Petition for Writ of Certiorari — Suncor Energy (U.S.A.) Inc., et al., Petitioners v. Board of County Commissioners of Boulder County, et al.

APPENDIX

TABLE OF CONTENTS

Appendix A:

Appendix B:

Appendix C:

Appendix D:

Court of appeals opinion,

July 7, 2020 ..................................................... 1a

District court opinion,

September 5, 2019 ....................................... 59a

District court order denying motion

to stay the remand order,

October 7, 2019 .......................................... 114a

Court of appeals order denying motion

to stay the remand order,

October 17, 2019 ........................................ 131a

1a

APPENDIX A

UNITED STATES COURT OF APPEALS

FOR THE TENTH CIRCUIT

No. 19-1330

BOARD OF COUNTY COMMISSIONERS OF BOULDER

COUNTY; BOARD OF COUNTY COMMISSIONERS OF SAN

MIGUEL COUNTY; CITY OF BOULDER,

PLAINTIFFS-APPELLEES,

v.

SUNCOR ENERGY (U.S.A.) INC.; SUNCOR ENERGY SALES

INC.; SUNCOR ENERGY INC.; EXXON MOBIL

CORPORATION, DEFENDANTS-APPELLANTS.

Filed: July 7, 2020

Before: LUCERO, HOLMES, and McHUGH, Circuit

Judges.

McHUGH, Circuit Judge.

This appeal concerns whether federal court is the

proper forum for a suit filed in Colorado state court by

local governmental entities for the global warming-related damage allegedly caused by oil and gas companies

in Colorado. Suncor Energy and ExxonMobil advanced

seven bases for federal subject matter jurisdiction in re-

2a

moving the action to federal court, each of which the district court rejected in its remand order. Suncor Energy

and ExxonMobil now appeal, relying on six of those bases

for federal jurisdiction. We hold, however, that 28 U.S.C.

§ 1447(d) limits our appellate jurisdiction to just one of

them—federal officer removal under 28 U.S.C.

§ 1442(a)(1). And because we conclude ExxonMobil failed

to establish grounds for federal officer removal, we affirm

the district court’s order on that basis and dismiss the remainder of this appeal.

I. BACKGROUND

Three local Colorado government entities—the

County Commissioners of Boulder and San Miguel Counties and the City of Boulder (Plaintiffs-Appellees; collectively, the “Counties”)—filed suit in Colorado state court

on June 11, 2018, against Suncor Energy1 and ExxonMobil Corporation (Defendants-Appellants, collectively,

“Defendants”). The complaint asserts that the Counties

face substantial and rising costs to protect people and

property within their jurisdictions from the threat of

global warming, including from increasing and intensified

heat waves, wildfires, droughts, and floods across Colorado. The Counties allege that Defendants have substantially contributed to this local environmental harm by engaging in unchecked fossil fuel activity—producing, promoting, refining, marketing, and selling—which has resulted in excess greenhouse gas emissions. For decades

after becoming aware of the dangers of global warming,

the Counties further allege, Defendants continued to produce, promote, refine, market, and sell fossil fuels at levels

that caused and contributed to negative climate alteration

“Suncor Energy” includes Suncor Energy (U.S.A.) Inc.; Suncor

Energy Sales Inc.; and Suncor Energy Inc.

1

3a

without disclosing the harms posed by continued fossil

fuel overuse. According to the complaint, Defendants misrepresented the dangers of unchecked fossil fuel use and

acted to prevent and forestall changes in energy use that

they knew were needed to limit the impact of global warming, thereby exacerbating the climate-related harm suffered by the Counties and their residents.

The complaint asserts state law claims for public and

private nuisance, trespass, unjust enrichment, civil conspiracy, and violation of the Colorado Consumer Protection Act. Among other forms of relief, the Counties seek

past and future compensatory damages to mitigate the

impact of global warming in their respective jurisdictions,

along with remediation and/or abatement of the attendant

global warming-related environmental hazards they now

face. The Counties do not seek “to enjoin any oil and gas

operations or sales in the State of Colorado, or elsewhere,

or to enforce emissions controls of any kind.” App. 195.

They ask the state court not “to stop or regulate” fossil

fuel production or emissions, but instead to ensure Defendants pay a pro rata share of the costs the Counties

have incurred and will incur based on Defendants’

averred contribution to climate alteration, and to help remediate the harm the Counties claim has been and will be

caused by Defendants’ allegedly tortious and illegal conduct. App. 74.

On June 29, 2018, Defendants filed a notice of removal

in federal district court for the District of Colorado, asserting seven grounds for federal jurisdiction. Five of

these grounds relied upon the general removal statute, 28

U.S.C. § 1441(a), which allows for removal of “any civil action brought in a State court of which the district courts

of the United States have original jurisdiction.” Of these

five grounds, four were based on general federal question

4a

jurisdiction2—that the Counties’ claims (1) arose under

federal common law; (2) were completely preempted by

federal law; (3) implicated disputed and substantial federal issues under Grable & Sons Metal Products, Inc. v.

Darue Engineering & Manufacturing, 545 U.S. 308

(2005); and (4) arose in part from incidents that occurred

on federal enclaves. The fifth claim of original federal jurisdiction was based on the Outer Continental Shelf

Lands Act, 43 U.S.C. § 1349(b). Additionally, Defendants

relied on two other removal provisions: the bankruptcy

removal statute, 28 U.S.C. § 1452, and the federal officer

removal statute, 28 U.S.C. § 1442(a)(1).

The Counties filed a motion to remand pursuant to 28

U.S.C. § 1447(c) based on lack of federal subject matter

jurisdiction. The district court granted this motion on

September 5, 2019, rejecting all seven grounds for removal and remanding to the Colorado state court. Bd. of

Cty. Comm’rs. of Boulder County v. Suncor Energy

(U.S.A.) Inc. (Boulder County I), 405 F. Supp. 3d 947,

954–55 (D. Colo. 2019).

Defendants appealed the district court’s remand order

with respect to six of their seven asserted bases for removal (omitting a challenge to bankruptcy removal). They

also moved in the district court for a stay of the remand

order pending appeal. Notwithstanding the general bar to

remand order appealability imposed by 28 U.S.C.

§ 1447(d), Defendants argued before the district court

that the exception in § 1447(d) permitting review of federal officer removal under 28 U.S.C. § 1442 creates appellate jurisdiction to consider all of their asserted removal

2

See 28 U.S.C. § 1331, which confers original jurisdiction on the

federal district courts “of all civil actions arising under the Constitution, laws, or treaties of the United States.”

5a

bases. While acknowledging that this court has yet to determine the scope of appellate review of remand orders

premised on the § 1447(d) exceptions, as well as circuit

disagreement on that issue, Defendants asserted that plenary review was compelled by a Seventh Circuit decision

interpreting the Supreme Court’s holding in Yamaha Motor Corp., U.S.A. v. Calhoun, 516 U.S. 199 (1996). Defendants further contended that this court’s interpretation of

the Class Action Fairness Act’s removal provision in

Coffey v. Freeport McMoran Copper & Gold, 581 F.3d

1240 (10th Cir. 2009), “strongly suggests that it would review the district court’s entire order, not simply the

ground that permitted appeal.” Defendants’ Mot. for Stay

of Remand Order, Dist. Ct. ECF No. 75 at 6.

The district court denied this motion to stay its remand order on October 7, 2019. Bd. of Cty. Comm’rs of

Boulder County v. Suncor Energy (U.S.A.) Inc. (Boulder

County II), 423 F. Supp. 3d 1066 (D. Colo. 2019). Noting

the split of authority on the scope of appellate review of

remand orders, as well as the lack of a controlling Tenth

Circuit opinion, the district court reasoned that this court

would likely “follow the weight of authority and find that

the only ground subject to appeal is federal officer jurisdiction under § 1442.” Id. at 1070. It disagreed with Defendants’ reading of Yamaha and Coffey, finding instead

that “Coffey suggests the Tenth Circuit would be unlikely

to review aspects of a remand order that would otherwise

be unreviewable”—here, all bases for federal question jurisdiction other than § 1442. Id. at 1071.

Defendants then filed motions in this court and the Supreme Court for a temporary stay of the remand order

pending appeal, which both courts denied. The Counties

filed a motion for partial dismissal based on the reviewa-

6a

bility bar in § 1447(d), seeking to narrow the issues on appeal to only the propriety of federal officer removal.3 It is

to this issue of the scope of our appellate jurisdiction that

we first turn.

II. SCOPE OF APPELLATE JURISDICTION

“‘The authority of appellate courts to review districtcourt orders remanding removed cases to state court is

substantially limited by statute,’ namely, 28 U.S.C.

§ 1447(d).” Mayor & City Council of Baltimore v. BP

P.L.C., 952 F.3d 452, 459 (4th Cir. 2020) (quoting Powerex

Corp. v. Reliant Energy Servs., 551 U.S. 224, 229 (2007)).

Consequently, “the threshold question in an appeal of a

remand order is whether the district court’s decision is reviewable notwithstanding the proscription set forth in 28

U.S.C. § 1447(d).”4 Am. Soda, LLP v. U.S. Filter

Wastewater Grp., 428 F.3d 921, 924 (10th Cir. 2005). Section 1447(d) of the Judicial Code, Title 28 U.S.C., provides:

An order remanding a case to the State court from

which it was removed is not reviewable on appeal or

otherwise, except that an order remanding a case to

the State court from which it was removed pursuant to

The Counties also moved for summary affirmance based on issue

preclusion, arguing the Fourth Circuit’s ruling in Mayor & City

Council of Baltimore v. B.P. PLC, 952 F.3d 452 (4th Cir. 2020)—

which rejected the same federal officer removal argument brought

here, in a case featuring ExxonMobil as a defendant—is a supervening change of law under 10th Cir. R. 27.3(A)(1)(b). See also County of

San Mateo v. Chevron Corp., 960 F.3d 586 (9th Cir. 2020) (rejecting

the same federal officer removal argument in a case also featuring

ExxonMobil as a defendant).

3

4

We can thoroughly explore this question because “federal courts

always have jurisdiction to consider their own jurisdiction.” Pritchett

v. Office Depot, Inc., 420 F.3d 1090, 1093 (10th Cir. 2005).

7a

section 1442 or 1443 of this title shall be reviewable by

appeal or otherwise.

The primary clause of this statute is construed together with 28 U.S.C. § 1447(c), which describes two

grounds for remand—lack of federal subject matter jurisdiction and a defect in removal procedure. See Things Remembered, Inc. v. Petrarca, 516 U.S. 124, 127 (1995); City

of Albuquerque v. Soto Enters., 864 F.3d 1089, 1092–95

(10th Cir. 2017). “If a district court orders remand on either of these grounds, § 1447(d) absolutely prohibits appellate review of the order, and we adhere firmly to this

prohibition even where we believe that the district court

was plainly incorrect.” Kennedy v. Lubar, 273 F.3d 1293,

1297 (10th Cir. 2001); see also Powerex Corp., 551 U.S. at

238–39 (“Appellate courts must take th[e] jurisdictional

prescription [of § 1447(d)] seriously, however pressing the

merits of the appeal might seem.”). “Thus, we have jurisdiction to review a remand order only if (1) the remand

was for a reason other than lack of subject matter jurisdiction or a defect in the removal procedure or (2) the ‘except’ clause of § 1447(d) gives us jurisdiction.” Miller v.

Lambeth, 443 F.3d 757, 759 (10th Cir. 2006).

The roots of § 1447(d)’s primary clause stretch back to

1887. Thermtron Prods. v. Hermansdorfer, 423 U.S. 336,

343 (1976); see Osborn v. Haley, 549 U.S. 225, 262 (2007)

(Scalia, J., dissenting) (stating that § 1447(d)’s “bar to appellate review is a venerable one”). The “except” clause

was added via the 1964 Civil Rights Act, and allowed for

appellate review only of remands of civil rights cases removed pursuant to 28 U.S.C. § 1443. See Thermtron, 423

U.S. at 342 n.7. Congress expanded this clause to provide

for review of remands of cases removed pursuant to the

federal officer removal statute, 28 U.S.C. § 1442, through

the Removal Clarification Act of 2011, Pub. L. No. 112-51,

8a

125 Stat. 545. This act amended § 1447(d) “by inserting

‘1442 or’ before ‘1443.’” 125 Stat. at 546.

Here, the district court’s remand order was premised

on lack of subject matter jurisdiction, a § 1447(c) ground

barred from review by § 1447(d). Boulder County I, 405

F. Supp. 3d at 955–56. This characterization was indisputably colorable. See Powerex Corp., 551 U.S. at 234

(“[R]eview of the District Court’s characterization of its

remand as resting upon lack of subject-matter jurisdiction, to the extent it is permissible at all, should be limited

to confirming that that characterization was colorable.”).

It was also indisputably in good faith. See Archuleta v.

Lacuesta, 131 F.3d 1359, 1363 (10th Cir. 1997) (“[W]here

a district court in good faith remands a case for lack of

jurisdiction under § 1447(c), we do not have the power to

review the remand.”). The jurisdictional dispute thus concerns only the effect of § 1447(d)’s “except” clause on the

scope of our appellate review of the district court’s order.5

Appellate jurisdiction is also constrained by 28 U.S.C. § 1291,

which empowers federal circuit courts to review only “final decisions

of the district courts.” In Thermtron Products, Inc. v. Hermansdorfer, 423 U.S. 336, 352–53 (1976), the Supreme Court stated that “an

order remanding a removed action does not represent a final judgment reviewable by appeal.” But the Court disavowed this assertion

in Quackenbush v. Allstate Ins. Co., 517 U.S. 706 (1996), reasoning

that while the abstention-based remand order at issue “d[id] not meet

the traditional definition of finality,” id. at 715, it was nonetheless appealable because it put the litigants “effectively out of [federal]

court,” id. at 714 (quoting Moses H. Cone Mem’l Hosp. v. Mercury

Constr. Corp., 460 U.S. 1, 10 n.11 (1983)). “We have acknowledged the

central point of Quackenbush, i.e., that a remand order may be reviewed under 28 U.S.C. § 1291 as a final order or as a collateral order

because [a] remand order puts the litigants effectively out of court.”

In re Stone Container Corp., 360 F.3d 1216, 1219 (10th Cir. 2004) (internal quotation marks omitted). Consequently, § 1291 does not present a jurisdictional hurdle here.

5

9a

Defendants assert that because their removal was

premised partly on federal officer removal under § 1442,

we have appellate jurisdiction to review the district

court’s entire remand order, not just the portion dispensing with the federal officer removal argument. The Counties disagree, asserting that the scope of our review must

be confined to the district court’s disposition of the § 1442

argument. We have yet to issue a precedential opinion deciding this question of appellate jurisdiction, which turns

on statutory construction.6 In doing so now, we adopt the

narrower interpretation of the scope of § 1447(d) review

advanced by the Counties.

A. The § 1447(d) Circuit Split

Before proceeding to the substantive statutory analysis, we pause to note disagreement among the courts of

appeals over whether invoking a § 1447(d) exception in a

petition for removal creates appellate jurisdiction over the

district court’s whole remand order, or only over that portion addressing the excepted basis. Six circuits—the Second, Third, Fourth, Eighth, Ninth, and Eleventh—hold

that a remand order premised on a § 1447(c) ground is reviewable only to the extent it addresses a § 1442 (federal

officer) or 1443 (civil rights) removal argument. See Jacks

v. Meridian Resource Co., 701 F.3d 1224, 1229 (8th Cir.

2012); Patel v. Del Taco, Inc., 446 F.3d 996, 998 (9th Cir.

2006); Alabama v. Conley, 245 F.3d 1292, 1293 n.1 (11th

6

In Sanchez v. Onuska, No. 93-2155, 1993 WL 307897, at *1 (10th

Cir. Aug. 13, 1993) (unpublished), we determined that § 1447(d) allowed for review of a remand order “[t]o the extent the removal is

based upon § 1443,” but that the remainder of the remand order was

“not reviewable and must be dismissed for lack of jurisdiction.” Unpublished decisions, of course, provide only persuasive authority. See

10th Cir. R. 32.1(A). After conducting our own analysis here, we adopt

a position consistent with Onuska.

10a

Cir. 2001); Davis v. Glanton, 107 F.3d 1044, 1047 (3d Cir.

1997); State Farm Mut. Auto Ins. Co. v. Baasch, 644 F.2d

94, 97 (2d Cir. 1981); Noel v. McCain, 538 F.2d 633, 635

(4th Cir. 1976); see also City of Baltimore, 952 F.3d at 459

(rejecting arguments to depart from circuit precedent on

the scope of § 1447(d) review via an appeal concerning

functionally identical global warming-related state law

claims); County of San Mateo v. Chevron Corp., 960 F.3d

586, 595–98 (9th Cir. 2020) (same).

In 2015, the Seventh Circuit fractured this unanimity

on the scope of appellate review created by § 1447(d),

holding that the invocation of a § 1447(d) exception allows

for plenary review of all other removal bases addressed in

a remand order. See Lu Junhong v. Boeing Co., 792 F.3d

805, 811 (7th Cir. 2015).7 Unlike the other courts to ad-

Two other circuits have since issued opinions following Lu Junhong on the scope of appellate review created by § 1447(d), but each

has conflicting precedent on the issue.

7

In Decatur Hospital Authority v. Aetna Health, Inc., 854 F.3d 292,

296 (5th Cir. 2017), the Fifth Circuit relied on Lu Junhong’s reasoning to hold the entire district court’s remand order reviewable when

one of the asserted grounds for removal is § 1442. In a subsequent

opinion dismissing in part an appeal from a remand order, however,

the Fifth Circuit noted in passing that while the defendant “d[id] not

argue that the § 1447(d) exception for federal officer jurisdiction allows us to review the entire remand order,” “[t]his court has rejected

similar arguments in the past.” City of Walker v. Louisiana, 877 F.3d

563, 566 n.2 (5th Cir. 2017) (citing Robertson v. Ball, 534 F.2d 63, 65–

66 (5th Cir. 1976)); see also Gee v. Texas, 769 F. App’x 134, 134 & n.2

(5th Cir. 2019) (unpublished) (following City of Walker, while not citing Decatur Hospital, in holding that “[w]here a party has argued for

removal on multiple grounds, we only have jurisdiction to review a

district court’s remand decision for compliance with [§ 1442 or

1443]”).

11a

dress the issue—which employed mostly summary analysis in refusing to extend the review granted by the

§ 1447(d) exceptions to any otherwise nonreviewable removal bases contained in a remand order—the Seventh

Circuit engaged in a comprehensive discussion of statutory text and policy. As Defendants lean heavily on this

reasoning, we examine it in some depth.

The Seventh Circuit’s reasoning in Lu Junhong relied

primarily on the Supreme Court’s decision in Yamaha

Motor Corp., U.S.A. v. Calhoun, 516 U.S. 199 (1996).

Yamaha addressed the meaning of 28 U.S.C. § 1292(b),

which concerns a district court’s certification of controlling questions of law to the courts of appeals for discretionary review. The Yamaha Court held that upon accepting an interlocutory appeal under § 1292(b), a federal

court of appeals has jurisdiction over the whole “order,”

rather than being limited to review of the individual question (or questions) framed by the district court. 516 U.S.

at 205. Per Lu Junhong’s interpretation of Yamaha’s

holding, “[t]o say that a district court’s ‘order’ is reviewable is to allow appellate review of the whole order, not just

of particular issues or reasons.” 792 F.3d at 811.

In Mays v. City of Flint, 871 F.3d 437, 442 (6th Cir. 2017), the Sixth

Circuit cited Lu Junhong in holding that its jurisdiction to review an

order remanding a case that was removed pursuant to § 1442 “also

encompasses review of the district court’s decision on the alternative

ground for removal under 28 U.S.C. § 1441”—there, “substantial federal question” jurisdiction. However, Mays failed to distinguish two

Sixth Circuit decisions from the 1970’s—Detroit Police Lieutenants

& Sergeants Ass’n v. City of Detroit, 597 F.3d 566, 567–68 (6th Cir.

1979), and Appalachian Volunteers, Inc. v. Clark, 432 F.2d 530, 534

(6th Cir. 1970)—that held appellate jurisdiction lacking to review any

portion of a district court’s remand order other than its ruling on

§ 1443 (at that time the only statutory exception in § 1447(d)).

12a

In determining that § 1447(d) is best construed the

same way, Lu Junhong analogized to another statute creating an exception to the general lack of appellate jurisdiction over remand orders. The Class Action Fairness

Act (“CAFA”), Pub. L. No. 109-2, 119 Stat. 4–14, creates

federal subject matter jurisdiction over certain types of

class actions and allows for appellate review “of ‘an order

of a district court’ that has remanded after finding that

the Act does not permit removal.” 792 F.3d at 811 (quoting

28 U.S.C. § 1453(c)(1)). A prior Seventh Circuit decision,

Brill v. Countrywide Home Loans, Inc., 427 F.3d 446,

451–52 (7th Cir. 2005), applied Yamaha in interpreting

§ 1453(c)(1) to allow for plenary review of remand orders

addressing CAFA removal, even if such orders also address other bases for removal. Lu Junhong reasoned that

Brill stood for the proposition “that once an appeal of a

remand ‘order’ has been authorized by statute, the court

of appeals may consider all of the legal issues entailed in

the decision to remand.” 792 F.3d at 811.

The Lu Junhong court deemed its interpretation of

the word “order” in § 1447(d) to be “entirely textual”:

The Court remarked in Kircher [v. Putnam Funds

Trust, 547 U.S. 633, 641 n.8 (2006)], that Congress has

on occasion made the rule of § 1447(d) inapplicable to

particular “orders”—and for this the Court cited,

among other statutes, § 1447(d) itself. We take both

Congress and Kircher at their word in saying that, if

appellate review of an “order” has been authorized,

that means review of the “order.” Not particular reasons for an order, but the order itself.

Id. at 812.

And the Lu Junhong court further determined that

§ 1447(d)’s statutory purpose led to the same outcome:

13a

[Section] 1447(d) was enacted to prevent appellate delay in determining where litigation will occur. . . . But

once Congress has authorized appellate review of a remand order—as it has authorized review of suits removed on the authority of § 1442—a court of appeals

has been authorized to take the time necessary to determine the right forum. The marginal delay from

adding an extra issue to a case where the time for

briefing, argument, and decision has already been accepted is likely to be small.

Id. at 813 (citations omitted). Any concern that unscrupulous defendants will use the § 1447(d) exceptions as “a

hook to allow appeal of some different subject” did not

counsel a different result, because frivolous removals can

lead to sanctions, and frivolous appeals can be dealt with

summarily. Id.

B. Statutory Analysis

To decide the scope of our appellate review of the district court’s remand order—and determine whether to

follow Lu Junhong or the opposing weight of circuit authority on the issue—we must construe the meaning of

§ 1447(d)’s “except” clause de novo. See United States v.

Porter, 745 F.3d 1035, 1040 (10th Cir. 2014).

“The goal of statutory interpretation is to ascertain

the congressional intent and give effect to the legislative

will.” In re Taylor, 899 F.3d 1126, 1129 (10th Cir. 2018)

(internal quotation marks omitted). “In conducting this

analysis, we first turn to the statute’s plain language,” id,

as “[a] statute clear and unambiguous on its face must be

interpreted according to its plain meaning,” In re Geneva

Steel Co., 281 F.3d 1173, 1178 (10th Cir. 2002).

“A statute is ambiguous when it is capable of being understood by reasonably well-informed persons in two or

14a

more different senses.” United States v. Quarrell, 310

F.3d 664, 669 (10th Cir. 2002) (internal quotation marks

omitted). “The plainness or ambiguity of statutory language is determined by reference to the language itself,

the specific context in which that language is used, and the

broader context of the statute as a whole.” Ceco Concrete

Const., LLC v. Centennial State Carpenters Pension Tr.,

821 F.3d 1250, 1258 (10th Cir. 2016) (quoting Robinson v.

Shell Oil Co., 519 U.S. 337, 341 (1997)). If statutory meaning cannot be derived “merely by reference to the text, we

may also look to traditional canons of statutory construction to inform our interpretation,” Conrad v. Phone Directories Co., 585 F.3d 1376, 1381 (10th Cir. 2009), and

“may seek guidance from Congress’s intent, a task aided

by reviewing the legislative history,” In re Geneva Steel

Co., 281 F.3d at 1178. “Ambiguous text can also be decoded by knowing the purpose behind the statute.” Id.

Because text alone does not clarify the meaning of

§ 1447(d)’s “except” clause, we rely upon this full toolkit

of statutory construction. Cf. Watson v. Philip Morris

Cos., 551 U.S. 142, 147 (2007) (using the “text’s language,

context, history, and purposes” to guide interpretation of

the federal officer removal statute).

1. Text and Context

The “except” clause states “that an order remanding a

case . . . removed pursuant to section 1442 or 1443 . . . shall

be reviewable[.]” 28 U.S.C. § 1447(d) (emphasis added).

Defendants seize upon this reference to “order,” contending the “plain text of Section 1447(d) provides that, when

a case is removed under Section 1442, the remand ‘order’—not just the applicability of the federal-officer

ground for removal—is reviewable on appeal.” Appellant

Br. at 4; see Lu Junhong, 792 F.3d at 811 (“To say that a

district court’s ‘order’ is reviewable is to allow appellate

15a

review of the whole order, not just of particular issues or

reasons.”). We do not interpret the word “order” in isolation, however, for “[t]he meaning—or ambiguity—of certain words or phrases may only become evident when

placed in context.” FDA v. Brown & Williamson Tobacco

Corp., 529 U.S. 120, 132 (2000); see also Deal v. United

States, 508 U.S. 129, 132 (1993) (“[T]he meaning of a word

cannot be determined in isolation, but must be drawn

from the context in which it is used.”); United States v.

Villa, 589 F.3d 1334, 1343 (10th Cir. 2009) (“[T]he meaning of statutory language, plain or not, depends on context.” (quoting Bailey v. United States, 516 U.S. 137, 145

(1995)). Here, the specific context of the “except” clause

adds ambiguity to the meaning of “order,” because

§ 1447(d) “treats Section 1442 and 1443 removal as distinct from other removals.” Appellee Mot. for Partial Dismissal at 12. As the Counties state, because the “except”

clause refers to removals “pursuant to section 1442 or

1443,” not pursuant to those sections in part, it “does not

expressly contemplate the situation in which removal is

done pursuant to one of these sections and other

grounds.” Id. And as a result, it also does not expressly

contemplate the situation in which remand is granted regarding such mixed grounds for removal.

By modifying its reference to appealability in such

way, § 1447(d)’s “except” clause leaves no clear answer to

what scope of appellate review is applied when both enumerated (§ 1442 or 1443) and unenumerated bases for federal subject matter jurisdiction are addressed in the same

remand order. The Lu Junhong court impliedly conceded

as much in asserting that “Section 1447(d) itself authorizes review of the remand order, because the case was removed (in part) pursuant to § 1442.” 792 F.3d at 811 (emphasis added). In other words, to convey its point that the

16a

plain language of § 1447(d) creates plenary review of a remand order upon invocation of a federal officer removal

basis, the Seventh Circuit was forced to modify that language with a clarifying parenthetical entirely absent from

the statutory text. Cf. BP Am., Inc. v. Oklahoma ex rel.

Edmondson, 613 F.3d 1029, 1033 (10th Cir. 2010) (“That

second, italicized condition, however, appears nowhere in

the statute, and we are not at liberty to take our editing

pencils to what Congress has written.”). We thus determine that the specific context in which “order” is used in

the “except” clause creates ambiguity regarding the ambit of our jurisdiction over appeals of mixed remand orders like the one here.

Contextual analysis next requires “examining the subsection’s structure.” In re Woods, 743 F.3d 689, 694 (10th

Cir. 2014); see Davis v. Mich. Dep’t of Treasury, 489 U.S.

803, 809 (1989) (“[T]he words of a statute must be read in

their context and with a view to their place in the overall

statutory scheme.”). That is, § 1447(d)’s primary clause—

“An order remanding a case to the State court from which

it was removed is not reviewable on appeal or otherwise”—must inform the reading of its secondary exception. See In re Woods, 743 F.3d at 694 (finding the statute

at issue “best understood by breaking the provision into

its two principal parts,” amounting to the general rule and

its exception); see also Shell Oil Co. v. Iowa Dep’t of Revenue, 488 U.S. 19, 25 (1988) (reasoning that a statutory

subsection should be “read in its entirety” to divine the

meaning of an exception). Because the structure of

§ 1447(d) exhibits “a scheme whereby a default rule is

subject to an exception, we are guided by the interpretive

principle that exceptions to a general proposition should

be construed narrowly.” In re Woods, 743 F.3d at 699; see

Comm’r of Internal Revenue v. Clark, 489 U.S. 726, 739

(1989) (“In construing provisions . . . in which a general

17a

statement of policy is qualified by an exception, we usually

read the exception narrowly in order to preserve the primary operation of the provision.”). “Flowing from this interpretive principle . . . is the related concept that exceptions must not be interpreted so broadly as to swallow the

rule.” In re Woods, 743 F.3d at 699; see Cuomo v. Clearing

House Ass’n, L.L.C., 557 U.S. 519, 530 (2009) (rejecting an

interpretation of a statutory exception that “would swallow the rule”); Minter v. Prime Equip. Co., 451 F.3d 1196,

1212 (10th Cir. 2006) (reading the impeachment exception

to Fed. R. Evid. 407 “narrowly, lest it swallow the rule”);

In re Annis, 232 F.3d 749, 753 (10th Cir. 2000) (rejecting

a broad construction of a statutory exemption that “would

swallow the rule”).

Application of these guidelines leads us to believe that

the “except” clause must be narrowly construed. See In re

Woods, 743 F.3d at 698. As the Counties note, § 1447(d)’s

“overall thrust,” embodied in its primary clause, “is to impose one of the most categorical bars to reviewability

found anywhere in federal law.” Appellee Mot. for Partial

Dismissal at 12; see Osborn, 549 U.S. at 262 (Scalia, J., dissenting) (noting that “[f]ew statutes read more clearly”

than the primary clause of § 1447(d)); Gravitt v. Sw. Bell

Tel. Co., 430 U.S. 723, 723 (1977) (per curiam) (noting the

clause’s “unmistakabl[e] command[]”); see also Kircher v.

Putnam Funds Tr., 547 U.S. 633, 642 (2006) (“Where the

order is based on one of the [grounds enumerated in

§ 1447(c)], review is unavailable no matter how plain the

legal error in ordering the remand.” (alterations in original) (quoting Briscoe v. Bell, 432 U.S. 404, 413 n.13 (1977)).

“Given that Congress has enacted [this] general rule”

against remand reviewability, “we should not eviscerate

that legislative judgment through an expansive reading of

a somewhat ambiguous exception.” Clark, 489 U.S. at 739.

An expansive reading of § 1447(d)’s ambiguous “except”

18a

clause to allow for plenary review would risk just such an

evisceration: it would let defendants skirt “the primary

operation of the provision,” see id.—its absolute prohibition against appeal of the vast majority of subject matter

jurisdiction-based remands—by simply including a colorable § 1442 or 1443 basis in their petition for removal. Cf.

Fed. Deposit Ins. Co. v. Alley, 820 F.2d. 1121, 1124 (10th

Cir. 1987) (holding that a prior version of Ҥ 1447(c) must

be read disjunctively in order not to eviscerate the thrust

of § 1447(d)”). A broad construction would likewise risk

the exception swallowing the general rule, by turning

§ 1447(d)’s secondary clause into a jurisdictional loophole

allowing appellants to do indirectly what they cannot do

directly. If, alongside the two removal grounds it explicitly exempted, Congress intended the “except” clause to

also lift the general bar to appellate jurisdiction over all

unenumerated subject matter jurisdiction removal

grounds, it could have clearly indicated this intent in the

statutory text—for example, by modifying “pursuant to

1442 or 1443” with “in part.” Cf. Lu Junhong, 792 F.3d at

811; Appellee Mot. for Partial Dismissal at 12.

Because Congress did not indicate any such intent, the

phrase ‘pursuant to section 1442 or 1443’ must be construed “in a way that allows the rule’s exception to function as just that—an exception.” In re Woods, 743 F.3d at

699. Interpreting the “except” clause to create review of

only its two enumerated removal bases, rather than all

other bases rejected by a district court in an order also

addressing those exceptions, serves to preserve, rather

than erode, the “strong legislative mandate” against remand order reviewability, Kennedy, 273 F.3d at 1300, conveyed through § 1447(d)’s “long established policy,” In re

Bear River Drainage Dist., 267 F.2d 849, 851 (10th Cir.

19a

1959).8 In thereby harmonizing § 1447(d)’s venerable

baseline rule with its exception, the narrower interpretation of the scope of review created by the “except” clause

preserves the subsection’s overall structure and prevents

“a serious and unacceptable risk of the exception consuming the rule.” In re Woods, 743 F.3d at 700.

Instead of addressing this statutory context, Defendants argue that the scope of § 1447(d) review is clarified

via extra-statutory context—namely, Yamaha’s interpretation of the word “order” in 28 U.S.C. § 1292(b). As introduced above, that provision permits a district court to certify an interlocutory order to the court of appeals for immediate discretionary review if the order “involves a controlling question of law as to which there is substantial difference of opinion.”9 In Yamaha, the Supreme Court determined whether, under § 1292(b), appellate courts can

“exercise jurisdiction over any question that is included

within the order that contains the controlling question of

law identified by the district court[.]” 516 U.S. at 204. Per

the text of § 1292(b), the Court held that “appellate juris-

8

Cf. Gardner v. Westinghouse Broadcasting Co., 437 U.S. 478, 480

(1978) (narrowly interpreting 28 U.S.C. § 1292(a)(1)’s “exception from

the long-established policy against piecemeal appeals”).

9

Section 1292(b) reads, in relevant part:

When a district judge, in making in a civil action an order not otherwise appealable under this section, shall be of the opinion that

such order involves a controlling question of law as to which there

is substantial ground for difference of opinion and that an immediate appeal from the order may materially advance the ultimate

termination of the litigation, he shall so state in writing in such

order. The Court of Appeals which would have jurisdiction of an

appeal of such action may thereupon, in its discretion, permit an

appeal to be taken from such order, if application is made to it

within ten days after the entry of the order[.]

20a

diction applies to the order certified to the court of appeals, and is not tied to the particular question formulated

by the district court.” Id. at 205. Therefore, “the appellate

court may address any issue fairly included within the certified order.” Id.

Even though Yamaha interpreted a distinct section of

the Judicial Code concerning neither removal nor remand, the Court’s interpretation of “order” might at first

glance appear analogous, as both § 1292(b) and § 1447(d)

contemplate the appealability of district court orders. Cf.

District of Columbia v. Carter, 409 U.S. 418, 421 (1973)

(“At first glance, it might seem logical simply to assume

. . . that identical words used in two related statutes were

intended to have the same effect.”). But Yamaha did not

“purport to establish a general rule governing the scope

of appellate jurisdiction for every statute that uses that

word.” City of Baltimore, 952 F.3d at 460. While “there is

a natural presumption that identical words used in different parts of the same act are intended to have the same

meaning,” Atl. Cleaners & Dyers v. United States, 286

U.S. 427, 433 (1932) (emphasis added), no such presumption applies to the same word used in different statutes.

And even regarding intra-statutory meaning, “the presumption is not rigid”—it “readily yields whenever there

is such variation in the connection in which the words are

used as reasonably to warrant the conclusion that they

were employed in different parts of the act with different

intent.” General Dynamics Land Sys., Inc. v. Cline, 540

U.S. 581, 595 (2004) (quoting Atl. Cleaners, 286 U.S. at

433). Put more succinctly, “[c]ontext counts.” Envtl. Defense v. Duke Energy Corp., 549 U.S. 561, 576 (2007). As

such, the Supreme Court has “several times affirmed that

identical language may convey varying content when used

in different statutes, sometimes even in different provisions of the same statute.” Yates v. United States, 574 U.S.

21a

528, 537 (2015) (plurality opinion); see id. at 537–38 (listing

examples).

Such is the case here: The contextual differences between § 1292(b), which speaks generally of any interlocutory district court order, and § 1447(d), which speaks specifically of remand orders with two express underlying bases, strongly suggest that the word “order” conveys varying content in the two statutes. Section 1292(b) broadly

“permit[s] an appeal to be taken from such order,” referring to “an order not otherwise appealable under this section”—that is, any non-final district court order besides

the three specialized interlocutory varieties outlined in

§ 1292(a). See In re Bear River, 267 F.2d at 851 (stating

that § 1292(b) “applies generally to ‘a civil action’ in which

‘an order not otherwise appealable under this section’ is

made”). Section 1447(d), on the other hand, specifies the

orders exempted from its general bar on reviewability

with multiple identifying layers: “an order remanding a

case . . . removed pursuant to section 1442 or 1443.” (emphasis added). Because § 1292(b) imposes limits on neither the type of order that may be certified for review nor

the underlying basis for such order, an appellate court

reasonably “may address any issue fairly included within

the certified order.” Yamaha, 516 U.S. at 205. But because § 1447(d) does limit the orders that shall be reviewable by both type (remand) and basis (those removed pursuant to § 1442 or 1443), such limiting language is sensibly

read to cabin appellate review to the two enumerated removal bases contemplated by the statute, thereby animating a discrete kind of district court remand order. Cf. Gustafson v. Alloyd Co., 513 U.S. 561, 577 (1995) (“Just as the

absence of limiting language in § 17(a) [of the Securities

Act of 1933] resulted in broad coverage, the presence of

limiting language in § 12(2) requires a narrow construction.”); Jarecki v. G.D. Searle & Co., 367 U.S. 303, 307

22a

(1961) (reading the words surrounding “discovery” in a

section of the tax code to “strongly suggest that a precise

and narrow application was intended”). In short, “there is

such variation in the connection in which the words are

used” in each statute “as reasonably to warrant the conclusion that they were employed . . . with different intent.”

Carter, 409 U.S. at 421 (quoting Atl. Cleaners & Dyers,

286 U.S. at 433).

Strengthening our determination that “order” was

employed with different intent in the two statutes is the

basic observation that, while both § 1292(b) and § 1447(d)

concern appellate review of lower court orders, they point

in opposite directions. As the district court reasoned in rejecting Defendants’ motion for a stay, “§ 1292(b) expressly authorizes appellate review of orders certified by

the district court, while § 1447(d) explicitly bars review of

any kind, with only two specified, narrow exceptions.”

Boulder County II, 423 F. Supp. 3d at 1071; see also Feidt

v. Owens Corning Fiberglas Corp., 153 F.3d 124, 130 (3d

Cir. 1998) (“Section 1447(d) prohibits review of a particular type of district court order, namely a remand order

under section 1447(c), whereas section 1292(b) is a more

general grant of appellate jurisdiction.”). The Fourth Circuit expanded on this fundamental divergence in its opinion rejecting the same Yamaha-based textual argument

advanced by Defendants:

[Section] 1292(b) permits appellate review of important issues before final judgment, but it does not

make otherwise non-appealable questions reviewable.

Reading “order” to authorize plenary review thus

makes sense in the § 1292(b) context, as § 1292(b) only

affects the timing of review for otherwise appealable

questions. But giving the word “order” the same

23a

meaning in the § 1447(d) context would mandate review of issues that are ordinarily unreviewable, period—even following a final judgment.

City of Baltimore, 952 F.3d at 460. We find this analysis

persuasive. Put another way, to read “order” the same

way in both § 1292(b) and § 1447(d) would ignore the distinction between a statute that “governs when an appellate court may review a particular question within its discretion” and one that “limits which issues are ‘reviewable

on appeal or otherwise.’” Id. (quoting § 1447(d)). Ignoring

this distinction between the “when” and “which” of appealability would cut against the Supreme Court’s directive to “take th[e] jurisdictional prescription [of

§ 1447(d)] seriously, however pressing the merits of the

appeal might seem,” Powerex Corp., 551 U.S. at 238–39,

contravene the mandate against expanding the limited

statutory jurisdiction of the federal courts by judicial decree, see Kokkonen v. Guardian Life Ins. Co. of Am., 511

U.S. 375, 377 (1994), and lead us into an interpretive pitfall

the Court has repeatedly flagged—that is, “[t]he tendency to assume that a word which appears in two or more

legal rules, and so in connection with more than one purpose, has and should have precisely the same scope in all

of them,” a tendency that “has all the tenacity of original

sin and must constantly be guarded against,” Wachovia

Bank v. Schmidt, 546 U.S. 303, 319 (2006) (quoting Walter

Wheeler Cook, “Substance” and “Procedure” in the Conflict of Laws, 42 Yale L.J. 333, 337 (1933)).

These differences between the two statutes, expressed

in terms of both structure and function, have important

practical application in assessing appellate jurisdiction, as

both this court and others have noted. For example, In re

Bear River addressed a district court’s use of § 1292(b) to

certify a controlling question of law contained in its order

24a

remanding a case to state court. 267 F.2d at 850. We held

that appellate jurisdiction to review the remand order was

lacking, because § 1447(d)’s specific prohibition overrode

§ 1292(b)’s general grant of jurisdiction: “While the generality of § 1292(b) might seem sufficient to encompass a

remand order, it does not expressly either amend or repeal § 1447(d),” which “applies specially to prohibit appeals from remand orders.” Id. at 851. In addressing the

same issue decades later, the Third Circuit likewise concluded that “the jurisdictional bar of section 1447(d)

trumps the power to grant leave to appeal in section

1292(b),” because “a statute dealing with a narrow, precise, and specific subject is not submerged by a later enacted statute covering a more generalized spectrum.”

Feidt, 153 F.3d at 130 (quoting Radzanower v. Touche

Ross & Co., 426 U.S. 148, 153 (1976)).

Bear River and Feidt provide added authority for our

conclusion that the contextual contrast between the two

statutes—§ 1292(b) being a general grant of appellate jurisdiction, and § 1447(d) being a specific prohibition of it—

leads to the natural conclusion that the same word employed in each provision conveys a distinct meaning. See,

e.g., United States v. Cleveland Indians Baseball Co., 532

U.S. 200, 213 (2001) (phrase “wages paid” means different

things in different parts of Title 26 of the United States

Code); Shell Oil Co., 519 U.S. at 343–44 (term “employee”

means different things in different parts of Title VII);

Carter, 409 U.S. at 420 (“Whether the District of Columbia constitutes a ‘State or Territory’ within the meaning

of any particular statutory or constitutional provision depends upon the character and aim of the specific provision

involved.”). Thus, Yamaha’s construction of “order” in

§ 1292(b) “does not compel symmetrical construction” of

the same word “in the discrete . . . context[]” of § 1447(d).

See Cleveland Indians Baseball Co., 532 U.S. at 213. To

25a

the contrary, our analysis of § 1292(b) and § 1447(d) indicates that while the word “order” in the former statute allows for plenary review of all issues contained in a certified order, its use in the “except” clause contemplates remand orders addressing cases removed solely pursuant to

§ 1442 or 1443, and thus favors limiting remand order review to those specifically delineated removal bases.

Besides marshalling Yamaha, Defendants assert that

our opinion in Coffey v. Freeport McMoran Copper &

Gold, 581 F.3d 1240 (10th Cir. 2009), also “counsels in favor of review of the district court’s entire order, not

simply the ground that permitted appeal.” Appellant Br.

at 11. Like the district court, we are not convinced.

Coffey concerned a provision of CAFA, 28 U.S.C.

§ 1453(c)(1), that states “notwithstanding section 1447(d),

a court of appeals may accept an appeal from an order of

a district court granting or denying a motion to remand a

class action to the State court from which it was removed.”

The defendants in Coffey removed to federal court based

on both CAFA and the Comprehensive Environmental

Response, Compensation and Liability Act (“CERCLA”),

and the district court remanded after determining it

lacked subject matter jurisdiction under either statute.

581 F.3d at 1242. When the defendants appealed that remand order under § 1453(c)(1), the plaintiffs argued that

appellate jurisdiction existed to review only whether removal was proper under CAFA, and not to review “the

district court’s order with respect to the CERCLA determination.” Id. at 1247.

We held that § 1453(c)(1) did allow for discretionary

review of the district court’s determination regarding

both the CAFA and CERCLA removal bases. Id. We

found support for this conclusion in both Yamaha’s inter-

26a

pretation of § 1292(b) and the Seventh Circuit’s application of Yamaha to § 1453(c)(1). In Brill, the Seventh Circuit determined it was “free to consider any potential error in the district court’s decision, not just a mistake in

application of [CAFA],” because “[w]hen a statute authorizes interlocutory appellate review, it is the district court’s

entire decision that comes before the court for review.”

427 F.3d at 451–52 (citing Yamaha, 516 U.S. at 205). In

Coffey, we “agree[d] with the Brill court that Yamaha’s

analysis applies equally to” § 1453(c)(1). 581 F.3d at 1247.

That statute “speaks in terms of the court of appeals accepting an appeal ‘from an order of a district court granting or denying a motion to remand a class action.’” Id.

(quoting § 1453(c)(1)). And it has “no language limiting

the court’s consideration solely to the CAFA issues in the

remand order.” Id.

We went on to hold, however, that while jurisdiction to

review the district court’s disposition of CERCLA removal existed, that jurisdiction was discretionary, and

was best declined under the circumstances. Id. at 1247–

48. We reasoned that if remand had been granted solely

on the CERCLA issue, § 1447(d) would bar review of the

district court’s order. Id. at 1247. Therefore, review of

that issue would not fit within § 1453(c)(1)’s purpose,

which is “to develop a body of appellate law interpreting

[CAFA] without unduly delaying the litigation of class actions.” Id. (alteration in original) (quoting S. Rep. No. 10914, at 49 (2005)).

Defendants thus correctly note that this circuit has

“already applied Yamaha’s rationale to another statutory

provision concerning removal.” Appellant Br. at 14. But

we reject their argument that the removal provision construed in Coffey “contains statutory language that mirrors the language of [§] 1447(d) in all relevant aspects.”

27a

Id. To reiterate, we emphasized in Coffey that § 1453(c)(1)

contains “no language limiting the court’s consideration

solely to the CAFA issues in the remand order.” 581 F.3d

at 1247. However, § 1447(d), as discussed above, does have

limiting language. While § 1453(c)(1) concerns “an order

. . . to remand a class action,” § 1447(d) concerns “an order

remanding a case . . . removed pursuant to section 1442 or

1443.” (emphasis added). “Class action” identifies a broad

category of case, which a defendant can remove to federal

court via any number of bases besides those created by

CAFA.10 See 28 U.S.C. § 1332(d)(1) (defining “class action” as “any civil action filed under [Federal Rule of Civil

Procedure 23] or similar State statute or rule of judicial

procedure authorizing an action to be brought by 1 or

more representative persons”); id. § 1453(b) (CAFA provision easing the requirements for class action removal).

But “removed pursuant to section 1442 or 1443” identifies

specific statutory removal bases that must be addressed

in any corresponding remand order. Thus, while the language of § 1453(c)(1) does not limit the reviewing court to

consider solely “CAFA issues in the remand order,” the

language of § 1447(d) can be read to limit the reviewing

court to consider solely “[§ 1442 or 1443] issues in the remand order.” See Coffey, 581 F.3d at 1247. If, as Defendants assert, § 1453(c)(1) mirrored the language of

§ 1447(d) in all relevant aspects, it would instead speak of

an order to remand a class action “removed pursuant to

section 1453(b),” the CAFA-specific removal provision.

Other textual differences between the statutes also

counsel against applying Coffey’s interpretation of

10

State court class actions were removable prior to the Class Action Fairness Act, provided they met the general requirements of 28

U.S.C. § 1446. CAFA simply made the removal of class actions easier.

See id. § 1453(b).

28a

§ 1453(c)(1) to § 1447(d)’s “except” clause. Section

1453(c)(1) allows for appellate jurisdiction over orders

“granting or denying a motion to remand a class action,”

while the § 1447(d) exceptions call only for appellate review of orders granting such motions. More significantly,

§ 1453(c)(1), like § 1292(b), vests discretion regarding

whether to allow review with the court, see Edmondson,

613 F.3d at 1033, while the appellate jurisdiction created

by the § 1447(d) exceptions is mandatory. Compare

§ 1453(c)(1) (“[A] court of appeals may accept an appeal

from an order of a district court.” (emphasis added)), and

§ 1292(b) (“The Court of Appeals . . . may thereupon, in

its discretion, permit an appeal to be taken from such order.” (emphasis added)), with § 1447(d) (“[A]n order remanding a case . . . removed pursuant to section 1442 or

1443 . . . shall be reviewable.” (emphasis added)). These

differences reflect opposing statutory thrusts: § 1447(d)

being a provision that forecloses appellate jurisdiction,

with two narrow exceptions, and § 1453(c)(1), like

§ 1292(b), being a provision that creates appellate jurisdiction—indeed, that explicitly carves it from § 1447(d)’s

general prohibition. See § 1453(c)(1) (“except that notwithstanding § 1447(d) . . . .”). The distinction between

granting control over appellate jurisdiction to the court,

and ceding such control to the defendant—who is sole

master of her petition for removal—further suggests the

definition of “order” applied to § 1292(b) in Yamaha and

imported to § 1453(c)(1) in Coffey is a poor fit for the

unique context of § 1447(d). In other words, a more expansive scope of jurisdiction is sensible when the appellate

courts may exercise their discretion as gatekeepers, but

29a

not when the defendant holds the key to appellate review.11

One further lesson relevant to our present task can be

drawn from Coffey’s construction of § 1453(c)(1). The appellate discretion granted by that statute over whether to

accept review of remand orders is framed as an either/or

proposition: “a court of appeals may accept an appeal

from an order . . . granting or denying a motion to remand

a class action,” not part of an appeal. 28 U.S.C.

§ 1453(c)(1) (emphasis added). Under Defendants’ reading of “an appeal from an order”—which would create

“appellate review of the whole order, not just of particular

issues or reasons,” Lu Junhong, 792 F.3d at 811—the

court of appeals would be required to exercise its discretion by either accepting review of the entire remand order

(in effect, review of all bases for removal rejected by the

district court and challenged by the defendant), or disclaiming appellate review entirely. It would not be permitted to chart a middle path by choosing to review only “particular issues or reasons” underlying the remand order.

See id.

But such a middle path is exactly what was chosen in

Coffey. We elected to review only one of the rejected bases

for removal challenged by the defendants (the CAFA basis) while declining to exercise jurisdiction over the other

(the CERCLA basis). See 581 F.3d at 1247–48. And we in-

11

Compare, for example, the Yamaha Court’s broad interpretation

of the discretionary appellate jurisdiction created by 28 U.S.C.

§ 1292(b) with the narrow interpretation given by federal courts to

the specific exceptions to the final judgment rule found in § 1291(a),

which create mandatory appellate jurisdiction. See generally United

States v. Solco I, LLC, — F.3d —, No. 19-4089, 2020 WL 3407013

(10th Cir. June 22, 2020).

30a

terpreted § 1453(c)(1) to allow for this jurisdictional partitioning based on our reading of the statutory purpose:

that § 1453(c)(1) was aimed at developing CAFA doctrine

in the courts of appeals, and that review of CERCLA removal would clearly not advance that purpose and would

also not otherwise be allowable under § 1447(d). Id. Likewise here: section 1447(d) was aimed at accelerating litigation on the merits, see Powerex Corp., 551 U.S. at 238,

and reviewing the non-§ 1442 grounds for removal would

clearly not advance that purpose and would also not otherwise be allowable under § 1447(d). Coffey therefore supports disclaiming appellate jurisdiction over aspects of a

remand order “that would otherwise be unreviewable.”

Boulder County II, 423 F. Supp. 3d at 1071; see also Parson v. Johnson & Johnson, 749 F.3d 879, 893 (10th Cir.

2014) (declining to exercise § 1453(c)(1) jurisdiction over

the district court’s decision to remand for lack of diversity

jurisdiction, based in part on the absence of “freestanding

appellate jurisdiction” over that non-CAFA ruling, “a factor we found significant in Coffey”).

In sum, bearing in mind that “[a]mbiguity is a creature

not of definitional possibilities but of statutory context,”

Brown v. Gardner, 513 U.S. 115, 118 (1994), our analysis

of Yamaha and Coffey indicates that the word “order” in

the singular statutory context of § 1447(d)’s “except”

clause should not be read the same as it is in § 1292(b) and

§ 1453(c)(1). Specifically, comparing the three statutes

convinces us that while “order” allows for plenary review

in both § 1292(b) and § 1453(c)(1), the same word used in

§ 1447(d) extends appellate jurisdiction to only the § 1442

or 1443 removal bases addressed in a district court’s remand. Statutory context is thus sufficient to lift the textual ambiguity that cloaks the “except” clause, revealing

the narrower construction of § 1447(d) appealability to be

the proper one.

31a

We recognize, however, that the question of ambiguity

is close, as our extended exegesis necessarily implies. And

the circuit split on which way § 1447(d)’s purportedly plain

meaning cuts also indicates that the “except” clause is “capable of being understood by reasonably well-informed

persons in two or more different senses.” Quarrell, 310

F.3d at 669 (quotation marks omitted). Compare, e.g., Lu

Junhong, 792 F.3d at 812 (calling its “application of

Yamaha Motor and Brill to the word ‘order’ in § 1447(d)

. . . entirely textual”), and Decatur Hosp. Auth. v. Aetna

Health, Inc., 854 F.3d 292, 296 (5th Cir. 2017) (stating its

conclusion that § 1442 removal creates plenary review

“flows from the text of § 1447(d)”), with Glanton, 107 F.3d

at 1047 (dismissing appeal insofar as it challenged non§ 1443 ground “follows from the clear text of § 1447(d)”),

and Jacks, 701 F.3d at 1229 (retaining jurisdiction over

part of remand order addressing § 1442, while rejecting

jurisdiction over part addressing federal common law,

based on “[t]he plain language of § 1447(d)”). In this circuit, such a clear divergence in the appellate courts on

statutory plain meaning is not conclusive evidence of ambiguity, but it is worthy of some consideration. In re S.

Star Foods, Inc., 144 F.3d 712, 715 (10th Cir. 1998). Because the text of § 1447(d) is “arguably ambiguous,” see

Pritchett v. Office Depot, Inc., 420 F.3d 1090, 1097 (10th

Cir. 2005), and has been interpreted inconsistently by the

circuit courts, we venture beyond text and context to seek

further elucidation of the “except” clause’s scope of review. As we now discuss, the additional tools of statutory

construction confirm our primary, context-based reading.

2. Presumption Against Jurisdiction

If an ambiguity is found in the text, “[w]e then look to

presumptions that might aid our analysis.” Pritchett, 420

F.3d at 1094. “Because the jurisdiction of federal courts is

32a

limited, there is a presumption against our jurisdiction.”

Merida Delgado v. Gonzales, 428 F.3d 916, 919 (10th Cir.

2005) (quotation marks omitted); see Kokkonen, 511 U.S.

at 377. This presumption is manifested in “the deeply felt

and traditional reluctance of th[e Supreme] Court to expand the jurisdiction of the federal courts through a broad

reading of jurisdictional statutes.” Romero v. Int’l Term.

Op. Co., 358 U.S. 354, 379 (1959). Thus, “statutes conferring jurisdiction on federal courts are to be strictly construed, and doubts resolved against federal jurisdiction.”

F & S Const. Co. v. Jensen, 337 F.2d 160, 161 (10th Cir.

1964). This includes statutes authorizing federal appellate

jurisdiction. Cal. Coastal Comm’n v. Granite Rock Co.,

480 U.S. 572, 579 (1987); see, e.g., Fornaris v. Ridge Tool

Co., 400 U.S. 41, 42 n.1 (1970) (“[O]ur practice of strict construction of statutes authorizing appeals dictates that we

not give an expansive interpretation to the word ‘State’ [in

28 U.S.C. § 1254].”).

The presumption against jurisdiction also applies with

full force to removal. Interpreting a precursor to the general removal statute, 28 U.S.C. § 1441, the Court determined in Shamrock Oil & Gas Corp. v. Sheets, 313 U.S.

100 (1941), that “[d]ue regard for the rightful independence of state governments, which should actuate federal

courts, requires that they scrupulously confine their own

jurisdiction to the precise limits which the statute has defined.” Id. at 108–09 (quoting Healy v. Ratta, 292 U.S. 263,

270 (1934)); see also Am. Fire & Cas. Co. v. Finn, 341 U.S.

6, 17 (1951) (“The jurisdiction of the federal courts is carefully guarded against expansion by judicial interpretation.” (interpreting § 1441)). As a result, “removal statutes[] are to be narrowly construed in light of our constitutional role as limited tribunals.” Pritchett, 420 F.3d at

1094–95; see also Syngenta Crop Protection, Inc. v. Henson, 537 U.S. 28, 32 (2002).

33a

Pritchett v. Office Depot, Inc. concerned the removal

provisions of CAFA. See 420 F.3d at 1092. We acknowledged in Pritchett that while Congress sought to expand

federal jurisdiction via those provisions, “when that expansion is made effective is what is at issue . . . , and that

is an issue we approach cautiously.” Id.. at 1097 n.7 (citing

Shamrock, 313 U.S. at 108–09); see also Becenti v. Vigil,

902 F.2d 777, 780 (10th Cir. 1990) (acknowledging that

while Congress could authorize removal of tribal court actions against federal officers, at issue was whether it “has

in fact done so” via 28 U.S.C. § 1442, and that the court

“must be careful not to expand the jurisdiction of the federal courts beyond Congressional mandates”). Because

this case concerns the scope of Congress’s desired expansion of the specific exceptions to § 1447(d)’s general bar

on remand order reviewability, we must likewise “approach cautiously.” And while Pritchett and Becenti referenced statutes governing the procedure for removal, rather than “[p]rocedure after removal generally,” see 28

U.S.C. § 1447, their logic should equally apply to

§ 1447(d), which governs removal’s jurisdictional corollary. See also 28 U.S.C. §§ 1441–1455 (containing the

chapter of the Judicial Code addressing “Removal of

Cases from State Courts”).

“Thus, if there is ambiguity as to whether the instant

statute confers federal jurisdiction over this case, we are

compelled to adopt a reasonable, narrow construction.”

Pritchett, 420 F.3d at 1095. By confining appellate review

to only the § 1442 basis for removal, and not the handful

of alternate § 1447(c) bases advanced by Defendants, the

Counties’ reading of § 1447(d) “is clearly the narrower of

the two.” See Conrad v. Phone Directories, Inc., 585 F.3d

1376, 1382 (10th Cir. 2009). And it is also a reasonable

reading, as evidenced by our contextual analysis and the

weight of circuit authority interpreting the “except”

34a

clause. The presumption against jurisdiction thus supports our decision to adopt that reading.

3. Legislative Ratification

A second presumption that can help parse ambiguous

text is the principle of legislative ratification—that “Congress is presumed to be aware of an administrative or judicial interpretation of a statute and to adopt that interpretation when it re-enacts a statute without change,” or

when it “adopts a new law incorporating sections of a prior

law.” Lorillard v. Pons, 434 U.S. 575, 580–81 (1978); see

Consolidation Coal Co. v. Dir., Office of Workers’ Comp.

Programs, 864 F.3d 1142, 1148 (10th Cir. 2017); Bd. of Cty.

Comm’rs v. E.E.O.C., 405 F.3d 840, 845 (10th Cir. 2005).

Both parties rely on this presumption to draw divergent meaning from Congress’s passage of the Removal

Clarification Act of 2011, which authorized appellate review of orders remanding cases removed pursuant to

§ 1442. Defendants contend that this revision to § 1447(d)

incorporated the Yamaha Court’s prior interpretation of

the word “order,” because “Congress is of course presumed to be aware of judicial interpretations of relevant

statutory text.” Appellant Br. at 10. As has been made

clear, however, “Yamaha did not interpret the scope of

§ 1447(d), let alone involve a remand order.” City of Baltimore, 952 F.3d at 460–61. And at the date of the Clarification Act’s passage, every court of appeals to address the

issue in a published opinion interpreted § 1447(d)’s “except” clause to create appellate jurisdiction only over the

asserted § 1443 basis for removal, not the entire remand

order. This included eight circuits12 in a line of authority

See Patel v. Del Taco, Inc., 446 F.3d 996, 998 (9th Cir. 2006); Alabama v. Conley, 245 F.3d 1292, 1293 n.1 (11th Cir. 2001); Davis v.

Glanton, 107 F.3d 1044, 1047 (3d Cir. 1997); Thornton v. Holloway,

12

35a

that continued unbroken following the 1996 decision in

Yamaha. See also County of San Mateo, 960 F.3d at 597

(stating that when the Clarification Act was passed, “no

circuit court had applied Yamaha to § 1447(d) or discussed its applicability in that context”).

Against this “backdrop of unanimous judicial interpretation,” id., the Clarification Act’s sole revision to

§ 1447(d) was to insert “1442 or” before “1443,” 125 Stat.

at 546. Such a minor change evidences Congress’s intent

to adopt the existing appellate consensus regarding

proper construction of the “except” clause. See Lindahl v.

Office of Personnel Mgmt., 470 U.S. 768, 782 (1985) (reasoning that the fact Congress amended a statute “without

explicitly repealing” the established interpretation given

it by the Court of Claims “gives rise to a presumption that

Congress intended to embody [that court’s interpretation] in the amended version”); see also Merrill Lynch,

Pierce, Fenner & Smith Inc. v. Dabit, 547 U.S. 71, 85

(2006) (“[W]hen ‘judicial interpretations have settled the

meaning of an existing statutory provision, repetition of

the same language in a new statute indicates, as a general

matter, the intent to incorporate its . . . judicial interpretations as well.” (quoting Bragdon v. Abbott, 524 U.S. 624,

645 (1998))). Legislative history affirms this intent to incorporate the established contemporaneous judicial interpretation: As the House Report on the Act stated, the revision to § 1447(d) “permit[ted] judicial review of § 1442

cases that are remanded, just as they are with civil rights

cases.” H.R. Rep. No. 112–17, pt. 1, at 7 (2011) (emphasis

added). Cf. Cannon v. Univ. of Chicago, 441 U.S. 677, 696–

70 F.3d 522, 523 (8th Cir. 1995); State Farm Mut. Auto Ins. Co. v.

Baasch, 644 F.2d 94, 97 (2d Cir. 1981); Noel v. McCain, 538 F.2d 633,

635 (4th Cir. 1976); Robertson, 534 F.2d at 65; AppalachianVolunteers, 432 F.2d at 534.

36a

98 (1979) (presuming Congress was aware of the prior federal district and circuit court interpretation of Title VI of

the 1964 Civil Rights Act “and that that interpretation reflects their intent” with respect to Title IX, whose drafters

“explicitly assumed that it would be interpreted and applied as Title VI had been”).

“Absent a clear statutory command to the contrary,

we assume that Congress is ‘aware of the universality of

th[e] practice’ of denying appellate review of remand orders when Congress creates a new ground for removal.”

Things Remembered, 516 U.S. at 128 (alteration in original) (quoting United States v. Rice, 327 U.S. 742, 752

(1946)). Likewise, we will assume Congress was aware of

the universality of denying plenary review of remand orders under the § 1447(d) “except” clause when it augmented that provision with a second narrow statutory avenue for appeal. Thus, if any judicial interpretation of relevant statutory text was ratified by Congress via 2011’s

Removal Clarification Act, it was the unanimous treatment of the scope of appellate review created by

§ 1447(d)’s civil rights exception by three quarters of the

courts of appeals, and not the Yamaha Court’s contrary

reading of a single word in a distinct statute.13

4. Statutory Purpose

“Where the language of a statute is arguably ambiguous, courts also look to public policy considerations to cast

further elucidation on Congress’[s] likely intent.” Pritchett, 420 F.3d at 1097. “Section 1447(d) reflects Congress’s

longstanding ‘policy of not permitting interruption of the

We join the Fourth and Ninth Circuits in reaching this conclusion. See City of Baltimore, 952 F.3d at 460–61; County of San Mateo,

960 F.3d at 597.

13

37a

merits of a removed case by prolonged litigation of questions of jurisdiction of the district court to which the cause

is removed.’” Powerex Corp., 551 U.S. at 238 (quoting

Rice, 327 U.S. at 751); see Dalrymple v. Grand River Dam

Auth., 145 F.3d 1180, 1185 n.8 (10th Cir. 1998) (referencing the “strong congressional policy against review of remand orders ‘in order to prevent delay in the trial of remanded cases by protracted litigation of jurisdictional issues’” (quoting Thermtron, 423 U.S. at 351)); see also Osborn, 549 U.S. at 227 (labeling § 1447(d) an “antishuttling

provision[]”).

Defendants argue that mandating review of the complete remand order “comports with” this statutory purpose of preventing delay, because

[o]nce Congress has permitted appellate review of a

remand order, an appellate court “has been authorized

to take the time necessary to determine the right forum,” and “[t]he marginal delay from adding an extra

issue to a case where the time for briefing, argument,

and decision has already been accepted is likely to be

small.”

Appellant Opp. to Mot. for Partial Dismissal at 9 (quoting

Lu Junhong, 792 F.3d at 813). The leading treatise on federal civil procedure agrees: Although “it has been held

that review [under § 1447(d)] is limited to removability

under § 1443,” it should “instead be extended to all possible grounds for removal underlying the order,” for “[o]nce

an appeal is taken there is little to be gained by limiting

review.” 15A Charles A. Wright et al., Federal Practice

and Procedure § 3914.11, at 706 (2d ed. 2019); see Appellant Opp. to Mot. for Partial Dismissal at 9.

38a

The Counties contend this argument “is not obvious on

its face,” because “a court of appeals may be able to summarily dispose—even in an expedited manner—of a weak

argument under Section 1442 . . . while it may require

more time to consider a range of other, more complex federal jurisdictional issues.” Appellee Mot. for Partial Dismissal at 10; see, e.g., Robertson v. Ball, 534 F.2d 63, 66

n.5 (5th Cir. 1976) (contemplating summary dismissal of

“an appeal from a remand when the removal purportedly

based on § 1443 does not even colorably fall” under that

statute). It was also not obvious to this court in Coffey:

there, we declined to exercise discretionary jurisdiction

over the remand order’s non-CAFA issue because doing

so would conflict with § 1453(c)(1)’s purpose of “develop[ing] a body of appellate law interpreting [CAFA]

without unduly delaying the litigation of class actions.”

581 F.3d at 1247 (second alteration in original) (emphasis

added) (quoting S. Rep. No. 109-14, at 49 (2005)).

This case provides a prime example of the potential

delay occasioned by adding more complex federal jurisdictional issues to the appellate docket. As the district

court reasoned in denying Defendants’ motion to stay the

remand order: “Unlike the situation in [Lu] Junhong,

where ‘the marginal delay from adding an extra issue to

[a] case . . . [’] would be small . . . the time needed to address the numerous additional jurisdictional issues in this

case would be significant.” Boulder County II, 423 F.

Supp. 3d at 1071. In Lu Junhong, besides § 1442, the Seventh Circuit needed to review only one other source of federal jurisdiction (admiralty jurisdiction under 28 U.S.C.

§ 1333). See 792 F.3d at 808. But here, expanding review

to the entire remand order would force this court to grapple with complex judge-made doctrines of “arising under”

jurisdiction—implicating federal common law, contested

and substantial embedded federal issues, see Grable, 545

39a

U.S. at 312–13, and the complete preemption doctrine14—

in addition to more “bespoke jurisdictional law,” Rhode

Island v. Chevron Corp., 393 F. Supp. 3d 142, 151 (D.R.I.

2019), pertaining to federal enclaves and the outer continental shelf. The pages of the Federal Supplement are

rapidly filling with the extended discussions occasioned

by application of these doctrines to global warming-based

state law actions. See, e.g., Boulder County I, 405 F. Supp.

3d at 956–79; Mayor & City Council of Baltimore v. BP

P.L.C., 388 F. Supp. 3d 538, 551–67 (D. Md. 2019), aff’d,

952 F.3d 452 (4th Cir. 2020).

It is thus not apparent that expanding the scope of

§ 1447(d) review will lead to merely marginal delay in litigation on the merits. To the contrary, the extra analysis

necessitated by a broad interpretation has significant potential to foment “protracted litigation of jurisdictional is-

Federal district courts have come out differently on these meaty

issues of federal question jurisdiction, further demonstrating the potential for delay if this court was forced to weigh in on their proper

resolution. Compare California v. BP P.L.C., Nos. 17-06011 & 1706012, 2018 WL 1064293 (N.D. Cal. Feb. 27, 2018) (unpublished)

(denying remand of global warming-related action and exercising federal subject matter jurisdiction based on federal common law), rev’d

sub nom City of Oakland v. BP P.L.C., 960 F.3d 570 (9th Cir. 2020),

and City of New York v. BP P.L.C., 325 F. Supp. 3d 466 (S.D.N.Y.

2018) (finding federal common law governed state common law global

warming-related claims), with Bd. of Cty.Comm’rs of Boulder County

v. Suncor Energy (U.S.A.) Inc., 405 F. Supp. 3d 947 (D. Colo. 2019)

(granting remand of similar global warming action and rejecting jurisdiction under federal common law, Grable, and complete preemption), Rhode Island v. Chevron Corp., 393 F. Supp. 3d 142 (D.R.I.

2019) (same), Mayor & City Council of Baltimore v. BP P.L.C., 388

F. Supp. 3d 538 (D. Md. 2019) (same), aff’d, 952 F.3d 452 (4th Cir.

2020), and County of San Mateo v. Chevron Corp., 294 F. Supp. 3d

934 (N.D. Cal. 2018) (same), aff’d, 960 F.3d 586 (9th Cir. 2020).

14

40a

sues,” Thermtron, 423 U.S. at 351, “and prolong the interference with state jurisdiction that § 1447(d) clearly seeks

to minimize,” Lambeth, 443 F.3d at 760, thereby frustrating the statute’s “clear Congressional policy of expedition,” Appalachian Volunteers, Inc. v. Clark, 432 F.2d

530, 533 (6th Cir. 1970). Statutory purpose thus lends further support to our conclusion that the review granted by

§ 1447(d)’s “except” clause must be confined to the enumerated removal bases, for “[a] textually permissible interpretation that furthers rather than obstructs the document’s purpose should be favored.” Medina v. Catholic

Health Initiatives, 877 F.3d 1213, 1226 (10th Cir. 2017)

(quoting Antonin Scalia & Bryan A. Garner, Reading Law

63–65 (2012)). This is especially so because a holding that

only the explicit exceptions in § 1447(d) are appealable,

besides shortening the travel time of this particular “intercourt shuttle,” Osborn, 549 U.S. at 244, could also prevent some gratuitous trips entirely—for example, by encouraging parties with weak § 1442 or 1443 removal arguments to forego appeals,15 or omit those two bases for removal in the first place.

The potential for this latter result speaks to the Counties’ “moral hazard” policy argument—that allowing for

an expanded scope of review “would encourage removing

parties to assert frivolous federal officer claims in order

to bring otherwise nonappealable removal arguments to

the court of appeals.” Appellee Mot. for Partial Dismissal

at 10. Similar moral hazard issues of appealability have

15

Cf. Coffey v. Freeport McMoran Copper & Gold, 581 F.3d 1240,

1242 n.2 (10th Cir. 2009) (“Defendants also argued that removal was

authorized under 28 U.S.C. § 1442(a)(1). The district court disagreed,

and that portion of the district court’s decision [wa]s not . . . challenged on appeal.”). Coffey was decided before Congress expanded

§ 1447(d)’s “except” clause to encompass § 1442 removal.

41a

not escaped judicial notice. In Abney v. United States, 431

U.S. 651 (1977), the Supreme Court held that a criminal

defendant may immediately appeal a district court’s rejection of her motion to dismiss an indictment on double

jeopardy grounds “based on the special considerations

permeating claims of that nature.” Id. at 663. But it further determined that “obviously, such considerations do

not extend” to allow the appeal of “other claims presented

to, and rejected by, the district court in passing on the accused’s motion to dismiss.” Id. “Any other rule would encourage criminal defendants to seek review of, or assert,

frivolous double jeopardy claims in order to bring more

serious, but otherwise nonappealable questions to the attention of the courts of appeals prior to conviction and

sentence.” Id. And while Abney was confined to the criminal context, “the concern expressed in Abney . . . bears on

civil cases as well.” Swint v. Chambers Cty. Comm’n, 514

U.S. 35, 49–50 (1995).

In Lu Junhong, the Seventh Circuit reasoned that

sanctions and summary resolutions are sufficient tools to

combat citing § 1442 or 1443 in a notice of removal merely

as “a hook to allow appeal of some different subject.” 792

F.3d at 813; see also Wright et al., supra, § 3914.11, at 706

(acknowledging the “plausible concern” that interpreting

§ 1447(d) to allow for review of otherwise nonreviewable

removal bases would lead to frivolous removal arguments,

but arguing that “[s]ufficient sanctions are available to deter” that “sorry possibility”). But should the scope of

§ 1447(d) review be expanded, we harbor serious doubt

that either tool will prove dexterous enough to prevent the

delay of litigation on the merits Congress so clearly

sought to avoid. As one Amicus notes, “[i]f alleging federal-officer removal opens the door to appellate review of

all other asserted bases for removal, no lawyer would neglect to find a defensible, if inadequate, way to assert that

42a

peculiar form of removal to avoid the bar on interlocutory

appeal for all other justifications for removal.” Brief of

Nat’l Lg. of Cities as Amicus Curiae at 17 n.4; cf. Robertson, 534 F.2d at 66 n.5 (expressing concern that appeals

from remands of removals under § 1443 could “be used as

a dilatory tactic”); County of San Mateo v. Chevron Corp.,

294 F. Supp. 3d 934, 939 (N.D. Cal. 2018) (describing defendants’ § 1442 argument as “dubious” in a case featuring substantially similar state law global warming-related

causes of action and asserted grounds for removal), aff’d,

960 F.3d 586 (9th Cir. 2020).

***

In sum, while the text of § 1447(d)’s “except” clause is

arguably ambiguous, statutory context clarifies that the

word “order” in that provision must be construed differently than the word “order” in 28 U.S.C. § 1292(b) and

§ 1453(c)(1). And the proper construction of the statute is

the narrower one adopted by the majority of federal circuits. We therefore hold that when a district court issues

a remand order premised on a § 1447(c) ground, we are

empowered to review that order only to the extent it addresses the removal bases explicitly excepted from

§ 1447(d)—in this case, removal under 28 U.S.C. § 1442.

III. FEDERAL OFFICER REMOVAL

Having determined 28 U.S.C. § 1447(d) supplies appellate jurisdiction only to review the district court’s rejection of removal based on federal officer jurisdiction, we

now address that issue. Questions of removal are reviewed de novo. Frederick v. Hartford Underwriters Ins.

Co., 683 F.3d 1242, 1245 (10th Cir. 2012). ExxonMobil, as

the party asserting federal officer removal, bears the burden of establishing jurisdiction by a preponderance of the

43a

evidence.16 Dutcher v. Matheson, 733 F.3d 980, 985 (10th

Cir. 2013). This burden is met by “a substantial factual

showing,” Wyoming v. Livingston, 443 F.3d 1211, 1225

(10th Cir. 2006), that supports “‘candid, specific and positive’ allegations,” In re MTBE Prods. Liab. Litig., 488

F.3d 112, 130 (2d Cir. 2007) (quoting Willingham v. Morgan, 395 U.S. 402, 408 (1969)).

The federal officer removal statute permits removal of

state court actions filed against “any officer (or any person acting under that officer) of the United States or of

any agency thereof, in an official or individual capacity, for

or relating to any act under color of such office.” 28 U.S.C.

§ 1442(a)(1). The statute’s “‘basic purpose’ is to protect

against the interference with federal operations that

would ensue if a state were able to arrest federal officers

and agents acting within the scope of their authority and

bring them to trial in a state court for an alleged state-law

offense.” City of Baltimore, 952 F.3d at 461 (quoting Watson, 551 U.S. at 150). Three fears animate this purpose:

that “[s]tate-court proceedings may reflect ‘local prejudice’ against unpopular federal laws or federal officials,”

Watson, 551 U.S. at 150, “disable federal officials from

taking necessary action designed to enforce federal law,”

id. at 152, or “deprive federal officials of a federal forum

in which to assert federal immunity defenses,”17 id. at 150.

16

Suncor Energy asserts no basis for federal officer removal. See

Appellant Br. at 38–39. However, unlike the typical removal petition,

which requires joinder of all defendants, § 1442 allows for independent removal of an entire case by only one of several named defendants. See Akin v. Ashland Chem Co., 156 F.3d 1030, 1034 (10th Cir.

1998).

Our precedent elevates this statutory concern above others. See

Christensen v. Ward, 916 F.2d 1462, 1484 (10th Cir. 1990) (“The primary purpose for the removal statute is to assure that defenses of

official immunity applicable to federal officers are litigated in federal

17

44a

In short, “the removal provision was an attempt to protect

federal officers from interference by hostile state courts.”

Willingham, 395 U.S. at 405. Unlike other removal statutes, it should “be liberally construed to give full effect to

th[at] purpose[].” Colorado v. Symes, 286 U.S. 510, 517

(1932).

Section 1442(a)(1) removal can apply to private persons “who lawfully assist” federal officers “in the performance of [their] official duty,” Davis v. South Carolina,

107 U.S. 597, 600 (1883), meaning the private person must

be “authorized to act with or for [federal officers or

agents] in affirmatively executing duties under . . . federal

law,” Watson, 551 U.S. at 151 (alterations in original)

(quoting City of Greenwood v. Peacock, 384 U.S. 808, 824

(1966)). And § 1442(a)(1) has also been interpreted to allow removal by private corporations that meet the statutory requirements. See, e.g., Isaacson v. Dow Chem. Co.,

517 F.3d 129, 135–36 (2d. Cir. 2008).

A private corporation may remove a case under

§ 1442(a)(1) if it can show: (1) that it acted under the

direction of a federal officer; (2) that there is a causal

nexus between the plaintiff’s claims and the acts the

private corporation performed under the federal officer’s direction; and (3) that there is a colorable federal defense to the plaintiff’s claims.18

court.” (citing Willingham v. Morgan, 395 U.S. 402, 406–07 (1969));

see also Jefferson County v. Acker, 527 U.S. 423, 447 (1999) (Scalia,

J., concurring in part) (asserting the “main point” of the statute “is to

give officers a federal forum in which to litigate the merits of immunity defenses”).

A colorable federal defense “constitutes the federal law under

which the action against the federal officer arises for Art. III purposes.” Mesa v. California, 489 U.S. 121, 136 (1989). This is required

because the statute itself does not create a federal question, but

18

45a

Greene v. Citigroup, Inc., No. 99-1030, 2000 WL 647190,

at *2 (10th Cir. May 19, 2000) (unpublished); see also Sawyer v. Foster Wheeler LLC, 860 F.3d 249, 254 (4th Cir.

2017).

ExxonMobil asserts federal officer removal jurisdiction based on its long-term mining of the Outer Continental Shelf (“OCS”) for fossil fuels under government leases.

Appellant Br. at 38; see, e.g., App. 49, 62 (“Oil and Gas

Lease of Submerged Lands Under the Outer Continental

Shelf Lands Act.”). To address this argument, we first lay

out the regulatory background of these mineral leases.

The OCS “is a vast underwater expanse” beginning

several miles off the coastline and extending seaward for

roughly two hundred miles. Ctr. for Sustainable Econ. v.

Jewell, 779 F.3d 588, 592 (D.C. Cir. 2015). Its “subsoil and

seabed appertain to the United States and are subject to

its jurisdiction and control.” 43 U.S.C. § 1331(a). “Billions

of barrels of oil and trillions of cubic feet of natural gas lie

beneath the OCS.” Jewell, 779 F.3d at 592. Pursuant to

the Outer Continental Shelf Lands Act (“OCSLA”), the

United States Department of the Interior (“DOI”) administers a federal leasing program to develop and exploit the

oil and gas resources in these submerged lands in a sustainable manner. App. 38; see 43 U.S.C. §§ 1331–1356(b);

Jewell, 779 F.3d at 592 (“The [OCSLA] created a framework to facilitate the orderly and environmentally responsible exploration and extraction of oil and gas deposits on

the OCS.”). Under OCSLA, the Interior Secretary “is authorized to grant to the highest responsible qualified bidder or bidders by competitive bidding . . . any oil and gas

“merely serves to overcome the ‘well-pleaded complaint’ rule which

would otherwise preclude removal even if a federal defense were alleged.” Id.

46a

lease” on the OCS, in exchange for payment of royalties.

43 U.S.C. § 1337(a)(1); see County of San Mateo, 960 F.3d

at 602 (“[T]he government grants the lessee the right to

explore and produce oil and gas resources in the submerged lands of the outer Continental Shelf, and in exchange the lessee agrees to pay the government rents and

royalties.”). ExxonMobil has participated in this competitive leasing program for decades and continues to conduct

oil and gas operations under OCS leases. App. 40; see App.

61 (June 2016 DOI letter notifying ExxonMobil that its

“bid for the [OCS] block described above is accepted”);

App 62 (Ten-year ExxonMobil OCS lease starting July 1,

2016).

OCS lessees are required to conduct drilling in accordance with federally approved exploration, development,

and production plans and conditions. App. 64 § 9 (2016

lease exemplar); see 30 C.F.R. §§ 550.200–.299 (outlining

the plans and documents that must be submitted to and

approved by the Bureau of Ocean Energy Management

before starting to drill under OCS leases). These plans

must “conform to sound conservation practices to preserve, protect, and develop minerals resources and maximize the ultimate recovery of hydrocarbons from the

leased area.” App. 64 § 10. Lessees are obligated to “exercise diligence in the development of the leased area and in

the production of wells located thereon,” to “prevent unnecessary damage to, loss of, or waste of leased resources,” and to “comply with all applicable laws, regulations and orders related to diligence, sound conservation

practices and prevention of waste.” App. 64 § 10. A much

earlier OCS lease, from 1979, further stated that “[a]fter

due notice in writing, the Lessee shall drill such wells and

produce at such rates as the Lessor may require in order

that the Leased Area or any part thereof may be properly

47a

and timely developed and produced in accordance with

sound operating principles.” App. 50 § 10.

DOI officials reserve the right to obtain “prompt access” to facilities and records of private OCS lessees for

the purpose of federal safety, health, or environmental inspections. App. 64 § 12 (2016 lease). The federal government can precondition an OCS lease on a right of first refusal to purchase all production “[i]n time of war or when

the President of the United States shall so prescribe.”

App. 68 § 15(d). The government also mandates that

twenty percent of all crude or natural gas produced pursuant to OCS leases be offered to small or independent

refiners, “as defined in the Emergency Petroleum Allocation Act of 1973.” App. 68 § 15(c).

ExxonMobil argues that its participation in the OCS

leasing program under these terms and conditions satisfies the “acting under” element of federal officer removal.

Appellant Br. at 38. We disagree.

“The statutory phrase ‘acting under’ describes ‘the

triggering relationship between a private entity and a federal officer.’” City of Baltimore, 952 F.3d at 462 (quoting

Watson, 551 U.S. at 149). While “[t]he words ‘acting under’ are broad,” they are “not limitless.” Watson, 551 U.S.

at 147. In this context, “under” describes a relationship

between private entity and federal superior typically involving “subjection, guidance, or control.” Id. at 151 (quoting Webster’s New International Dictionary 948 (2d ed.

1953)). Thus, a “private person’s ‘acting under’ must involve an effort to assist, or to help carry out, the duties or

tasks of the federal superior.” Id. at 152. This “help or assistance necessary to bring a private person within the

scope of the statute does not include simply complying

with the law . . . , even if the regulation is highly detailed

48a

and even if the private firm’s activities are highly supervised and monitored.” Id. at 152–53. Rather, “there must

exist a ‘special relationship’ between” private firm and

federal superior that goes beyond the fulfillment of regulatory or statutory requirements. Isaacson, 517 F.3d at

137 (quoting Watson, 551 U.S. at 157).

In Watson, the Supreme Court addressed whether the

Philip Morris Companies were “acting under” a federal

officer or agency when they advertised cigarettes as

“light” in compliance with detailed Federal Trade Commission supervision of cigarette testing. 551 U.S. at 146–

47. As private contracting was not at issue, the Court disclaimed deciding “whether and when particular circumstances may enable private contractors to invoke the statute.” Id. at 154. In an effort to establish the necessary

amount of federal direction, however, the defendants

highlighted various lower court cases that held government contractors could invoke § 1442 removal, “at least

when the relationship between the contractor and the

Government is an unusually close one involving detailed

regulation, monitoring, or supervision.” Id. at 153. The

Court unanimously rejected this attempt to analogize the

highlighted “close supervision” over contractors to “intense regulation” of firms, because “the private contractor in such cases is helping the Government to produce an

item that it needs.” Id. That is, “[t]he assistance that private contractors provide federal officers goes beyond simple compliance with the law and helps officers fulfill other

basic governmental tasks.” Id.

The Watson Court illustrated this point by reference

to a Fifth Circuit case, Winters v. Diamond Shamrock

Chemical Co., 149 F.3d 387 (5th Cir. 1998). Winters involved tort claims brought against chemical firms premised on their production of the defoliant known as Agent

49a

Orange under a Department of Defense contract for use

in the Vietnam War. The Fifth Circuit concluded that both

the “acting under” and causal nexus elements needed for

a private company to remove under § 1442 were satisfied,

due to “the government’s detailed specifications concerning the make-up, packaging, and delivery of Agent Orange, the compulsion to provide the product to the government’s specifications, and the on-going supervision the

government exercised over the formulation, packaging,

and delivery of Agent Orange.” Id. at 400. The chemical

companies “provid[ed] the Government with a product

that it used to help conduct a war,” and “at least arguably

. . . performed a job that, in the absence of a contract with

a private firm, the Government itself would have had to

perform.” Watson, 551 U.S. at 154. As such, they had a

“special relationship” with the government, see id. at 157,

whereby they “help[ed] carry out[] the duties or tasks of

the federal superior,” id. at 152.

The Phillip Morris Companies also claimed § 1442 removal was appropriate because the FTC had delegated

testing authority to an industry-financed laboratory and

the companies were “acting pursuant to that delegation.”

Id. at 153–54. The Court disagreed, finding “no evidence

of any delegation of legal authority from the FTC to the

industry association to undertake testing on the Government agency’s behalf.” Id. at 156.

Watson teaches that a private contractor’s compliance

with statutory or regulatory mandates, even if complex, is

insufficient to satisfy the “acting under” requirement for

federal officer removal. Rather, the company must agree

to help carry out the duties of the federal superior under

that superior’s strict guidance and control. See In re

MTBE, 488 F.3d at 125 (“describing the need for some

50a

government intervention or control, other than that contemplated by a generally applicable regulatory scheme, as

‘regulation plus’” (quoting Bakalis v. Crossland Sav.

Bank, 781 F. Supp. 140, 145 (E.D.N.Y. 1991))). In addition, this closely supervised and directed work must help

federal officers fulfill basic government needs, accomplish

key government tasks, or produce essential government

products—that is, it must stand in for critical efforts the

federal superior would be required to undertake itself in

the absence of a private contract, with wartime production

being the paradigmatic example. Compare Ruppel v. CBS

Corp., 701 F.3d 1176, 1181 (7th Cir. 2016) (“Cases in which

the Supreme Court has approved removal involve defendants working hand-in-hand with the federal government

to achieve a task that furthers an end of the federal government.”), with County of San Mateo, 960 F.3d at 600

(“[A] person is not ‘acting under’ a federal officer when

the person enters into an arm’s- length business arrangement with the federal government or supplies it with

widely available commercial products or services.”). Alternately, the requisite “special relationship” can be established through the explicit delegation of legal authority

to act on the federal superior’s behalf.

Here, ExxonMobil’s OCS leases do not contemplate

the “close supervision of the private entity by the Government,” Isaacson, 517 F.3d at 137, needed to bring a federal contractor relationship within these strict parameters. We agree with the district court’s determination that

under the OCS leases “the government does not control

the manner in which Defendants drill for oil and gas, or

develop and produce the product.” 405 F. Supp. 3d at 976;

accord City of Baltimore, 952 F.3d at 466 (“[T]he leases

do not appear to dictate that Defendants extract fossil

fuels in a particular manner. . . . [n]or do they appear to

vest the government with control over the composition of

51a

oil or gas to be refined and sold to third parties.” (citations

and quotation marks omitted)); see also County of San

Mateo, 960 F.3d at 602–03 (holding the OCS leases do not

require lessees to act under the government’s “close direction”). As the physical mining of OCS fuels is not subject to DOI’s “detailed and ongoing control,” see Betzner

v. Boeing Co., 910 F.3d 1010, 1015 (7th Cir. 2018), and as

OCS-produced fuel need not conform to “highly detailed

. . . specifications,” see Sawyer, 860 F.3d at 253, ExxonMobil was not “acting under” a federal superior within the

meaning of the federal officer statute. Compare Bennett

v. MIS Corp., 607 F.3d 1076, 1087–88 (6th Cir. 2010) (holding a mold remediation firm whose workers were directly

supervised by on-site federal officers and escorted at all

times by federal personnel, and whose “closely monitored” contract work was subject to “explicit parameters

for site containment and waste disposal,” satisfied the

“acting under” requirement), with Cabalce v. Thomas E.

Blanchard & Assocs., 797 F.3d 720, 728 (9th Cir. 2015)

(holding a company that contracted to store and destroy

fireworks seized by the government did not act under a

federal officer due to a “lack of any evidence of the requisite federal control or supervision over the handling of the

seized fireworks”).

ExxonMobil disputes the district court’s finding of insufficient government control by asserting that “the operative leases explicitly afford the federal government the

right to control the rates of mining and production.” Appellant Br. at 40. It supports this contention by reference

to a single clause in the 1979 lease: “After due notice in

writing, the Lessee shall drill such wells and produce at

such rates as the Lessor may require in order that the

leased area . . . may be properly and timely developed[.]”

App. 50 § 10. There is no similar clause in the 2016 lease,

52a

however, and no indication that the 1979 language remains in effect. See App. 50 § 3 (stating that the 1979 lease

shall cover an initial five-year period, to be extended “so

long thereafter” as production from or operation on the

leased parcel continues). Additionally, there is no showing

the government ever gave notice of its intent to direct

ExxonMobil’s drilling activity or rates of production by

means of the OCS leases. The same is true with respect to

the government’s wartime right of first refusal over ExxonMobil’s OCS output. Even if the exercise of these rights

could create the necessary level of federal supervision, an

issue we do not decide, ExxonMobil points us to no authority for the proposition that the reservation of such

rights alone creates the “special relationship” needed for

a private firm to invoke § 1442. Cf. Mays v. City of Flint,

871 F.3d 437, 447 (6th Cir. 2017) (disagreeing with the argument that the government’s potential ability to intervene supports the invocation of federal officer removal in

the absence of actual intervention). As a result, ExxonMobil has not met its “burden of providing ‘candid, specific

and positive’ allegations that [it] w[as] acting under federal officers.” In re MTBE, 488 F.3d at 130 (quoting

Willingham, 395 U.S. at 408); see also City of Baltimore,

952 F.3d at 466 n.9 (“[T]he lack of any specificity as to federal direction leaves us unable to conclude that the leases

rise to the level of an unusually close relationship, as required by the first ‘acting under’ prong.”).

ExxonMobil’s other attempts to parse the lease language in support of federal officer removal are likewise

unavailing, see Cabalce, 797 F.3d at 729, because most of

the contractual terms “are mere iterations of the

OCSLA’s regulatory requirements.” City of Baltimore,

952 F.3d at 465; accord County of San Mateo, 960 F.3d at

603; see, e.g., 43 U.S.C. § 1337(a)(1) (authorizing OCS

leases to be granted “under regulations promulgated in

53a

advance”); Jewell, 779 F.3d at 594 (describing OCSLA as

“a statute with a ‘structure for every conceivable step to

be taken’ on the path to development of an OCS leasing

site.” (quoting California v. Watt, 668 F.2d 1290, 1297

(D.C. Cir. 1981))). For example, the plans and documents

required by DOI to drill under OCS leases, which ExxonMobil advances as evidence of the government’s “extensive control,” Appellant Br. at 39, are detailed in Bureau

of Ocean Energy Management regulations. See 30 C.F.R.

§§ 550.211–.228 (“Contents of Exploration Plans”); id.

§ 550.241–.262 (“Contents of Development and Production Plans and Development Operations Coordination

Documents”). And other lease terms cited by ExxonMobil

as proof of close federal oversight—the requirement that

a fifth of OCS production be offered to small or independent refiners, and the government’s reservation of a wartime right of first refusal—are also duplications of regulatory details furnished by OCSLA. See 43 U.S.C.

§ 1337(b)(7) (OCS lessees must “offer 20 per centum of the

crude oil, condensate, and natural gas liquids produced on

such lease . . . to small or independent refiners”); id.

§ 1341(b) (“In time of war, or when the President shall so

prescribe, the United States shall have the right of first

refusal to purchase at the market price all or any portion

of any mineral produced from the outer Continental

Shelf.”). Compliance with such legal requirements, no

matter their complexity, cannot by itself create the “acting under” relationship required to support a federal officer claim. Watson, 551 U.S. at 153. Something more is

needed—there must be “regulation plus.” In re MTBE,

488 F.3d at 125 (quoting Bakalis, 781 F. Supp. at 145).

And here, this “plus” factor is absent from what appear to

be “standard-form” leases containing mostly “boilerplate”

provisions. See County of San Mateo, 960 F.3d at 602;

City of Baltimore, 952 F.3d at 465.

54a

A holding that “simple compliance” with the statutory

and regulatory requirements embedded in these standard-form, boilerplate lease terms satisfies the “acting under” relationship would risk “expand[ing] the scope of the

statute considerably” to include “state-court actions filed

against private firms in many highly regulated industries.” See Watson, 551 U.S. at 153 (“Neither language,

nor history, nor purpose lead us to believe that Congress

intended any such expansion.”). Such a result is incompatible with the Watson Court’s careful articulation of when

a private firm can invoke federal officer removal. We thus

agree with the Fourth and Ninth Circuits that “the willingness to lease federal property or mineral rights to a

private entity for the entity’s own commercial purposes,

without more[,]’ cannot be ‘characterized as the type of

assistance that is required’ to show that the private entity

is ‘acting under’ a federal officer.” County of San Mateo,

960 F.3d at 603 (quoting City of Baltimore, 952 F.3d at

465).

Additionally, the OCS leases do not meet the “acting

under” parameters because they do not call for production

specially conformed to government use—the type of contract that “involve[s] an effort to assist, or to help carry

out, the duties or tasks of the federal superior.” Watson,

551 U.S. at 152. See Sawyer, 860 F.3d at 255 (stating that

courts often find the “acting under” requirement satisfied

“where a contractor seeks to remove a case involving injuries arising from equipment that it manufactured for

the government”); Mays, 871 F.3d at 445 (“[A] government contractor entitled to removal would presumably be

contractually required to follow the federal government’s

specifications in making products or providing services.”).

55a

In the Agent Orange cases, for example, the military

provided precise specifications to private firms that “included use of the two active chemicals in unprecedented

quantities for the specific purpose of stripping certain areas of Vietnam of their vegetation.” Winters, 149 F.3d at

399; see also Betzner, 910 F.3d at 1015 (holding that Boeing “acted under the military’s detailed and ongoing control” in “manufactur[ing] heavy bomber aircraft for the

United States Air Force”); Sawyer, 860 F.3d at 253, 255

(holding that a contractor “acted under the Navy” in manufacturing boilers “to match highly detailed ship specifications and military specifications provided by the

Navy”). Here, ExxonMobil is not tailoring its output to

detailed federal formulations customized to meet pressing

federal needs. Rather, it is leasing federal land to facilitate commercial production of a standardized, undifferentiated consumer product. See Jewell, 779 F.3d at 607 (determining DOI’s decision “not to earmark the point of consumption of OCS-derived energy” was rational “[b]ecause

oil and natural gas are fungible and traded on integrated

global markets”). And even assuming federal authorities

purchase some of the fuel extracted by ExxonMobil from

the OCS—the same as other buyers on the global markets—supplying the government “with widely available

commercial products or services” does not create the special relationship or assistance necessary to trigger “acting

under” removal. County of San Mateo, 960 F.3d at 600.

Clearly, then, this “arrangement is not the procurement

relationship that in previous cases has allowed a private

firm to enjoy the benefit of federal officer removal.” City

of Walker v. Louisiana, 877 F.3d 563, 571 (5th Cir. 2017).

Lastly, ExxonMobil cannot show the delegation of legal authority that the Watson Court hypothesized would

be sufficient to conclude a private corporation was “acting

56a

under” a government superior. No highlighted lease provision “establish[es] the type of formal delegation that

might authorize [ExxonMobil] to remove the case.” Watson, 551 U.S. at 156; see County of San Mateo, 960 F.3d

at 602 (“The leases do not require that lessees act on behalf of the federal government.”). And “neither Congress

nor federal agencies normally delegate legal authority to

private entities without saying that they are doing so.”

Watson, 551 U.S. at 157.

Our determination that ExxonMobil was not “acting

under” federal officers in drilling pursuant to OCS leases

is not altered by the OCS’s status as a “vital national resource reserve held by the Federal Government for the

public.” 43 U.S.C. § 1332(3). While the leasing of OCS

mining rights at least arguably implicates national energy

needs, the facilitation of fossil fuel resource development

by private companies is not a critical federal function in

the same vein as law enforcement, see Watson, 551 U.S.

at 151 (referencing a “private person” who “acts as an assistant to a federal official in helping that official to enforce federal law”); Fidelitad, Inc. v. Insitu, Inc., 904 F.3d

1095, 1099 (9th Cir. 2018) (stating that the “paradigm” for

a private party’s § 1442 removal is a “person acting under

the direction of a federal law enforcement officer”), military manufacturing, see Papp v. Fore-Kast Sales Co., 842

F.3d 805, 813 (3d Cir. 2016) (labeling a government contract to manufacture military aircraft “an archetypal

case” of a private firm acting under a federal officer), or

wartime production, see Isaacson, 517 F.3d at 137 (reasoning that defendants “provide[d] a product that the

Government was using during war” and that it otherwise

“would have had to produce itself”). This conclusion is “a

matter of statutory purpose,” Watson, 551 U.S. at 152: As

the Ninth Circuit reasoned in rejecting an identical § 1442

removal argument, by leasing government land for the

57a

commercial extraction of fossil fuels, private oil and gas

firms are not “engaged in an activity so closely related to

the government’s function” that they might face the “significant risk of state-court ‘prejudice’” that animates federal officer removal. County of San Mateo, 960 F.3d at 603

(quoting Watson, 551 U.S at 152); see Watson, 551 U.S. at

152 (“When a company subject to a regulatory order (even

a highly complex order) complies with the order, it does

not ordinarily create a significant risk of state-court ‘prejudice.’”).19

While “private contractors performing tasks for the

government are sometimes covered under section 1442,”

ExxonMobil “take[s] this idea too far.” Panther Brands,

LLC v. Indy Racing Lg., LLC, 827 F.3d 586, 590 (7th Cir.

2016). The OCS leases “represent arms-length commercial transactions whereby ExxonMobil agreed to certain

terms (that are not at issue in this case) in exchange for

the right to use government-owned land for [its] own commercial purposes.” Boulder County I, 405 F. Supp. 3d at

977. Such mineral rights leases—which call for neither

products nor services specially tailored to meet fundamental federal needs—do not fulfill the “acting under” el-

19

State-court claims against oil and gas firms operating under federal mineral leases also do not “disable federal officials from taking

necessary action designed to enforce federal law.” Watson, 551 U.S.

at 152. As an example of this risk, Watson cited Tennessee v. Davis,

100 U.S. 257 (1879), where a federal revenue officer was charged with

murder in state court for killing a man during a sanctioned raid on an

illegal distillery. That type of hostile provincial proceeding, and others that might similarly “paralyze the operations of the [federal] government,” id. at 263, is inapposite to the typical suit against a government contractor, which does not center on federal officers “enforcing

a locally unpopular national law,” Wyoming v. Livingston, 443 F.3d

1211, 1222 (10th Cir. 2006).

58a

ement of federal officer removal. The district court therefore correctly rejected the attempt to remove this action

under 28 U.S.C. § 1442(a)(1). Because ExxonMobil has

not established it sufficiently assisted a federal superior’s

duties through its participation in the OCS leasing program, we decline to reach the additional § 1442(a)(1) removal requirements of a causal nexus and a colorable federal immunity defense. See Riggs v. Airbus Helicopters,

Inc., 939 F.3d 981, 990 n.9 (9th Cir. 2019).

IV. CONCLUSION

Title 28, U.S. Code § 1447(d) empowers us to review

only the district court’s decision regarding removal under

28 U.S.C. § 1442(a)(1). ExxonMobil failed to establish

proper grounds for federal officer removal. We therefore

AFFIRM the district court’s remand order to the extent

it rejects removal under § 1442(a)(1) and DISMISS the

remainder of this appeal. The Counties’ motions for partial dismissal and for summary affirmance are granted

and dismissed as moot, respectively.

59a

APPENDIX B

IN THE UNITED STATES DISTRICT COURT

FOR THE DISTRICT OF COLORADO

Civil Action No. 18-cv-01672-WJM-SKC

BOARD OF COUNTY COMMISSIONERS OF BOULDER

COUNTY; BOARD OF COUNTY COMMISSIONERS OF SAN

MIGUEL COUNTY; AND CITY OF BOULDER,

PLAINTIFFS,

v.

SUNCOR ENERGY (U.S.A.) INC.; SUNCOR ENERGY SALES

INC.; SUNCOR ENERGY INC.; AND EXXON MOBIL

CORPORATION, DEFENDANTS.

Filed: September 5, 2019

ORDER

MARTINEZ, United States District Judge.

Plaintiffs brought Colorado common law and statutory

claims in Boulder County, Colorado District Court for injuries occurring to their property and citizens of their jurisdictions, allegedly resulting from the effects of climate

change. Plaintiffs sue Defendants in the Amended Complaint (“Complaint”) “for the substantial role they played

and continue to play in causing, contributing to and exacerbating climate change.” (ECF No. 7 ¶ 2.) Defendants

60a

filed a Notice of Removal (ECF No. 1) on June 29, 2018.

Plaintiffs filed a Motion to Remand (ECF No. 34) on July

30, 2018.

For the reasons explained below, the Court grants

Plaintiffs’ Motion to Remand. Defendants’ Motion to Reschedule Oral Argument on Plaintiffs’ Motion to Remand

(ECF No. 67), is denied as the Court finds that a hearing

is not necessary.

I. BACKGROUND

Plaintiffs assert six state law claims: public nuisance,

private nuisance, trespass, unjust enrichment, violation of

the Colorado Consumer Protection Act, and civil conspiracy. The Complaint alleges that Plaintiffs face substantial

and rising costs to protect people and property within

their jurisdictions from the dangers of climate alteration.

(ECF No. 7 ¶¶ 1–4, 11, 221–320.) Plaintiffs allege that Defendants substantially contributed to the harm through

selling fossil fuels and promoting their unchecked use

while concealing and misrepresenting their dangers. (Id.

¶¶ 2, 5, 13–18, 321–435.) The fossil fuel activities have

raised the emission and concentration of greenhouse

gases (“GHGs”) in the atmosphere. (Id. ¶¶ 7, 15, 123–138,

321–38.)

As a result of the climate alterations caused and contributed to by Defendants’ fossil fuel activities, Plaintiffs

allege that they are experiencing and will continue to experience rising average temperatures and harmful

changes in precipitation patterns and water availability,

with extreme weather events and increased floods,

drought, and wild fires. (ECF No. 7 ¶¶ 145–179.) These

changes pose a threat to health, property, infrastructure,

and agriculture. (Id. ¶¶ 1–4, 180–196.) Plaintiffs allege

that they are sustaining damage because of services they

61a

must provide and costs they must incur to mitigate or

abate those impacts. (Id. ¶¶ 1, 4–5, 221–320.) Plaintiffs

seek monetary damages from Defendants, requiring them

to pay their pro rata share of the costs of abating the impacts on climate change they have allegedly caused

through their tortious conduct. (Id. at ¶ 6.) Plaintiffs do

not ask the Court to stop or regulate Defendants’ emissions of fossil fuels (id. at ¶¶ 6, 542), and do not seek injunctive relief.

Defendants’ Notice of Removal asserts the following:

(1) federal question jurisdiction— that Plaintiffs’ claims

arise under federal common law, and that this action necessarily and unavoidably raises disputed and substantial

federal issues that give rise to jurisdiction under Grable

& Sons Metal Products, Inc. v. Darue Eng’g & Mfg., 545

U.S. 308 (2005) (“Grable”); (2) complete preemption; (3)

federal enclave jurisdiction; (4) jurisdiction because the

allegations arise from action taken at the direction of federal officers; (5) jurisdiction under the Outer Continental

Shelf Lands Act, 43 U.S.C. § 1349(b); and (6) jurisdiction

under 28 U.S.C. § 1452(a) because the claims are related

to bankruptcy proceedings.

While there are no dispositive cases from the Supreme

Court, the United States Court of Appeals for the Tenth

Circuit, or other United States Courts of Appeal, United

States District Court cases throughout the country are divided on whether federal courts have jurisdiction over

state law claims related to climate change, such as raised

in this case. Compare California v. BP p.l.c. (“CA I”), 2018

WL 1064293 (N.D. Cal. Feb. 27, 2018); City of Oakland v.

BP p.l.c. (“CA II), 325 F. Supp. 3d 1017 (N.D. Cal. June

25, 2018); City of New York v. BP p.l.c., 325 F. Supp. 3d

466 (S.D.N.Y. July 19, 2018) with State of Rhode Island v.

Chevron Corp., 2019 WL 3282007 (D. R.I. July 22, 2019);

62a

Mayor and City Council of Baltimore v. BP P.L.C. (“Baltimore”), 2019 WL 2436848 (D. Md. June 10, 2019), appeal

docketed, No. 19-1644 (4th Cir. June 18, 2019); and Cnty.

of San Mateo v. Chevron Corp., 294 F. Supp. 3d 934 (N.D.

Cal. 2018), appeal docketed, No. 18-15499 (9th Cir. May

27, 2018).

II. LEGAL STANDARD

Plaintiffs’ Motion to Remand is brought pursuant to

28 U.S.C. § 1447(c). The Motion to Remand asserts that

the Court lacks subject matter jurisdiction over the claims

in this case, which Plaintiffs contend are state law claims

governed by state law.

Federal courts are courts of limited jurisdiction, “possessing ‘only that power authorized by Congress and statute.’” Gunn v. Minton, 568 U.S. 251, 256 (2013) (citation

omitted). Thus, “[f]ederal subject matter jurisdiction is elemental.” Firstenberg v. City of Santa Fe, 696 F.3d 1018,

1022 (10th Cir. 2012). “It cannot be consented to or

waived, and its presence must be established” in every

case in federal court. Id.

Here, Defendants predicate removal on the ground

that the federal court has original jurisdiction over the

claims. 28 U.S.C. § 1441(a). Diversity jurisdiction has not

been invoked. Removal is appropriate “if, but only if, ‘federal subject-matter jurisdiction would exist over the

claim.”’ Firstenberg, 696 F.3d at 1023 (citation omitted).

If a court finds that it lacks subject matter jurisdiction at

any time before final judgment is entered, it must remand

the case to state court. 28 U.S.C. § 1447(c).

The burden of establishing subject matter jurisdiction

is on the party seeking removal to federal court, and there

is a presumption against its existence. Salzer v. SSM

Health Care of Okla. Inc., 762 F.3d 1130, 1134 (10th Cir.

63a

2014). “Removal statutes are to be strictly construed,. . .

and all doubts are to be resolved against removal.” Fajen

v. Found. Reserve Ins. Co., 683 F.2d 331, 333 (10th Cir.

1982). The party seeking removal must show that jurisdiction exists by a preponderance of the evidence. Dutcher v.

Matheson, 840 F.3d 1183, 1189 (10th Cir. 2016).

III. ANALYSIS

A. Federal Question Jurisdiction

Defendants first argue that federal question jurisdiction exists. Federal question jurisdiction exists for “all

civil actions arising under the Constitution, laws, or treaties of the United States.” 28 U.S.C. § 1331. In determining whether such jurisdiction exists, a court must “look to

the ‘face of the complaint’” and ask whether it is “‘drawn

so as to claim a right to recover under the Constitution

and laws of the United States’[.]” Firstenberg, 696 F.3d at

1023 (quoting Bell v. Hood, 327 U.S. 678, 681 (1946)).

“[T]he presence or absence of federal-question jurisdiction is governed by the ‘well-pleaded complaint rule’,

which provides that federal jurisdiction exists only when

a federal question is presented on the face of the plaintiff’s

properly pleaded complaint.” Caterpillar Inc. v. Williams, 482 U.S. 386, 392 (1987) (citation omitted). Under

this rule, a case arises under federal law ‘only when the

plaintiff’s statement of his own cause of action shows that

it is based’ on federal law.” Devon Energy Prod. Co., L.P.

v. Mosaic Potash Carlsbad, Inc., 693 F.3d 1195, 1202 (10th

Cir. 2012) (citation omitted). The court need only examine

“the well-pleaded allegations of the complaint and ignore

potential defenses. . . .’” Id. (citation omitted).

The well-pleaded complaint rule makes “the plaintiff

the master of the claim; he or she may avoid federal jurisdiction by exclusive reliance on state law.” Caterpillar,

64a

482 U.S. at 392; see also Devon Energy, 693 F.3d at 1202

(“By omitting federal claims from a complaint, a plaintiff

can generally guarantee an action will be heard in state

court.”) (internal quotation marks omitted). While the

plaintiff may not circumvent federal jurisdiction by artfully drafting the complaint to omit federal claims that are

essential to the claim, Caterpillar, 482 U.S. at 392, the

plaintiff “can elect the judicial forum–state of federal” depending on how the plaintiff drafts the complaint. Firstenberg, 696 F.3d at 1023. “Neither the plaintiff’s anticipation of a federal defense nor the defendant’s assertion of

a federal defense is sufficient to make the case arise under

federal law.” Id. (internal quotation marks omitted).

For a plaintiff’s well-pleaded complaint to establish

that the claims arise under federal law within the meaning

of § 1331, it “must establish one of two things: ‘either that

federal law creates the cause of action or that the plaintiff’s right to relief necessarily depends on a resolution of

a substantial question of federal law.’” Firstenberg, 696

F.3d at 1023 (citation omitted). The “creation’ test” in the

first prong accounts for the majority of suits that raise under federal law.” See Gunn, 568 U.S. at 257. However,

where a claim finds its origins in state law, the Supreme

Court has identified a “‘special and small category’ of

cases” in which jurisdiction lies under the substantial

question prong as they “implicate significant federal interests.” Id. at 258; see also Grable, 545 U.S. at 312.

Defendants argue that both prongs of federal question

jurisdiction are met. The Court will address each of these

arguments in turn.

65a

1. Whether Federal Law Creates the Cause of

Action

Defendants first assert that federal question jurisdiction exists because Plaintiffs’ claims arise under federal

law; namely, federal common law, such that federal law

creates the cause of action. The Supreme Court has “held

that a few areas, involving ‘uniquely federal interests,’ . . .

are so committed by the Constitution and laws of the

United States to federal control that state law is preempted and replaced, where necessary, by federal law of

a content prescribed (absent explicit statutory directive)

by the courts—so-called ‘federal common law.’” Boyle v.

United Technologies Corp., 487 U.S. 500, 504 (1988) (citations omitted); see also Nat’l Farmers Union Ins. Cos. v.

Crow Tribe of Indians, 471 U.S. 845, 850 (1985). The issue

must involve “an area of uniquely federal interest”, and

federal common law will displace state law only where “a

‘significant conflict’ exists between an identifiable ‘federal

policy or interest and the [operation] of state law,’ . . or

the application of state law would ‘frustrate specific objectives’ of federal legislation.” Boyle, 487 U.S. at 507 (citations omitted).

Defendants assert that this case belongs in federal

court because it threatens to interfere with longstanding

federal policies over matters of uniquely national importance, including energy policy, environmental protection, and foreign affairs. They note that two courts have

held that claims akin to those brought by Plaintiffs are

governed by federal common law, citing the decisions in

CA I, CA II, and City of New York.1

Notably, in another case ExxonMobil appeared to argue the opposite of what it argues here: that there is no uniquely federal interest

in this type of case and a suit does not require “‘the application of

1

66a

a. Relevant Case Law

Defendants state over the past century that the federal government has recognized that a stable energy supply is critical for the preservation of our economy and national security, taken steps to promote fossil fuel production, and worked to decrease reliance on foreign oil. The

government has also worked with other nations to craft a

workable international framework for responding to

global warming. This suit purportedly challenges those

decisions by requiring the court to delve into the thicket

of the “worldwide problem of global warming”— the solutions to which Defendants assert for “sound reasons”

should be “determined by our political branches, not by

our judiciary.” See CA II, 2018 WL 3109726, at *9.

Plaintiffs thus target global warming, and the transnational conduct that term entails. (ECF No. 7 ¶¶ 125–38.)

Defendants contend that the claims unavoidably require

adjudication of whether the benefits of fossil fuel use outweigh its costs—not just in Plaintiffs’ jurisdictions, or

even in Colorado, but on a global scale. They argue that

these claims do not arise out of state common law. Defendants further assert that this is why similar lawsuits have

been brought in federal court, under federal law, and why,

when those claims were dismissed, the plaintiffs made no

effort to pursue their claims in state courts. See, e.g., Am.

Elec. Power Co., Inc. v. Connecticut (“AEP”), 564 U.S. 410

(2011); Kivalina v. ExxonMobil Corp. (“Kivalina”), 696

F.3d 849 (9th Cir. 2012). Defendants thus contend that the

federal common law, merely because the conflict is not confined

within the boundaries of a single state.’” (See ECF No. 50-1 at 55–60)

(citation omitted). Instead, it asserted that “only suits by [states] implicating a sovereign interest in abating interstate pollution give rise

to federal common law.” (Id. at 58–60) (emphasis added).

67a

court has federal question jurisdiction because federal law

creates the cause of action.

The Court first addresses the cases relied on by Defendants that address similar claims involving injury from

global warming, beginning its analysis with the Supreme

Court’s decision in AEP. The AEP plaintiffs brought suit

in federal court against five domestic emitters of carbon

dioxide, alleging that by contributing to global warming,

they had violated the federal common law of interstate

nuisance, or, in the alternative, state tort law. 564 U.S. at

418 (citation omitted). They brought both federal and

state claims, and asked for “a decree setting carbon-dioxide emission for each defendant.” Id. The plaintiffs did not

seek damages.

The Court in AEP stated what while there is no federal general common law, there is an “emergence of a federal decisional law in areas of national concern”, the “new”

federal common law. 564 U.S. at 421 (internal quotation

marks omitted). This law “addresses ‘subjects within national legislative power where Congress has so directed’

or where the basic scheme of the Constitution so demands.” Id. (citation omitted). The Court found that environmental protection is “undoubtedly an area within national legislative power, one in which federal courts may

fill in statutory interstices, and, if necessary, even fashion

federal law.” Id. (internal quotation marks omitted). It

further stated that when the court “deal[s] with air and

water in their ambient or interstate aspects, there is federal common law.’” Id. (quoting Illinois v. City of Milwaukee, 406 US. 91, 103 (1972)).

AEP also found that when Congress addresses a question previously governed by federal common law, “‘the

need for such an unusual exercise of law-making by federal courts disappears.’” 564 U.S. at 423 (citation omitted).

68a

The test for whether congressional legislation excludes

the declaration of federal common law is “whether the

statute ‘speak[s] directly to [the] questions at issue.” Id.

at 424 (citation omitted). The Court concluded that “the

Clean Air Act and the EPA actions it authorizes displace

any federal common law right to seek abatement of carbon-dioxide emissions from fossil-fuel fired power plants,”

i.e., the Clean Air Act spoke directly “to emissions of carbon dioxide from the defendants’ plants.” Id. Since it

found that federal common law was displaced, AEP did

not decide the scope of federal common law, or whether

the plaintiffs had stated a claim under it. Id. at 423 (describing the question as “academic”). It also did not address the state law claims. Id. at 429.

In Kivalina, the plaintiffs alleged that massive greenhouse gas emissions by the defendants resulted in global

warming which, in turn, severely eroded the land where

the City of Kivalina sat and threatened it with imminent

destruction. 696 F.3d at 853. Relying on AEP, the Ninth

Circuit found that the Clean Air Act displaced federal

common law nuisance claims for damages caused by

global warming. Id. at 856. It recognized that “federal

common law includes the general subject of environmental law and specifically includes ambient or interstate air

and water pollution.” Id. at 855 (citing City of Milwaukee,

406 US. at 103). Thus, Kivalina stated that “federal common law can apply to transboundary pollution suits,” and

noted that most often such suits are, as in that case,

founded on a theory of public nuisance. Id. The Kivalina

court found that the case was governed by AEP and the

finding that Congress had “directly addressed the issue of

greenhouse gas commissions from stationary sources,”

thereby displacing federal common law. Id. at 856. The

fact that the plaintiffs sought damages rather than an

69a

abatement of emissions did not impact the analysis, according to Kivalina, because “the type of remedy asserted is not relevant to the applicability of the doctrine of

displacement.” Id. at 857. The Kivalina court affirmed the

district court’s dismissal of plaintiffs’ claims. Id. at 858.

Both AEP and Kivalina were brought in federal court

and asserted federal law claims. They did not address the

viability of state claims involving climate change that were

removed to federal court, as is the case here. This issue

was addressed by the United States District Court for the

Northern District of California in CA I and CA II. In the

CA cases, the Cities of Oakland and San Francisco asserted a state law public nuisance claim against ExxonMobil and a number of other worldwide producers of

fossil fuels, asserting that the combustion of fossil fuels

produced by the defendants had increased atmospheric

levels of carbon dioxide, causing a rise in sea levels with

resultant flooding in the cities. CA I, 2018 WL 1064293, at

*1. Like the instant case, the plaintiffs did not seek to impose liability for direct emissions of carbon dioxide.

Instead, they alleged “that—despite long-knowing

that their products posed severe risks to the global climate—defendants produced fossil fuels while simultaneously engaging in large scale advertising and public relations campaigns to discredit scientific research on global

warming, to downplay the risks of global warming, and to

portray fossil fuels as environmentally responsible and essential to human well-being.” Id. The plaintiffs sought an

abatement fund to pay for infrastructure necessary to address rising sea levels. Id.

CA I found that the plaintiffs’ state law “nuisance

claims—which address the national and international geophysical phenomenon of global warming—are necessarily governed by federal common law,” citing AEP, City

70a

of Milwaukee, and Kivalina. CA I, 2018 WL 1064293, at

*2–3. It stated that, as in those cases, “a uniform standard

of decision is necessary to deal with the issues,” explaining:

If ever a problem cried out for a uniform and comprehensive solution, it is the geophysical problem described by the complaints, a problem centuries in the

making (and studying) with causes [including] the

combustion of fossil fuels. The range of consequences

is likewise universal—warmer weather in some places

that may benefit agriculture but worse weather in others, . . . and—as here specifically alleged—the melting

of the ice caps, the rising of the oceans, and the inevitable flooding of coastal lands. . . . [T]he scope of the

worldwide predicament demands the most comprehensive view available, which in our American court

system means our federal courts and our federal common law. A patchwork of fifty different answers to the

same fundamental global issue would be unworkable.

Id. at *3.

The CA I court also found that federal common law

applied despite the fact that “plaintiffs assert a novel theory of liability,” i.e., against the sellers of a product rather

than direct dischargers of interstate pollutants. CA I,

2018 WL 1064293, at *3 (emphasis in original). Again, that

is the situation in this case. The CA I court stated that “the

transboundary problem of global warming raises exactly

the sort of federal interests that necessitate a uniform solution,” which is no “ less true because plaintiffs’ theory

mirrors the sort of state-law claims that are traditionally

applied to products made in other states and sold nationally.” Id. The court found, however, that federal common

law was not displaced by the Clean Air Act and the EPA

as in AEP and Kivalina because the plaintiffs there

71a

sought only to reach domestic conduct, whereas the plaintiffs’ claims in CA I “attack behavior worldwide.” Id. at 4.

It stated that those “foreign emissions are outside of the

EPA and Clean Air Acts’ reach.” Id. Nonetheless, as the

claims were based in federal law, the court found that federal jurisdiction existed and denied the plaintiffs’ motions

to remand. Id. at 5.

In CA II, the court granted the defendants’ motion to

dismiss. 325 F. Supp. 3d at 1019. It reaffirmed that the

plaintiffs’ nuisance claims “must stand or fall under federal common law,” including the state law claims. CA II,

325 F. Supp. 3d at 1024. It then held that the claims must

be dismissed because they ran counter to the presumption

against extraterritoriality and were “foreclosed by the

need for federal courts to defer to the legislative and executive branches when it comes to such international

problems.” Id. at 1024–25. The CA II court concluded that

“[i]t may seem peculiar that an earlier order refused to

remand this action to state court on the ground that plaintiffs’ claims were necessarily governed by federal law,

while the current order concludes that federal common

law should not be extended to provide relief.” Id. at 1028.

But it found “no inconsistency,” as “[i]t remains proper for

the scope of plaintiffs’ claims to be decided under federal

law, given the international reach” of the claims. Id. at

1028–29.

The City of New York case followed the rationale of

CA I and CA II, and dismissed New York City’s claims of

public and private nuisance and trespass against multinational oil and gas companies related to the sale and production of fossil fuels. 325 F. Supp. 3d at 471–76. On a motion to dismiss, the court found that the City’s claims were

governed by federal common law, not state tort law, because they were “based on the ‘transboundary’ emission

72a

of greenhouse gases” which “require a uniform standard

of decision.” Id. at 472 (citing CA I, 2018 WL 10649293, at

*3). It also found that to the extent the claims involved domestic greenhouse emissions, the Clean Air Act displaced

the federal common law claims pursuant to AEP. Id. To

the extent the claims implicated foreign greenhouse emissions, they were “barred by the presumption against extraterritoriality and the need for judicial caution in the

face of ‘serious foreign policy consequences.’” Id. at 475

(citation omitted). The court in City of New York did not

address federal jurisdiction or removal jurisdiction.

In summary, the above cases suggest that claims related to the emission or sale, production, or manufacture

of fossil fuels are governed by federal common law, even

if they are asserted under state law, but may displaced by

the Clean Air Act and the EPA. At first blush these cases

appear to support Defendants’ assertion that Plaintiffs’

claims arise under federal law and should be adjudicated

in federal court, particularly given the international scope

of global warming that is at issue.

However, the Court finds that AEP and Kivalina are

not dispositive. Moreover, while the CA I decision has a

certain logic, the Court ultimately finds that it is not persuasive. Instead, the Court finds that federal jurisdiction

does not exist under the creation prong of federal question

jurisdiction, consistent with San Mateo and the two most

recent cases that have addressed the applicable issues, as

explained below.

The Court first notes that in AEP and Kivalina, the

plaintiffs expressly invoked federal claims, and removal

was neither implicated nor discussed. Moreover, both

cases addressed interstate emissions, which are not at issue here. Finally, the cases did not address whether the

state law claims were governed by federal common law.

73a

The AEP Court explained that “the availability vel non of

a state lawsuit depend[ed], inter alia, on the preemptive

effect of the federal Act,” and left the matter open for consideration on remand. 564 U.S. at 429. Thus, “[f]ar from

holding (as the defendants bravely assert) that state

claims related to global warming are superseded by federal common law, the Supreme Court [in AIG] noted that

the question of whether such state law claims survived

would depend on whether they are preempted by the federal statute that had displaced federal common law (a

question the Court did not resolve).” San Mateo, 294 F.

Supp. 3d at 937.

Moreover, while AEP found that federal common law

governs suits brought by a state to enjoin emitters of pollution in another state, it noted that the Court had never

decided whether federal common law governs similar

claims to abate out-of-state pollution brought by “political

subdivisions” of a State, such as in this case. 564 U.S. at

421–22. Thus, AEP does not address whether state law

claims, such as those asserted in this case and brought by

political subdivisions of a state, arise under federal law for

purposes of removal jurisdiction. The Ninth Circuit in Kivalina also did not address this issue.

The Court disagrees with the finding in CA I that removal jurisdiction is proper because the case arises under

federal common law. CA I found that the well-pleaded

complaint rule did not apply and that federal jurisdiction

exists “if the claims necessarily arise under federal common law. 2018 WL 1064293, at *5. It based this finding on

a citation to a single Ninth Circuit case, Wayne v. DHL

Worldwide Express, 294 F.3d 1179, 1184–85 (9th Cir.

2002). Id. Wayne, however, recognized the well-pleaded

complaint rule, and did not address whether a claim that

arises under federal common law is an exception to the

74a

rule. 294 F.3d at 1183-85. Moreover, Wayne cited City of

Milwaukee in support of its finding that federal jurisdiction would exist if the claims arose under federal law. City

of Milwaukee was, however, filed in federal court and invoked federal jurisdiction such that the well-pleaded complaint rule was not at issue.

Thus, CA I failed to discuss or note the significance of

the difference between removal jurisdiction, which implicates the well pleaded complaint rule, and federal jurisdiction that is invoked at the outset such as in AEP and

Kivalina. This distinction was recognized by the recent

decision in Baltimore, which involved similar state law

claims as to climate change that were removed to federal

court. 2019 WL 2436848, at *1. Baltimore found CA I was

“well stated and presents an appealing logic,” but disagreed with it because the court looked beyond the face of

the plaintiffs’ well pleaded complaint. Id. at *7–8. It also

noted that CA I “did not find that the plaintiffs’ state law

claims fell within either of the carefully delineated exceptions to the well-pleaded complaint rule—i.e., that they

were completely preempted by federal law or necessarily

raised substantial, disputed issues of federal law.” Id. at

*8. Baltimore found that the well-pleaded complaint rule

was plainly not satisfied in that case because the City did

not plead any claims under federal law. Id. at *6.

b. The Well-Pleaded Complaint Rule as Applied to Plaintiffs’ Claims

In a case that is removed to federal court, the presence

or absence of federal-question jurisdiction is governed by

the well-pleaded complaint rule, which gives rise to federal jurisdiction only when a federal question is presented

on the face of the complaint. Caterpillar, 482 U.S. at 392.

The Tenth Circuit has held that to support removal jurisdiction, “the required federal right or immunity must be

75a

an essential element of the plaintiff’s cause of action, and

. . . the federal controversy must be disclosed upon the

face of the complaint, unaided by the answer or by the petition for removal.” Fajen, 683 F.2d at 333 (citation and

internal quotation marks omitted).

In this case, the Complaint on its face pleads only state

law claims and issues, and no federal law or issue is raised

in the allegations. While Defendants argue that the Complaint raises inherently federal questions about energy,

the environment, and national security, removal is not appropriate under the well-pleaded complaint rule because

these federal issues are not raised or at issue in Plaintiffs’

claims. A defendant cannot transform the action into one

arising under federal law, thereby selecting the forum in

which the claim will be litigated, as to do so would contradict the well-pleaded complaint rule. Caterpillar, 489 U.S.

at 399. Defendants, “in essence, want the Court to peek

beneath the purported state-law facade of the State’s public nuisance claim, see the claim for what it would need to

be to have a chance at viability, and convert it to that (i.e.,

into a claim based on federal common law) for purposes of

the present jurisdiction analysis.” State of Rhode Island,

2019 WL 3282007, at *2. That court found nothing in the

artful-pleading doctrine which sanctioned the defendants’

desired outcome. Id.

Defendants cite no controlling authority for the proposition that removal may be based on the existence of an

unplead federal common law claim—much less based on

one that is questionable and not settled under controlling

law. Defendants rely on the Supreme Court’s holding that

the statutory grant of jurisdiction over cases arising under the laws of the United States “will support claims

founded upon federal common law.” Nat’l Farmers Union Ins. Cos., 471 U.S. at 850–53. However, the plaintiffs

76a

invoked federal jurisdiction in that case. The same is true

in other cases cited by Defendants, including City of Milwaukee and Boyle, both of which were filed by plaintiffs

in federal court and invoked federal jurisdiction. See, e.g.,

State of Rhode Island, 2019 WL 3282007, at *2 n. 2 (Boyle

“does not help Defendants” as it “was not a removal case,

but rather one brought in diversity”); Arnold by and

Through Arnold v. Blue Cross & Blue Shield, 973 F.

Supp. 726, 737 (S.D. Tex. 1997) (Boyle did not address removal jurisdiction, nor did it modify the Caterpillar rule

that federal preemption of state law, even when asserted

as an inevitable defense to a . . . state law claim, does not

provide a basis for removal”), overruled on other grounds,

Winters v. Diamond Shamrock Chem. Co., 149 F.3d 387

(5th Cir. 1997). Removal based on federal common law being implicated by state claims was not discussed or sanctioned in Defendants’ cases.

A thoughtful analysis of the limits that removal jurisdiction poses on federal question jurisdiction was conducted in E. States Health & Welfare Fund v. Philip Morris, Inc., 11 F. Supp. 2d 384 (S.D.N.Y. 1998). That court

noted that removal jurisdiction is “a somewhat different

animal than original federal question jurisdiction—i.e.,

where the plaintiff files originally in federal court.” Id. at

389. It explained:

When a plaintiff files in federal court, there is no clash

between the principle that the plaintiff can control the

complaint—and therefore, the choice between state

and federal forums—and the principle that federal

courts have jurisdiction over federal claims; the plaintiff, after all, by filing in a federal forum is asserting

reliance upon both principles, and the only question a

defendant can raise is whether plaintiff has a federal

claim.

77a

On the other hand, when a plaintiff files in state court

and purports to only raise state law claims, for the federal court to assert jurisdiction it has to look beyond

the complaint and partially recharacterize the plaintiffs’ claims—which places the assertion of jurisdiction

directly at odds with the principle of plaintiff as the

master of the complaint. It is for this reason that removal jurisdiction must be viewed with a somewhat

more skeptical eye; the fact that a plaintiff in one case

chooses to bring a claim as a federal one and thus invoke federal jurisdiction does not mean that federal

removal jurisdiction will lie in an identical case if the

plaintiff chooses not to file a federal claim.

Id. at 389–90. The Court agrees with this well-reasoned

analysis.

The cases cited by Defendants from other jurisdictions that found removal of state law claims to federal

court was appropriate because the claims arose under or

were necessarily governed by federal common law are not

persuasive. See Wayne, 294 F.3d at 1184–85; Sam L. Majors Jewelers v. ABX, Inc., 117 F.3d 922, 926 (5th Cir.

1997); CA I, 2018 WL 1064293, at *2; Blanco v. Fed. Express Corp., No. 16-561, 2016 WL 4921437, at *2–3 (W.D.

Okla. Sept. 15, 2016). Those cases contradict Caterpillar

and the tenets of the well-pleaded complaint rule. They

also fail to cite any Supreme Court or other controlling

authority authorizing removal based on state law claims

implicating federal common law. While many of those

cases relied on City of Milwaukee as authority for their

holdings, the plaintiff in that case invoked federal common

law and federal jurisdiction. City of Milwaukee does not

support a finding that a defendant can create federal jurisdiction by re-characterizing a state claim.

78a

c. Ordinary Preemption

Ultimately, Defendants’ argument that Plaintiffs’

state law claims are governed by federal common law appears to be a matter of ordinary preemption which—in

contrast to complete preemption, which is discussed in

Section III.B, infra,–would not provide a basis for federal

jurisdiction. See Geddes v. Am. Airlines, Inc., 321 F.3d

1349, 1352 (11th Cir. 2003) (cited with approval in Devon

Energy, 693 F.3d at 1203).2 “Ordinary preemption ‘regulates the interplay between federal and state laws when

they conflict or appear to conflict . . . .’” Baltimore, 2019

WL 2436848, at *6 (citation omitted). The distinction between ordinary and complete preemption “is important

because if complete preemption does not apply, but the

plaintiff’s state law claim is arguably preempted . . . the

district court, being without removal jurisdiction, cannot

resolve the dispute regarding preemption.” Colbert v. Union Pac. R. Co., 485 F. Supp. 2d 1236, 1243 (D. Kan. 2007)

(internal quotation marks omitted).

When ordinary preemption applies, the federal court

“‘lacks the power to do anything other than remand to the

state court where the preemption issue can be addressed

and resolved.’” Colbert, 485 S. Supp. 2d at 1243 (citation

omitted). Ordinary preemption is thus a defense to the

complaint, and does not render a state- law claim removable to federal court. Hansen v. Harper Excavating, Inc.,

641 F.3d 1216, 1221 (10th Cir. 2011); see also Caterpillar,

482 U.S. at 392–93 (under the well-pleaded complaint rule,

2

The three forms of preemption that are frequently discussed in

judicial opinions— express preemption, conflict preemption, and field

preemption—are characterized as ordinary preemption. Devon Energy, 693 F.3d at 1203 n. 4.

79a

courts must ignore potential defenses such as preemption).

Thus, the fact that a defendant asserts that federal

common law is applicable “does not mean the plaintiffs’

state law claims ‘arise under’ federal law for purposes of

jurisdictional purposes.” E. States Health, 11 F. Supp. 2d

at 394. As that court explained, “[c]ouch it as they will in

‘arising under’ language, the defendants fail to explain

why their assertion that federal common law governs . . .

is not simply a preemption defense which, while it may

very well be a winning argument on a motion to dismiss in

the state court, will not support removal jurisdiction.” Id.

This finding is consistent with the decision in Baltimore. The court there found the defendants’ assertion

that federal question jurisdiction existed because the

City’s nuisance claim “is in fact ‘governed by federal common law’” was “‘a cleverly veiled [ordinary] preemption

argument.” Baltimore, 2019 WL 2436848, at *6 (citing

Boyle, 487 U.S. at 504). As the Baltimore defendants’ argument amounted to an ordinary preemption defense, it

did “not allow the Court to treat the City’s public nuisance

claim as if it had been pleaded under federal law for jurisdictional purposes.” Id. The court also found that the CA

I ruling was “at odds with the firmly established principle

that ordinary preemption does not give rise to federal

question jurisdiction.” Id. at *8.

Because an ordinary preemption defense does not

support remand, Defendants’ federal common law argument could only prevail under the doctrine of complete

preemption. Unlike ordinary preemption, complete

preemption “is so ‘extraordinary’ that it ‘converts an ordinary state law common-law complaint into one stating a

federal claim for purposes of the well-pleaded complaint

rule.’” Caterpillar, 482 U.S. at 393 (citation omitted).

80a

2. Whether Plaintiffs’ Right to Relief Necessarily Depends on Resolution of a Substantial Question of Federal Law (Grable Jurisdiction)

Defendants also argue that federal jurisdiction exists

under the second prong of the “arising under” jurisdiction, as Plaintiffs’ claims necessarily depend on a resolution of a substantial question of federal law under Grable.

They contend that the Complaint raises federal issues under Grable “because it seeks to have a court determine for

the entire United States, as well as Canada and other foreign actors, the appropriate balance between the production, sale, and use of fossil fuels and addressing the risks

of climate change.” (ECF No. 1 ¶ 37.) Such an inquiry, according to Defendants, “necessarily entails the resolution

of substantial federal questions concerning important federal regulations, contracting, and diplomacy.” (Id.) Thus,

they assert that the “state-law claim[s] necessarily raise a

stated federal issue, actually disputed and substantial,

which a federal forum may entertain without disturbing

. . . federal and state judicial responsibilities.” Grable, 545

U.S. at 313–14.

The substantial question doctrine “captures the commonsense notion that a federal court ought to be able to

hear claims recognized under state law that nonetheless

turn on substantial questions of federal law, and thus justify resort to the experience, solicitude, and hope of uniformity that a federal forum offers on federal issues.”

Grable, 545 U.S. at 312. To invoke this branch of federal

question jurisdiction, the Defendants must show that “a

federal issue is: (1) necessarily raised, (2) actually disputed, (3) substantial, and (4) capable of resolution in federal court without disrupting the federal-state balance approved by Congress.” Gunn, 568 U.S. at 258.

81a

Jurisdiction under the substantial question doctrine

“is exceedingly narrow—a special and small category of

cases.” Firstenberg, 696 F.3d at 1023 (citation and internal quotation marks omitted). “[M]ere need to apply federal law in a state-law claim will not suffice to open the

‘arising under’ door” of jurisdiction. Grable, 545 U.S. at

313. Instead, “‘federal jurisdiction demands not only on a

contested federal issue, but a substantial one, indicating a

serious federal interest in claiming the advantages

thought to be inherent in a federal forum.’” Id. (citation

omitted).

a. Necessarily Raised

The Court finds that the first prong of substantial

question jurisdiction is not met because Plaintiffs’ claims

do not necessarily raise or depend on issues of federal law.

The discussion of this issue in Baltimore is instructive. In

that case, the defendants contended that Grable jurisdiction existed because the claims raised a host of federal issues. Baltimore, 2019 WL 2436848, at *9. For example,

the defendants asserted that the claims “‘intrude upon

both foreign policy and carefully balanced regulatory considerations at the national level, including the foreign affairs doctrine.’” Id. (citation omitted). They also asserted

that the claims “‘have a significant impact on foreign affairs,’ ‘require federal-law-based cost-benefit analyses,’”

and “‘amount to a collateral attack on federal regulatory

oversight of energy and the environment.’” Id. (citation

omitted). These allegations are almost identical to what

Defendants assert in this case. (See ECF No. 48 at 22—

“Plaintiffs’ claims gravely impact foreign affairs”; 24—

“Plaintiffs’ claims require reassessment of cost-benefit

analyses committed to, and already conducted by the Gov-

82a

ernment”; 26—the claims “are a collateral attack on federal regulatory oversight of energy and the environment”).

Baltimore found that these issues were not “‘necessarily raised’ by the City’s claims, as required for Grable

jurisdiction.” 2019 WL 2436848, at *9–10. As to the alleged

significant effect on foreign affairs, the court agreed that

“[c]limate change is certainly a matter of serious national

and international concern.” Id. at *10. But it found that

defendants did “not actually identify any foreign policy

that was implicated by the City's claims, much less one

that is necessarily raised.” Id. “They merely point out that

climate change ‘has been the subject of international negotiations for decades.’” Id. Baltimore found that “defendants’ generalized references to foreign policy wholly

fail to demonstrate that a federal question is ‘essential to

resolving’ the City’s state law claims.” Id. (citation omitted).

The Court finds the analysis in Baltimore equally persuasive as to Defendants’ reliance on foreign affairs in this

case, as they point to no specific foreign policy that is essential to resolving the Plaintiffs’ claims. Instead, they

cite only generally to non-binding, international agreements that do not apply to private parties, and do not explain how this case could supplant the structure of such

foreign policy arrangements. Certainly Defendants have

not shown that any interpretation of foreign policy is an

essential element of Plaintiffs’ claims. Gilmore v. Weatherford, 694 F.3d 1160, 1173 (10th Cir. 2012).

The CA I and City of New York decisions do not support Defendants’ argument that the foreign policy issues

raise substantial questions of law. Defendants note, for

example, that the City of New York court dismissed the

claims there on the merits “for severely infring[ing] upon

83a

the foreign-policy decisions that are squarely within the

purview of the political branches of the U.S. Government.” 325 F. Supp. 3d at 476. But as Defendants have

acknowledged, at least at this stage of these proceedings,

the Court is not considering the merits of Plaintiffs’ claims

or whether they would survive a motion to dismiss, only

whether there is a basis for federal jurisdiction. (See ECF

No. 1 ¶ 20.) While CA I and City of New York may ultimately be relevant to whether Plaintiffs’ claims should be

dismissed, they do not provide a basis for Grable jurisdiction. See Becker v. Ute Indian Tribe of the Uintah and

Ouray Reservation, 770 F.3d 944, 948 (10th Cir. 2014)

(federal law that is alleged as a barrier to the success of a

state law claim “is not a sufficient basis from which to conclude that the questions are ‘necessarily raised’”) (citation

omitted).

Baltimore also rejected cost-benefit analysis and collateral attack arguments as a basis for Grable jurisdiction,

finding that they “miss[ ] the mark.” 2019 WL 2436848, at

*10. This is because the nuisance claims were, as here,

based on the “extraction, production, promotion, and sale

of fossil fuel products without warning consumers and the

public of their known risks”, and did “not rely on any federal statutes or regulations” or violations thereof. Id. “Although federal laws and regulations governing energy production and air pollution may supply potential defenses,”

the court found that federal law was “plainly not an element” of the City’s state law nuisance claims. Id.

The same analysis surely applies here. Plaintiffs’ state

law claims do not have as an element any aspect of federal

law or regulations. Plaintiffs do not allege that any federal

regulation or decision is unlawful, or a factor in their

claims, nor are they asking the Court to consider whether

84a

the government’s decisions to permit fossil fuel use and

sale are appropriate.

As to jurisdiction under Grable, the Baltimore court

concluded that, “[t]o be sure, there are federal interests in

addressing climate change.” 2019 WL 2436848, at *11

(emphasis in original). “Defendants have failed to establish, however, that a federal issue is a ‘necessary element’

of the City’s state law claims.” Id. (citation omitted) (emphasis in original). Thus, even without considering the remaining requirements for Grable jurisdiction, the Baltimore court rejected the defendants’ assertion that the

case fell within “the ‘special and small category’ of cases

in which federal question jurisdiction exists over a state

law claim. Id. (citation omitted).

Two other courts have recently arrived at the same

conclusion. The court in State of Rhode Island found that

the defendants had not shown that federal law was “‘an

element and an essential one, of the [State]’s cause[s] of

action.’” 2019 WL 3282007, at *4 (citation omitted). Instead, the court noted that the State’s claims “are thoroughly state-law claims”, and “[t]he rights, duties, and

rules of decision implicated by the complaint are all supplied by state law, without reference to anything federal.”

Id. The court concluded:

By mentioning foreign affairs, federal regulations, and

the navigable waters of the United States, Defendants

seek to raise issues that they may press in the course

of this litigation, but that are not perforce presented

by the State's claims. . . .These are, if anything, premature defenses, which even if ultimately decisive, cannot support removal.

Id. (internal citations omitted).

85a

Similarly, the court in San Mateo found that the defendants had not pointed to a specific issue of federal law

that necessarily had to be resolved to adjudicate the state

law claims. 294 F. Supp. 3d at 938. Instead, “the def

endants mostly gesture to federal law and federal concerns in a generalized way.” Id. The court found that

“[t]he mere potential for foreign policy implications”, the

“mere existence of a federal regulatory regime”, or the

possibility that the claims involved a weighing of costs and

benefits did not raise the kind of actually disputed, substantial federal issue necessary for Grable jurisdiction. Id.

San Mateo concluded, “[o]n the defendants’ theory, many

(if not all) state tort claims that involve the balancing of

interests and are brought against federally regulated entities would be removable”, and “Grable does not sweep

so broadly.” Id.

The Court agrees with the well-reasoned analyses in

Baltimore, State of Rhode Island, and San Mateo, and

adopts the reasoning of those decisions. To the extent Defendants raise other issues not addressed in those cases,

the Court finds that they also are not necessarily raised in

Plaintiffs’ Complaint.

Defendants here assert that Plaintiffs’ claims raise a

significant issue under Grable because they attack the decision of the federal government to enter into contracts

with Defendant ExxonMobil to develop and sell fossil

fuels. (ECF No. 1 ¶ 43.) Further, they argue that the

Complaint seeks to deprive the federal government of a

mechanism for carrying out vital governmental functions,

and frustrates federal objectives. (Id. ¶ 44.)

Plaintiffs’ claims, however, assert no rights under the

contracts referenced by Defendants. Nor do they challenge the contracts’ validity, or require a court to interpret their meaning or importance. The Complaint does

86a

not even mention the contracts. Defendants’ argument

appears to be based solely on their unsupported speculation about the potential impact that Plaintiffs’ success

would have on the government’s ability to continue purchasing fossil fuels. (Id. ¶¶ 43–44.) Even if Defendants’

speculation was well-founded, this would be relevant only

to the substantiality prong of the Grable analysis. See

Bennett v. Sw. Airlines Co., 484 F.3d 907, 910 (10th Cir.

2007). Defendants have not established the first requirement—that the issue is necessarily raised by the Plaintiffs.

b. Substantiality

The Court also finds that the second prong, substantiality, is not met. To determine substantiality, courts

“look[] to whether the federal law issue is central to the

case.” Gilmore, 694 F.3d at 1175. Courts distinguish “between ‘a nearly pure issue of law’ that would govern ‘numerous’ cases and issues that are ‘fact-bound and situation-specific.’” Id. at 1174 (quoting Empire Healthchoice

Assurance, Inc. v. McVeigh, 547 U.S. 677, 700–11 (2006)).

When a case “‘involve[s] substantial questions of state as

well as federal law,’ this factor weighs against asserting

federal jurisdiction.” Id. at 1175 (citation omitted).

The Court finds that the issues raised by Defendants

are not central to Plaintiffs’ claims, and the claims are

“rife with legal and factual issues that are not related” to

the federal issues. See Stark-Romero v. Nat’l R.R. Passenger Co. (Amtrak), No. CIV-09- 295, 2010 WL

11602777, at *8 (D.N.M. Mar. 31, 2010). This case is quite

different from those where jurisdiction was found under

the substantial question prong of jurisdiction. For example, in Grable, “the meaning of the federal statute . . . appear[ed] to be the only legal or factual issue contested in

the case.” 545 U.S. at 315. Similarly, in a Tenth Circuit

87a

case finding jurisdiction under Grable, “construction of

the federal land grant” at issue “appear[ed] to be the only

legal or factual issue contested in the case.” Nicodemus v.

Union Pac. Corp., 440 F.3d 1227, 1236 (10th Cir. 2006).

Here, it is plainly apparent that the federal issues raised

by Defendants are not the only legal or factual issue contested in the case. Plaintiffs’ claims also do not involve a

discrete legal question, and are “fact-bound and situationspecific,” unlike Grable. See Empire Healthchoice Assurance, 547 U.S. at 701; Bennett, 484 F.3d at 910–11. Finally,

the case does not involve a state-law cause of action that

“is ‘brought to enforce’ a duty created by [a federal statute],” where “the claim’s very success depends on giving

effect to a federal requirement.” Merrill Lynch, Pierce,

Fenner & Smith, Inc. v. Manning, ___U.S. ___, 136 S. Ct.

1562, 1570 (2016).

The cases relied upon by Defendants are distinguishable, as Plaintiffs have shown in their briefing. For example, while Defendants cite Crosby v. National Foreign

Trade Council, 530 U.S. 363 (2000), that case involved

preemption under the Supremacy Clause because of a

conflict between a state law and Congress’s imposition of

sanctions. It did not address Grable jurisdiction, and thus

does not support Defendants’ assertion that it is “irrelevant” to the jurisdictional issue that the “foreign agreements are not ‘essential elements of any claim.’” (ECF

No. 48 at 23.)

Base

This text is long and has been trimmed here. Open the source document for the complete record.