante, at 27, n. 14. History says oth-
erwise. For example, the Five Tribes actively lobbied for inclusion of this
language in the Indian Gaming Regulatory Act. See Hearing on S. 902
et al. before the Senate Select Committee on Indian Affairs, 99th Cong.,
2d Sess., 299–300 (1986). They observed that the term “reservation,” as
originally defined, did not pertain to the “eastern Oklahoma tribes, in-
cluding the Five Civilized Tribes.” Ibid. (statement of Charles Blackwell,
representative of the Chickasaw Nation of Oklahoma). Accordingly, they
“recommend[ed] inclu[ding] . . . the wording ‘or in the case of Oklahoma
tribes, their former jurisdictional and/or reservation boundaries in Okla-
homa.’ ” Id., at 300 (emphasis added). The National Indian Gaming As-
sociation, which proposed the language on which the final act was ulti-
mately modeled, made the same point, observing that in Oklahoma
“reservation boundaries have been extinguished for most purposes” so
the statute should refer to “former reservation[s] in Oklahoma.” Id., at
312 (Memorandum from the National Indian Gaming Assn. to the Senate
Select Committee on Indian Affairs (June 17, 1986)).
32 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

Territory on the understanding that it is not a reservation,
without any objection by the Five Tribes until recently (or
by McGirt for the first 20 years after his convictions). See
Brief for Respondent 4, 40. The same goes for major cities
in Oklahoma. Tulsa, for example, has exercised jurisdiction
over both Indians and non-Indians for more than a century
on the understanding that it is not a reservation. See Brief
for City of Tulsa as Amicus Curiae 27–28.
All the while, the federal government has operated on the
same understanding. Brief for United States as Amicus Cu-
riae 24. No less than Felix Cohen, whose authoritative
treatise the Court repeatedly cites, agreed while serving as
Acting Solicitor of the Interior in 1941 that “all offenses by
or against Indians” in the former Indian Territory “are sub-
ject to State laws.” App. to Supp. Reply Brief for Petitioner
in Carpenter v. Murphy, O. T. 2018, No. 17–1107, p. 1a
(Memorandum for Commissioner of Indian Affairs (July 11,
1941)). In the view of the Department of the Interior, such
state jurisdiction was appropriate because the reservations
in the Territory “lost their character as Indian country” by
the time Oklahoma became a State. App. to Brief for
United States as Amicus Curiae 4a (Letter from O. Chap-
man, Assistant Secretary of the Interior, to the Attorney
General (Aug. 17, 1942)); see also supra, at 28, n. 6.
Indeed, far from disputing Oklahoma’s jurisdiction, the
Five Tribes themselves have repeatedly and emphatically
agreed that no reservation exists. After statehood, tribal
leaders and members frequently informed Congress that
“there are no reservations in Oklahoma.” App. to Brief for
Respondent 19a (Testimony of Hon. Bill Anoatubby, Gover-
nor, Chickasaw Nation, Hearings before the Subcommittee
on Indian, Insular and Alaska Native Affairs of the House
Committee on Natural Resources (Feb. 24, 2016)).8 They
——————
8 See App. to Brief for Respondent 18a–19a (excerpting various state-

ments before Congress, including: “[w]e are not a reservation tribe”
 Cite as: 591 U. S. ____ (2020) 33

ROBERTS, C. J., dissenting

took the same position before federal courts. Before this
litigation started, the Creek Nation represented to the
Tenth Circuit that there is only “ ‘checkerboard’ Indian
country within its former reservation boundaries.” Reply
Brief in No. 09–5123, p. 5 (emphasis added). And the Na-
tion never once contended in this Court that a sprawling
reservation still existed in the more than a century that
preceded the present disputes.
Like the Creek, this Court has repeatedly described the
area in question as the “former” lands of the Creek Nation.
See Grayson v. Harris, 267 U. S. 352, 353 (1925) (lands “ly-
ing within the former Creek Nation”); Woodward, 238 U. S.,
at 285 (lands “formerly part of the domain of the Creek Na-
tion”); Washington v. Miller, 235 U. S. 422, 423 (1914)
(lands “within what until recently was the Creek Nation”).
Yet today the Court concludes that the lands have been a
Creek reservation all along—contrary to the position
shared for the past century by this Court, the United
States, Oklahoma, and the Creek Nation itself.
Under our precedent, Oklahoma’s unquestioned, century-
long exercise of jurisdiction supports the conclusion that no
reservation persisted past statehood. See Yankton Sioux
Tribe, 522 U. S., at 357; Hagen, 510 U. S., at 421; Rosebud
Sioux Tribe, 430 U. S., at 604–605. “Since state jurisdiction
over the area within a reservation’s boundaries is quite lim-
ited, the fact that neither Congress nor the Department of
Indian Affairs has sought to exercise its authority over this
area, or to challenge the State’s exercise of authority is a

——————
(Principal Cherokee Chief, 1982), “Oklahoma, . . . of course, is not a res-
ervation State” (Chickasaw Governor, 1988), “Oklahoma is not [a reser-
vation State]” and “[w]e have no surface reservations in Oklahoma”
(Chickasaw advisor, 2011), as well as references to the boundaries and
lands of “former reservation[s]” (Chickasaw nominee for Assistant Sec-
retary of Indian Affairs, 2012; Inter-Tribal Council of the Five Civilized
Tribes, 2016)).
34 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

factor entitled to weight as part of the ‘jurisdictional his-
tory.’ ” Id., at 603–604 (citations omitted).
Third, consider the “subsequent demographic history” of
the lands at issue, which provides an “ ‘additional clue’ ” as
to the meaning of Congress’s actions. Parker, 577 U. S., at
___ (slip op., at 10) (quoting Solem, 465 U. S., at 472). Con-
tinuing from statehood to the present, the population of the
lands has remained approximately 85%–90% non-Indian.
See Brief for Respondent 43; Murphy v. Royal, 875 F. 3d
896, 965 (CA10 2017). “[T]hose demographics signify a di-
minished reservation.” Yankton Sioux Tribe, 522 U. S., at
357. The Court questions whether the consideration of de-
mographic history is appropriate, ante, at 18–19, 27, but we
have determined that it is a “necessary expedient.” Solem,
465 U. S., at 472, and n. 13 (emphasis added); see Parker,
577 U. S., at ___ (slip op., at 10). And for good reason. Our
precedents recognize that disestablishment cases call for a
wider variety of tools than more workaday questions of stat-
utory interpretation. Supra, at 12. In addition, the use of
demographic data addresses the practical concern that
“[w]hen an area is predominately populated by non-Indians
with only a few surviving pockets of Indian allotments, find-
ing that the land remains Indian country seriously burdens
the administration of state and local governments.” Solem,
465 U. S., at 471–472, n. 12.
Here those burdens—the product of a century of settled
understanding—are extraordinary. Most immediately, the
Court’s decision draws into question thousands of convic-
tions obtained by the State for crimes involving Indian de-
fendants or Indian victims across several decades. This in-
cludes convictions for serious crimes such as murder, rape,
kidnapping, and maiming. Such convictions are now sub-
ject to jurisdictional challenges, leading to the potential re-
lease of numerous individuals found guilty under state law
 Cite as: 591 U. S. ____ (2020) 35

ROBERTS, C. J., dissenting

of the most grievous offenses.9 Although the federal gov-
ernment may be able to reprosecute some of these crimes,
it may lack the resources to reprosecute all of them, and the
odds of convicting again are hampered by the passage of
time, stale evidence, fading memories, and dead witnesses.
See Brief for United States as Amicus Curiae 37–39. No
matter, the court says, these concerns are speculative be-
cause “many defendants may choose to finish their state
sentences rather than risk reprosecution in federal court.”
Ante, at 38. Certainly defendants like McGirt—convicted of
serious crimes and sentenced to 1,000 years plus life in
prison—will not adopt a strategy of running out the clock
on their state sentences. At the end of the day, there is no
escaping that today’s decision will undermine numerous
convictions obtained by the State, as well as the State’s abil-
ity to prosecute serious crimes committed in the future.
Not to worry, the Court says, only about 10%–15% of Ok-
lahoma citizens are Indian, so the “majority” of prosecu-
tions will be unaffected. Ibid. But the share of serious
crimes committed by 10%–15% of the 1.8 million people in
eastern Oklahoma, or of the 400,000 people in Tulsa, is no
small number.
Beyond the criminal law, the decision may destabilize the
governance of vast swathes of Oklahoma. The Court, de-
spite briefly suggesting that its decision concerns only a
narrow question of criminal law, ultimately acknowledges
that “many” federal laws, triggering a variety of rules,
spring into effect when land is declared a reservation. Ante,
at 39–40.
——————
9 The Court suggests that “well-known” “procedural obstacles” could

prevent challenges to state convictions. Ante, at 38. But, under Okla-
homa law, it appears that there may be little bar to state habeas relief
because “issues of subject matter jurisdiction are never waived and can
therefore be raised on a collateral appeal.” Murphy v. Royal, 875 F. 3d
896, 907, n. 5 (CA10 2017) (quoting Wallace v. State, 935 P. 2d 366, 372
(Okla. Crim. App. 1997)).
36 MCGIRT v. OKLAHOMA

ROBERTS, C. J., dissenting

State and tribal authority are also transformed. As to the
State, its authority is clouded in significant respects when
land is designated a reservation. Under our precedents, for
example, state regulation of even non-Indians is preempted
if it runs afoul of federal Indian policy and tribal sover-
eignty based on a nebulous balancing test. This test lacks
any “rigid rule”; it instead calls for a “particularized inquiry
into the nature of the state, federal, and tribal interests at
stake,” contemplated in light of the “broad policies that un-
derlie” relevant treaties and statutes and “notions of sover-
eignty that have developed from historical traditions of
tribal independence.” White Mountain Apache Tribe v.
Bracker, 448 U. S. 136, 142, 144–145 (1980). This test
mires state efforts to regulate on reservation lands in sig-
nificant uncertainty, guaranteeing that many efforts will be
deemed permissible only after extensive litigation, if at
all.10
In addition to undermining state authority, reservation
status adds an additional, complicated layer of governance
over the massive territory here, conferring on tribal govern-
ment power over numerous areas of life—including powers
over non-Indian citizens and businesses. Under our prece-
dents, tribes may regulate non-Indian conduct on reserva-
tion land, so long as the conduct stems from a “consensual

——————
10 See, e.g., White Mountain Apache Tribe, 448 U. S., at 148–151 (bar-

ring State from imposing motor carrier license tax and fuel use taxes on
non-Indian logging companies that harvested timber on a reservation);
Warren Trading Post Co. v. Arizona Tax Comm’n, 380 U. S. 685, 690–
692 (1965) (barring State from taxing income earned by a non-Indian
who operated a trading post on a reservation); New Mexico v. Mescalero
Apache Tribe, 462 U. S. 324, 325 (1983) (barring State from regulating
hunting and fishing by non-Indians on a reservation); see also Brendale
v. Confederated Tribes and Bands of Yakima Nation, 492 U. S. 408, 448
(1989) (opinion of Stevens, J.) (arguing that it is “impossible to articulate
precise rules that will govern whenever a tribe asserts that a land use
approved by a county board is pre-empted by federal law”).
 Cite as: 591 U. S. ____ (2020) 37

ROBERTS, C. J., dissenting

relationship[ ] with the tribe or its members” or directly af-
fects “the political integrity, the economic security, or the
health or welfare of the tribe.” Montana v. United States,
450 U. S. 544, 565–566 (1981); see Cohen §6.02(2)(a), at
506–507. Tribes may also impose certain taxes on non-In-
dians on reservation land, see Kerr-McGee Corp. v. Navajo
Tribe, 471 U. S. 195, 198 (1985), and in this litigation, the
Creek Nation contends that it retains the power to tax non-
members doing business within its borders. Brief for Mus-
cogee (Creek) Nation as Amicus Curiae 18, n. 6. No small
power, given that those borders now embrace three million
acres, the city of Tulsa, and hundreds of thousands of Ok-
lahoma citizens. Recognizing the significant “potential for
cost and conflict” caused by its decision, the Court insists
any problems can be ameliorated if the citizens of Okla-
homa just keep up the “spirit” of cooperation behind exist-
ing intergovernmental agreements between Oklahoma and
the Five Tribes. Ante, at 41. But those agreements are
small potatoes compared to what will be necessary to ad-
dress the disruption inflicted by today’s decision.
The Court responds to these and other concerns with the
truism that significant consequences are no “license for us
to disregard the law.” Ibid. Of course not. But when those
consequences are drastic precisely because they depart
from how the law has been applied for more than a cen-
tury—a settled understanding that our precedents demand
we consider—they are reason to think the Court may have
taken a wrong turn in its analysis.
* * *
As the Creek, the State of Oklahoma, the United States,
and our judicial predecessors have long agreed, Congress
disestablished any Creek reservation more than 100 years
ago. Oklahoma therefore had jurisdiction to prosecute
McGirt. I respectfully dissent.
 Cite as: 591 U. S. ____ (2020) 1

THOMAS, J., dissenting

SUPREME COURT OF THE UNITED STATES
_________________

No. 18–9526
_________________

JIMCY MCGIRT, PETITIONER v. OKLAHOMA
ON WRIT OF CERTIORARI TO THE COURT OF CRIMINAL
APPEALS OF OKLAHOMA
[July 9, 2020]

JUSTICE THOMAS, dissenting.
I agree with THE CHIEF JUSTICE that the former Creek
Nation Reservation was disestablished at statehood and
Oklahoma therefore has jurisdiction to prosecute petitioner
for sexually assaulting his wife’s granddaughter. Ante, at
1–2 (dissenting opinion). I write separately to note an ad-
ditional defect in the Court’s decision: It reverses a state-
court judgment that it has no jurisdiction to review. “[W]e
have long recognized that ‘where the judgment of a state
court rests upon two grounds, one of which is federal and
the other non-federal in character, our jurisdiction fails if
the non-federal ground is independent of the federal ground
and adequate to support the judgment.’ ” Michigan v. Long,
463 U. S. 1032, 1038, n. 4 (1983) (quoting Fox Film Corp. v.
Muller, 296 U. S. 207, 210 (1935)). Under this well-settled
rule, we lack jurisdiction to review the Oklahoma Court of
Criminal Appeals’ decision, because it rests on an adequate
and independent state ground.
In his application for state postconviction relief, peti-
tioner claimed that Oklahoma lacked jurisdiction to prose-
cute him because his crime was committed on Creek Nation
land and thus was subject to the exclusive jurisdiction of
the Federal Government under the Major Crimes Act, 18
U. S. C. §1153. In support of his argument, petitioner cited
the Tenth’s Circuit’s decision in Murphy v. Royal, 875 F. 3d
896 (2017).
2 MCGIRT v. OKLAHOMA

THOMAS, J., dissenting

The Oklahoma Court of Criminal Appeals concluded that
petitioner’s claim was procedurally barred under state law
because it was “not raised previously on direct appeal” and
thus was “waived for further review.” 2018 OK CR 1057 ¶2,
___ P. 3d ___, ___ (citing Okla. Stat., Tit. 22, §1086 (2011)).
The court found no grounds for excusing this default, ex-
plaining that “[p]etitioner [had] not established any suffi-
cient reason why his current grounds for relief were not pre-
viously raised.” ___ P. 3d, at ___. This state procedural bar
was applied independent of any federal law, and it is ade-
quate to support the decision below. We therefore lack ju-
risdiction to disturb the state court’s judgment.
There are two possible arguments in favor of jurisdiction,
neither of which hold water. First, one might claim that the
state procedural bar is not an “adequate” ground for deci-
sion in this case. In Murphy, the Tenth Circuit suggested
that Oklahoma law permits jurisdictional challenges to be
raised for the first time on collateral review. 875 F. 3d, at
907, n. 5 (citing Wallace v. State, 1997 OK CR 18, 935 P. 2d
366). But the Oklahoma Court of Criminal Appeals did not
even hint at such grounds for excusing petitioner’s default
here. More importantly, however, we may not go beyond
“the four corners of the opinion” and delve into background
principles of Oklahoma law to determine the adequacy of
the independent state ground. Long, 463 U. S., at 1040.
This Court put an end to that approach in Long, noting that
“[t]he process of examining state law is unsatisfactory be-
cause it requires us to interpret state laws with which we
are generally unfamiliar, and which often, as in this case,
have not been discussed at length by the parties.” Id., at
1039. Moreover, such second-guessing disrespects “the in-
dependence of state courts,” id., at 1040, and the State it-
self, Coleman v. Thompson, 501 U. S. 722, 738–739 (1991).
Second, one might argue, as the Court does, that we have
jurisdiction because the decision below rests on federal, not
state, grounds. See ante, at 38, n. 15. It is true that the
 Cite as: 591 U. S. ____ (2020) 3

THOMAS, J., dissenting

Oklahoma Court of Criminal Appeals briefly recited the
procedural history of Murphy and recognized that the
Tenth Circuit’s decision—which we granted certiorari to re-
view—is not yet final. But contrary to the Court’s assertion
that brief discussion of federal case law did not come close
to “address[ing] the merits of [petitioner’s] federal [Major
Crimes Act] claim.” Ante, at 38, n. 15. The state court did
not analyze the relevant statutory text or this Court’s deci-
sions in Solem v. Bartlett, 465 U. S. 463 (1984), and Ne-
braska v. Parker, 577 U. S. 481 (2016). It reads far too
much into the opinion to claim that the court’s brief refer-
ence to the Tenth Circuit’s decision in Murphy transformed
the state court’s decision into one that “fairly appear[s] to
rest primarily on federal law or to be interwoven with fed-
eral law,” Long, supra, at 1040–1041; see also ante, at 38,
n. 15. Nothing in the court’s opinion suggests that its judg-
ment was at all based on federal law. Thus, even if we were
to set aside the fact that the state court “clearly and ex-
pressly state[d] that [its decision] was based on state proce-
dural grounds,” we could not presume jurisdiction here.
Coleman, supra, at 735–736 (internal quotation marks
omitted).
The Court might think that, in the grand scheme of
things, this jurisdictional defect is fairly insignificant. Af-
ter all, we were bound to resolve this federal question
sooner or later. See Royal v. Murphy, 584 U. S. ___ (2018).
But our desire to decisively “settle [important disputes] for
the sake of convenience and efficiency” must yield to the
“overriding and time-honored concern about keeping the
Judiciary’s power within its proper constitutional sphere.”
Hollingsworth v. Perry, 570 U. S. 693, 704–705 (2013) (in-
ternal quotation marks omitted). Because the Oklahoma
court’s “judgment does not depend upon the decision of any
federal question[,] we have no power to disturb it.” Enter-
prise Irrigation Dist. v. Farmers Mut. Canal Co., 243 U. S.
157, 164 (1917).
4 MCGIRT v. OKLAHOMA

THOMAS, J., dissenting

I agree with THE CHIEF JUSTICE that the Court misap-
plies our precedents in granting petitioner relief. Ante, at
6–38 (dissenting opinion). But in doing so, the Court also
overrides Oklahoma’s statutory procedural bar, upsetting a
violent sex offender’s conviction without the power to do so.
The State of Oklahoma deserves more respect under our
Constitution’s federal system. Therefore, I respectfully dis-
sent.