Cheryl Harris v. Kellogg Brown & Root Services

PRECEDENTIAL

UNITED STATES COURT OF APPEALS
FOR THE THIRD CIRCUIT
_____________

No. 12-3204
_____________

CHERYL HARRIS, Co-Administratrix of the Estate
of Tyan D. Maseth, deceased; DOUGLAS MASETH;
Co-Administrator of the Estate of Ryan D. Maseth,
deceased,
Appellants

v.

KELLOGG BROWN & ROOT SERVICES, INC.

On Appeal from the United States District Court
for the Western District of Pennsylvania
District Court No. 2-08-cv-00563
District Judge: The Honorable Nora B. Fischer

Argued May 14, 2013

Before: SMITH, FISHER, and CHAGARES, Circuit
Judges

(Filed: August 1, 2013)
Patrick K. Cavanaugh
Stephen J. Del Sole
William S. Stickman, IV [ARGUED]
Del Sole Cavanaugh Strotd
200 First Avenue
Suite 300
Pittsburgh, PA 15222
Counsel for Appellant

Raymond B. Biagini
Michelle L. Hylton
Lawrence S. Ebner [ARGUED]
Kurt J. Hamrock
Shannon G. Konn
Daniel L. Russell, Jr.
William J. Sayers
McKenna, Long & Aldridge
1900 K Street, N.W.
Washington, DC 20006

John R. Dingess
Kari Horner
Dingess, Foster, Luciana, Davidson
& Chleboski
20 Stanwix Street
PNC Center, Third Floor
Pittsburgh, PA 15222

2
William D. Wickard
K&L Gates
210 Sixth Avenue
Pittsburgh, PA 15222
Counsel for Appellee

________________

OPINION
________________

SMITH, Circuit Judge.

This case returns to us for resolution of the
“important questions about the scope of the political
question doctrine and the Federal Tort Claims Act’s
‘combatant activities’ exception” in suits against defense
contractors. We did not have the opportunity to reach
these issues when this case was before us previously.
Harris v. Kellogg, Brown & Root Servs., Inc., 618 F.3d
398, 399 (3d Cir. 2010). Having jurisdiction now to reach
these questions, we will provide a framework that
establishes the contours of each of these doctrines. And
while explaining the two frameworks can be simple,
applying them is complicated by a number of case-by-
case factors. Illustrating this is our conclusion that one
such crucial factor still needs to be decided before the
political-question doctrine aspect of this case can be
3
resolved: which state’s law controls the claims and
defenses presented. This, in addition to our conclusion
that the combatant-activities exception does not preempt
the plaintiffs’ claims, requires that we reverse and
remand to the District Court for further proceedings.
I

During the Iraq War, the United States military
established the Radwaniyah Palace Complex as a base of
operations. Staff Sergeant Ryan Maseth was stationed
there and assigned to live in the barracks called Legion
Security Forces Building 1, a building that predated the
war and was known to have significant electrical
problems. On January 2, 2008, Staff Sergeant Maseth
died by electrocution while taking a shower in his
barracks. The shower was electrified by an ungrounded
and unbonded water pump.

Staff Sergeant Maseth’s estate and his parents sued
Kellogg, Brown & Root Services (“KBR”), a military
contractor hired to perform certain maintenance services
at the barracks. They allege that KBR caused Staff
Sergeant Maseth’s death by negligently performing its
maintenance duties under two contracts with the United
States. According to the plaintiffs, these contracts set
standards of care for work performed under them, which
KBR did not meet because it failed to ground and bond
the water pump either when KBR installed it or

4
responded to work orders complaining of electrified
water in Staff Sergeant Maseth’s barracks.

The merits of the plaintiffs’ claims have not yet
been resolved. Instead, KBR raises two antecedent
challenges through a Rule 12(b)(1) motion to dismiss.
First, KBR argues that the District Court should not
exercise its proper 28 U.S.C. § 1332 diversity jurisdiction
because this case presents a nonjusticiable political
question. Second, KBR argues that the plaintiffs’ claims
are preempted by the federal policy underlying the
combatant-activities exception in 28 U.S.C. § 2860(j) to
the United States’ waiver of sovereign immunity for
torts.
The District Court first denied the motion before
extensive discovery took place. Harris v. Kellogg, Brown
& Root Servs., Inc., 618 F. Supp. 2d 400, 403 (W.D. Pa.
2009). KBR sought review of this denial through an
interlocutory appeal under the collateral-order doctrine,
which we dismissed for lack of appellate jurisdiction.
Harris, 618 F.3d at 400, 404. On remand, the District
Court ordered discovery on the plaintiffs’ claims and
KBR’s defenses. After discovery was mostly complete,
KBR renewed its Rule 12(b)(1) motion to dismiss. This
time, the District Court granted the motion, holding that
the case was nonjusticiable and—alternatively—that the
plaintiffs’ claims were preempted by the federal policy
embodied in § 2680(j). Harris v. Kellogg, Brown & Root
Servs., Inc., 878 F. Supp. 2d 543, 547–58 (W.D. Pa.
5
2012). The plaintiffs appeal the dismissal, and this Court
has jurisdiction under 28 U.S.C. § 1291.

II

Jurisdictional and justiciability questions must be
resolved before a court reaches the merits of a case.
Larsen v. Senate of Com. of Pa., 152 F.3d 240, 245–46
(3d Cir. 1998); see also Sinochem Int’l Co. v. Malaysia
Int’l Shipping Corp., 549 U.S. 422, 431 (2007)
(“Jurisdiction is vital only if the court proposes to issue a
judgment on the merits.”). Whether a case contains a
political question is a matter of justiciability. Gross v.
German Found. Indus. Initiative, 456 F.3d 363, 376 (3d
Cir. 2006). Absent complete preemption, whether a
plaintiff’s claims are preempted relates to the merits. See
In re U.S. Healthcare, Inc., 193 F.3d 151, 160 (3d Cir.
1999) (explaining that “ordinary preemption” arises only
“as a federal defense to a state-law claim”). Neither party
argues, and no court has held, that § 2860(j) combatant-
activities preemption constitutes complete preemption.
Accordingly, we must review the District Court’s
political-question holding before addressing its
preemption holding.1

1
The parties do not discuss whether Rule 12(b)(1) was
the appropriate vehicle for KBR to assert its § 2680(j)
preemption argument. The District Court, however, noted
that Rule 56 may have been the appropriate vehicle.
6
Nevertheless, it analyzed KBR’s arguments under Rule
12(b)(1) because the plaintiffs did not dispute its
applicability and because the District Court believed that
“it can be reasonably inferred from [our prior decision in
this case] that Rule 12(b)(1) is the appropriate standard.”
Harris, 878 F. Supp. 2d at 568 & n.17. Although the
parties do not dispute the appropriate standard, we must
clarify that our prior decision did not imply, as the
District Court believed, that Rule 12(b)(1) is the right
vehicle for ordinary preemption arguments.
As the District Court acknowledged, our first decision in
this case did not address whether Rule 12(b)(1) was the
appropriate vehicle in which to advance any of the
arguments KBR has made. Id. at 568 n.17. Nevertheless,
the District Court inferred from our statement that
“because the presence or absence of a political question
is such a fact-intensive inquiry, a better-developed record
could give rise to another colorable motion to dismiss,”
Harris, 618 F.3d at 403, one could reasonably conclude
that Rule 12(b)(1) is the appropriate mechanism for
making KBR’s arguments. Our prior decision’s statement
is arguably dicta. At all events, it is nothing more than a
statement about the appropriate procedural posture for
analyzing political-question arguments rather than a
statement about the method to review § 2680 preemption
arguments.
This narrow reading is necessary because § 2680
7
A. The Political-Question Doctrine
KBR asserts its political-question argument as a
factual challenge to the District Court’s jurisdiction. See
Gould Elecs., Inc. v. United States, 220 F.3d 169, 176 (3d

questions like the one in this case are about preemption
rather than sovereign immunity. To be sure, § 2680 is
often invoked under Rule 12(b)(1) because it is an
assertion of sovereign immunity by the federal
government, which is jurisdictional. See Smith v. Johns-
Manville Corp., 795 F.2d 301, 306 n.8 (3d Cir. 1986)
(“The discretionary function exception [found in 28
U.S.C. § 2680(a)] operates as a bar to jurisdiction.”).
KBR, however, does not assert sovereign immunity.
Instead, it argues that § 2680(j) represents a federal
policy that preempts the plaintiffs’ claims. See Boyle, 487
U.S. at 508–10. Preemption arguments, other than
complete preemption, relate to the merits of the case. In
re U.S. Healthcare, Inc., 193 F.3d at 160. Therefore, the
appropriate procedural device for reviewing the § 2680(j)
preemption argument is not a motion pursuant to Rule
12(b)(1), but rather a motion under either Rule 12(b)(6)
or for summary judgment, Fisher v. Halliburton, 667
F.3d 602, 608–09 (5th Cir. 2012)—as the District Court
seemed to intuit.
The plaintiffs have waived any argument related to this
error, however, so it is beyond the scope of our review.
See Kost v. Kozakiewicz, 1 F.3d 176, 183 (3d Cir. 1993).
8
Cir. 2000) (“A Rule 12(b)(1) motion may be treated as
either a facial or factual challenge to the court’s subject
matter jurisdiction.”). In such a challenge, the burden of
proving that jurisdiction exists lies with the plaintiff, and
“the trial court is free to weigh the evidence and satisfy
itself as to the existence of its power to hear the case.”
Mortensen v. First Fed. Sav. & Loan Ass’n, 549 F.2d
884, 891 (3d Cir. 1977). Furthermore, “no presumptive
truthfulness attaches to plaintiff’s allegations,” and when
jurisdiction is intertwined with the merits, “the existence
of disputed material facts will not preclude the trial court
from evaluating for itself the merits of jurisdictional
claims.” Id.
We exercise plenary review over the District
Court’s legal conclusions but review its factual findings
for clear error. CNA v. United States, 535 F.3d 132, 139
(3d Cir. 2008). Here, the plaintiffs challenge both factual
findings and legal conclusions. The plaintiffs’ factual
arguments, however, are without merit because the Court
did not make the factual findings they argue that it did.2
We will thus exercise plenary review over the District

2
The plaintiffs argue that the District Court erroneously
found that KBR did not install or work on the pump that
caused Staff Sergeant Maseth’s death. The District Court
made no such findings. Instead, it explained that whether
KBR did install or work on the pump could be
reasonably disputed by the parties.
9
Court’s legal conclusion that this case presents a
nonjusticiable political question.

A case presents a nonjusticiable political question
when one of the following characteristics is “inextricable
from the case”:

(1) a textually demonstrable constitutional
commitment of the issue to a coordinate
political department; (2) or a lack of
judicially discoverable and manageable
standards for resolving it; (3) or the
impossibility of deciding without an initial
policy determination of a kind clearly for
nonjudicial discretion; (4) or the
impossibility of a court’s undertaking
independent resolution without expressing
lack of the respect due coordinate branches
of government; (5) or an unusual need for
unquestioning adherence to a political
decision already made; (6) or the potentiality
of embarrassment from multifarious
pronouncements by various departments on
one question.

Baker v. Carr, 369 U.S. 186, 217 (1962). KBR argues
that resolving the plaintiffs’ claims will require judicial
intrusion into issues textually committed to the executive,
present issues that lack judicially manageable standards,
and express a lack of respect due to coordinate branches
10
of government. Assessing this argument requires a
“discriminating inquiry into the precise facts and posture
of the particular case,” id., in a level of detail and
complexity that is rare even in the political-question
context.
Often, when the political-question doctrine is
asserted, nonjusticiability arises from the possibility that
one branch of government has exceeded its powers and
the court must decide whether it has the authority and
competence to regulate the alleged abuse. See, e.g.,
Zivotofsky v. Clinton, 132 S. Ct. 1421, 1430–31 (2012)
(holding that determining whether a statute allowing
Americans born in Jerusalem to indicate Israel as their
place of birth, which was argued to represent a
congressional infringement on executive prerogatives,
was not a political question). As such, when deciding
whether a case presents a political question, we rarely
need to look beyond the complaint and any of its obvious
implications.
This is not so with complaints against defense
contractors. Defense contractors do not have independent
constitutional authority and are not coordinate branches
of government to which we owe deference. See Taylor v.
Kellogg, Brown & Root Servs., Inc., 658 F.3d 402, 409
(4th Cir. 2011) (recognizing that “KBR is not a part of
the military”). Consequently, complaints against them for
conduct that occurs while they are providing services to
the military in a theater of war rarely, if ever, directly
11
implicate a political question. Nonetheless, these suits
may present nonjusticiable issues because military
decisions that are textually committed to the executive
sometimes lie just beneath the surface of the case. For
example, a contractor’s apparently wrongful conduct may
be a direct result of an order from the military,
Carmichael v. Kellogg, Brown & Root Servs., Inc., 572
F.3d 1271, 1281–83 (11th Cir. 2009) (holding that a
nonjusticiable issue is introduced when contractor-caused
harm was a result of following orders from a convoy
commander), or a plaintiff’s contributory negligence may
be directly tied to the wisdom of an earlier military
decision, Taylor, 658 F.3d at 411–12 (holding that a
nonjusticiable issue is introduced when contributory
negligence is based on the plaintiff’s disregard of an
earlier military decision). In these situations, the political
question appears not from the plaintiff’s claims but from
the broader context made relevant by a contractor’s
defenses. As such, to avoid infringing on other branches’
prerogatives in war-time defense-contractor cases, courts
must apply a particularly discriminating inquiry into the
facts and legal theories making up the plaintiff’s claims
as well as the defendant’s defenses. Lane v. Halliburton,
529 F.3d 548, 565 (5th Cir. 2008) (“We must look
beyond the complaint, considering how the Plaintiffs
might prove their claims and how KBR would defend.”).

1. Textual Commitment
12
 The Fifth and Eleventh Circuits have provided a
helpful framework for deciding whether a suit against a
defense contractor contains issues textually committed to
another branch. Because defense contractors are not
coordinate branches of government, a determination must
first be made whether the case actually requires
evaluation of military decisions. If so, those military
decisions must be of the type that are unreviewable
because they are textually committed to the executive.
See id. at 560; McMahon v. Presidential Airways, Inc.,
502 F.3d 1331, 1359–60 (11th Cir. 2007).3 According to

3
Although the Fourth and Ninth Circuits—the only two
circuits to have previously addressed this issue—do not
use this framework, their analyses are consistent with it.
In Taylor v. Kellogg, Brown & Root Servs., Inc., the
Fourth Circuit decided that a marine’s negligence claim
against KBR was nonjusticiable. 658 F.3d at 404. The
marine was electrocuted while installing a second
generator to a tank ramp that the military had not
authorized. Id. Relying heavily on the Fifth and Eleventh
Circuits’ decisions, the Fourth Circuit held that the case
presented a political question because resolving KBR’s
contributory-negligence defense would require
evaluating whether the military was correct to not
authorize the second generator installed by the marine—a
question “beyond the scope of judicial review.” Id.
at 411–12 & n.13.
13
KBR, this case would require judicial review of the
military’s decisions about where to house soldiers on a
battlefield—decisions that are unreviewable because they
involve strategic calculi about how best to defend against
threats. See McMahon, 502 F.3d at 1359 (“‘The strategy
and tactics employed on the battlefield are clearly not
subject to judicial review.’” (quoting Tiffany v. United
States, 931 F.2d 271, 277 (4th Cir. 1991)). Consequently,
the parties have focused on the first element of the
framework: whether the plaintiffs’ claims can be resolved

Koohi v. United States, 976 F.2d 1328 (9th Cir. 1992),
involved a suit against the United States and a defense
contractor for shooting down a civilian airliner off the
Iranian coast. Id. at 1330–31. The Ninth Circuit decided
that although the suit involved “conduct [that] took place
as part of an authorized military operation,” the suit was
not barred by the political-question doctrine because the
claims were for “judicially cognizable injury” that
resulted from “military intrusion into the civilian sector.”
Id. at 1331–31 (quoting Laird v. Tatum, 408 U.S. 1, 15–
16 (1972)); see also The Paquete Habana, 175 U.S. 677
(1900) (reviewing military’s seizure of two Spanish
fishing vessels during the Spanish-American war).
Viewed under the Fifth and Eleventh Circuits’
framework, the case was justiciable because the second
condition was not met—although the case required
evaluation of military decisions, they were the type that
is reviewable.
14
without evaluating these military decisions.
Military control over a contractor’s actions is one
common way that evaluation of strategic military
decisions becomes necessary. Military control requires
evaluation of military decisions because if the contractor
is simply doing what the military ordered it to do, then
review of the contractor’s actions necessarily includes
review of the military order directing the action. See
Carmichael, 572 F.3d at 1281–83 (holding that a suit for
damages arising from a convoy crash included a
nonjusticiable issue because of the degree of control the
military had over the convoy, such as selection of path,
speed, and distance between vehicles). However, where
the military does not exercise control but merely provides
the contractor with general guidelines that can be
satisfied at the contractor’s discretion, contractor actions
taken within that discretion do not necessarily implicate
unreviewable military decisions. See McMahon, 502 F.3d
at 1360–61 (holding that a defense contract for aviation
transportation in Afghanistan did not include sufficient
military control to introduce a political question because
the contractor retained authority over the type of plane,
flight path, and safety of the flight).

In this case, the contracts between the military and
KBR fit within the latter category. They provide KBR
with significant discretion over how to complete
authorized work orders. This discretion is best evidenced

15
by the lack of detailed instructions in the work orders4
and the lack of military involvement in completing
authorized work orders. See id. Military control over
KBR’s relevant activities therefore does not introduce an
unreviewable military decision into the case.
Our analysis does not end here however. Plaintiffs’
claims might still present unreviewable military decisions
if proving those claims or KBR’s defenses necessarily
requires evaluating such decisions. See Taylor, 658 F.3d
at 410–12. Accordingly, we must review every claim and
defense in the case. Ultimately, whether the claims or
defenses introduce a political question depends on which
state’s law applies. We will thus remand so the District
Court may undergo a choice-of-law analysis.
a) The Plaintiffs’ Claims

4
For example, one work order contained the problem
complained of—“pipes (shower & sink) have voltage[,]
get shocked in shower & sink”—but did not instruct
KBR how to solve this problem. J.A. at 2013. KBR
marked the project complete but did not explain what it
did. J.A. at 2014. And when the military gave directions,
those directions were quite minimal. See J.A. at 2015
(work order to solve “[w]ater pump leaking on top of
bldg thru roof” that directs KBR to fix by “replac[ing]
pressure switch”).
16
 The plaintiffs’ claims center on KBR’s failure to
ground or bond the water pump when KBR allegedly
installed or maintained the pump. As to installation, the
plaintiffs allege that if KBR installed the pump, then it
was negligent for not grounding or bonding the pump as
required by the standard of care set by KBR’s contract
with the military. As to maintenance of the pump, the
plaintiffs allege that (1) KBR had a contractual duty to
respond to work orders with safe work, (2) soldiers in
Staff Sergeant Maseth’s barracks complained of shocks
that were reported to KBR in authorized work orders, (3)
KBR could have eliminated the risk of electrocution
under these work orders, but (4) it was negligent in
failing to eliminate or recognize that risk.5 Although
determining the validity of these claims will require
acknowledgement of some strategic military decisions,
neither theory requires second-guessing the wisdom of
those decisions.

5
The plaintiffs did include several other claims in their
complaint, which the District Court dismissed because
they directly called into question strategic military
decisions. Harris, 878 F. Supp. 2d at 574. These liability
theories were the failure to warn, remedy the risk, rewire
the building, provide safe alternatives, and properly
maintain the facility. Id. We do not understand the
plaintiffs to appeal this ruling because their briefs focus
solely on the theories explained above.
17
 The installation theory is based on KBR’s alleged
installation of the pump between March 2006 and
February 2007. At that time, KBR was operating under a
CENTCOM6 contract with the U.S. Army Corps of
Engineers. This contract acknowledged that “[e]xisting
. . . electrical systems are in poor condition” and required
KBR only to maintain the systems in their “existing”
state. J.A. at 1645. Nonetheless, any completed electrical
work was required to “operate as originally intended and
designed, and in a safe manner.” J.A. at 1644. The parties
dispute what “safe manner” means. KBR argues that it is
not associated with any particular standard, while the
plaintiffs argue that it refers to American and British
electrical safety standards. So if the plaintiffs can show
that KBR actually installed the pump—a disputed factual
question—then whether KBR was negligent depends
entirely on the standard of care established by the
contract.

To be sure, determining that standard will require a
court to interpret the contract, which may require
testimony from military officials. But such testimony
would do no more than provide information about how to
interpret the term “safe manner”; their testimony would
not require the fact finder to determine whether the

6
CENTCOM stands for United States Central
Command—the United States’ military command in the
Middle East.
18
military was negligent in setting the “safe manner”
standard in the contract. And once the meaning of “safe
manner” is determined, evaluating whether KBR’s work
complied with that standard is a factual question for the
fact finder—a question that, again, does not require
evaluating any military decision. The plaintiff’s
installation theory therefore does not require evaluating
any unreviewable military decisions.
The same is true for the plaintiffs’ maintenance
theory. KBR allegedly performed, or should have
performed, maintenance to the pump under a different
contract, the LOGCAP7 III, Statement of Work and Task
Order 139. This contract divided buildings located on the
base into three categories—Level A, B, or C. KBR was
tasked with refurbishing and providing preventative
maintenance to Level A buildings. However, for Level B
buildings like the one in which Staff Sergeant Maseth
was electrocuted, KBR was not to perform preventative
maintenance—it was required only to complete
maintenance requested through work orders. These work
orders were initiated through complaints submitted to on-
base field officers, known as “camp mayors,” who would

7
LOGCAP—the Logistics Civil Augmentation
Program—is a program to “preplan for the use of civilian
contractors to perform selected [support] services in
wartime to augment Army forces.” U.S. Army
Regulation 700-137 § 1-1 (1985).
19
review the complaints and submit work orders to KBR if
the work was to cost less than a fixed amount. J.A. at
1718. If a work order exceeded KBR’s contractual
authority, then KBR was to return it to the camp mayor.
J.A. at 1718.
According to the plaintiffs’ maintenance theory,
KBR should have properly grounded and bonded the
pump when it responded to one of several work orders.
Although none of these work orders requested
maintenance on the pump that caused Staff Sergeant
Maseth’s death, the plaintiffs argue that KBR’s
completion of other work orders complaining of shocks
in the same building is circumstantial evidence that KBR
must have (or, at least, should have) performed some
maintenance on that water pump.8

8
Underlying their argument that KBR must have
performed this maintenance is a factual dispute over
whether KBR could have performed such maintenance
within the scope of the contract. Resolving this issue
depends on whether the maintenance would have
required KBR to rewire the entire building or just to
ground and bond the water pumps—the former is
presumably beyond the cost constraints of the contract
while the latter is not. This is a question for the fact
finder to resolve through evaluation of the competing
experts’ testimony.
20
 This theory, like the installation theory, is based
solely on whether KBR satisfied its contract duties. The
plaintiffs do not, for example, argue that the military
should have categorized Staff Sergeant Maseth’s
barracks as Level A or should have submitted a work
order for the pump. They argue only that KBR failed to
satisfy the contractual standards for maintaining Level B
buildings. The LOGCAP contract’s standard of care is
currently unresolved—this time because the contract is
silent on the question. Interpreting the contract’s standard
of care will again require applying principles of contract
interpretation, and may require some military officers to
testify. But just like the installation theory, this
interpretive question can be resolved without second-
guessing military decisions.

As a result, neither of the plaintiffs’ liability
theories requires evaluating the wisdom of the military’s
decisions. Accordingly, neither justifies dismissing this
case on political-question grounds.
b) KBR’s Assumption-of-the-Risk Defense
While the plaintiffs’ liability theories do not
implicate strategic military decisions, KBR asserts three
defenses that may: assumption of the risk, proximate
cause, and contributory negligence. When analyzing
whether a proposed defense implicates a nonjusticiable
issue in a Rule 12(b)(1) factual challenge, courts must
first decide whether the defendant has “present[ed]
21
sufficient evidence to permit a jury to conclude that he
established the [elements of the] defense by a
preponderance of the evidence.” United States v. Stewart,
185 F.3d 112, 125 (3d Cir. 1999). If there is sufficient
evidence to support the defense, then the District Court
must determine whether the defense actually presents a
nonjusticiable issue. If it does introduce such an issue,
then the case is dismissed.9 But if there is insufficient
evidence to support the defense, or if the defense does
not present a nonjusticiable issue, then the case goes
forward. Applying this framework, we conclude that
KBR’s assumption-of-the-risk defense is justiciable
because that defense does not require evaluating
unreviewable military decisions. Yet KBR’s contributory
negligence and proximate cause defenses may present
nonjusticiable issues, depending on which state’s law

9
The parties do not dispute that the introduction of a
nonjusticiable issue by a defense requires the dismissal of
the entire case rather than elimination of the defense.
This assumption is also made by several of our sister
courts of appeals. See, e.g., Taylor, 658 F.3d at 409;
Carmichael, 572 F.3d at 1292; Lane, 529 F.3d at 565.
We follow suit and, at least for now, adopt this
assumption. But we acknowledge that dismissing the
entire case is not the only possible conclusion, as
evidenced by the remedy for the introduction of
nonjusticiable issues by damages estimates discussed in
Part II.A.1.c infra.
22
applies.
The District Court analyzed KBR’s assumption-of-
the-risk defense under Pennsylvania law.10 This defense
bars any recovery if a defendant can show that the
injured party knew of the dangerous condition, which
was both obvious and avoidable, yet still voluntarily
encountered it. Kaplan v. Exxon Corp., 126 F.3d 221,
226 (3d Cir. 1997) (quoting Carrender v. Fitterer, 469

10
The District Court has not yet determined if
Pennsylvania, Tennessee, or Texas law applies. Harris,
878 F. Supp. 2d at 567. It sensibly restricted its analysis
of KBR’s assumption-of-the-risk defense to Pennsylvania
law because the parties relied on it alone and because this
defense is not available under Tennessee and Texas law.
Id. at 567 & n.32; see also Del Lago Partners, Inc. v.
Smith, 307 S.W.3d 762, 772 n.34 (Tex. 2010) (explaining
that in Farley v. M.M. Cattle Co., 529 S.W.2d 751, 758
(Tex. 1978), the Texas Supreme Court “abolish[ed]
implied assumption of the risk but retain[ed] [the]
affirmative defense of express assumption of the risk”—
the latter of which is when a plaintiff “explicitly consents
to take personal responsibility for potential injury-
causing risks”); Baggett v. Bedford Cnty., 270 S.W.3d
550, 554 (Tenn. Ct. App. 2008) (explaining that “[t]he
Tennessee Supreme Court abolished the defense of
implied assumption of risk in [Perez v. McConkey, 872
S.W.2d 897, 905 (Tenn. 1994)]”).
23
A.2d 120, 125 (Pa. 1983)). Voluntariness requires that
the injured party “had a real ‘choice.’” Id. (citing Howell
v. Clyde, 620 A.2d 1107, 1112 (Pa. 1993)). KBR argues
that Staff Sergeant Maseth assumed the risk of
electrocution because when he took the fatal shower, he
was aware of the risks of taking a shower in his barracks
but chose to do so despite the military’s provision of safe
alternative showering facilities. KBR is entitled to
present this defense to a jury because it has presented
evidence supporting Staff Sergeant Maseth’s awareness
of and voluntary exposure to the risk of electrocution.
Importantly, the voluntariness of his choice to use the
shower is evidenced by the availability of alternative
showering facilities provided by the military.
The District Court found that analyzing
voluntariness would draw strategic military decisions
into the case because it would require the plaintiffs
“either [to] admit that Maseth voluntarily encountered the
risk in the shower, an admission which would undermine
their case, or [to] take the position that his actions were
involuntary such that he was acting in response to
military orders and directly challenge the military’s
decision concerning the shower facilities which were
made available to him at the base.” Harris, 878 F. Supp
2d at 587. But those are not the only possibilities. The
plaintiffs may argue, for instance, that the alternative
facilities were not available to Staff Sergeant Maseth or,
if they were, that he was not aware of them. If either of
24
these propositions is true, then he could not have avoided
the risk under Pennsylvania law. Neither of these
arguments implicates strategic military decisions.
Whether the military should have provided Staff Sergeant
Maseth with alternative showering facilities, as KBR
intends to argue, is entirely irrelevant to whether such
facilities were available to him and, if they were, whether
he was aware of them. Furthermore, although the
evidence appears to weigh against them, the plaintiffs
may still dispute whether Staff Sergeant Maseth was
aware of the risk, which has nothing to do with
unreviewable military decisions. KBR’s assumption-of-
the-risk defense thus does not introduce a nonjusticiable
question because the merits of this defense depend solely
on facts that do not implicate strategic military decisions.

c) KBR’s Proximate-Cause Defense

KBR also argues that its proximate-cause defense
makes this case nonjusticiable. KBR emphasizes its
intent to argue that the military’s actions were the sole
cause of Staff Sergeant Maseth’s death. A variation of
this defense, which the District Court referred to, is the
ability of KBR to argue that the military was a proximate
cause of the death. The District Court found that both
versions of KBR’s proximate-cause defense would
require evaluating military judgments.
KBR has presented sufficient evidence to support
both of these defenses. Under relevant state law, a
25
defendant can avoid liability by demonstrating that a
third party is the true proximate cause of the harm.
Harris, 796 F. Supp. 2d at 658–60 (collecting cases).
KBR has presented sufficient evidence to support its
argument that the military, rather than KBR, was the
exclusive proximate cause of Staff Sergeant Maseth’s
death. Under both contracts between KBR and the
military, the parties shared responsibility for maintaining
buildings in the Radwaniyah Palace Complex. The
military retained authority to perform its own
maintenance. See, e.g., J.A. at 701 (recording Specialist
Michael Skaggs’ testimony regarding maintenance work
he completed while serving in the complex).
Furthermore, the military was ultimately responsible for
life support functions at the base—which is exemplified
by the military’s retention of authority to approve
projects before KBR could perform any work and by
occasional decisions to ignore KBR’s maintenance
advice. See, e.g., J.A. at 649 (explaining that Level B
facilities were to be maintained only on request); J.A.
at 500 (recounting KBR’s initial desire to estimate a
Level A maintenance cost for the barracks that the
military rejected).

This shared responsibility leaves open the
possibility that the military alone caused Staff Sergeant
Maseth’s death. As the District Court explained, KBR
could prove that the military is the sole cause if the
military (1) installed the pump improperly and never
26
subsequently grounded or bonded it, (2) performed
maintenance on the pump that caused it to be ungrounded
and unbonded, (3) never provided KBR the authority to
fix it because it was outside of the contract’s scope, or
(4) never submitted a work order to fix the pump. Any of
these possibilities would mean that KBR had no
contractual duty to repair the pump. Because KBR has
provided sufficient evidence of these possibilities, this
defense may go forward.
The plaintiffs argue that the defense that the
military was a proximate cause is unavailable because
the relevant evidence shows only military actions that are
outside of the scope of their claims, and whose
connection to this case is too attenuated to be a proximate
cause. Deciding whether a party is a proximate cause
varies slightly between the relevant states. The District
Court has determined that Iraqi law does not apply but
has not decided if Pennsylvania, Tennessee, or Texas law
applies. Harris, 878 F. Supp 2d at 567. Starting with
Texas law, defendants are the proximate cause of an
injury if their conduct was the cause in fact of the harm
suffered and if the harm is the foreseeable result of that
conduct. Travis v. City of Mesquite, 830 S.W.2d 94, 98
(Tex. 1992). Harm is foreseeable when “the actor, as a
person of ordinary intelligence, should have anticipated
the dangers that his negligent act created for others.” Id.

Electrocution was a reasonably foreseeable result
of several strategic military decisions. The military was
27
aware that the buildings in the Radwaniyah Palace
Complex had substandard electrical systems that posed
the specific risk of electrocution in shower facilities. J.A.
at 324 (discussing the military’s warning to troops about
the risks of electrocution from showers in existing
buildings); J.A. 431–32 (recording the statement of an
Army general explaining that the military was aware of
the risks of placing troops in existing buildings); J.A.
at 1645 (recognizing in the contract that the electrical
systems were in poor condition). Nevertheless, the
military chose to assign personnel to live in these
barracks because the risk of electrocution was minor
compared to the risks from external threats, such as
missile and mortar attacks. J.A. at 432. From KBR’s
perspective, the military foresaw the exact harm suffered
by Staff Sergeant Maseth. Indeed, KBR’s argument is
bolstered by the military’s decision to contract with KBR
to repair the electrical problems in buildings like Staff
Sergeant Maseth’s only in response to a work order, even
though (1) KBR initially recommended that Staff
Sergeant Maseth’s barracks be categorized as Level A,
(2) KBR informed the military of the barracks’
significant electrical problems, J.A. at 500, and (3) the
military was aware of shocking in the building from
service-member complaints.
KBR argues that the military therefore must have
anticipated that electrocutions were a risk of its decision
not to categorize the building Level A and not to have
28
KBR repair the building’s electrical system. See J.A.
at 433 (recounting a general’s testimony that an event
like Staff Sergeant Maseth’s death could have occurred
in any number of facilities throughout Iraq because of
military decisions). Additionally, these decisions
establish cause in fact: but for the military’s decisions to
house troops in dangerous buildings that were not to be
repaired, the staff sergeant’s death would not have
occurred. KBR has therefore presented sufficient
evidence to invoke its proximate-cause defense under
Texas law.
The same is true under Pennsylvania and
Tennessee law. Although the tests are not identical, both
states essentially ask whether “(1) the tortfeasor’s
conduct must have been a ‘substantial factor’ in bringing
about the harm being complained of; and (2) there is no
rule or policy that should relieve the wrongdoer from
liability because of the manner in which the negligence
has resulted in the harm.” Hale v. Ostrow, 166 S.W.3d
713, 719 (Tenn. 2005) (evaluating proximate cause based
on these two factors and whether the harm could have
been reasonably foreseen are considered in evaluating
proximate cause); see also Lux v. Gerald E. Ort
Trucking, Inc., 887 A.2d 1281, 1287 (Pa. Super. Ct.
2005) (evaluating proximate cause based only on these
two factors). The second of these elements is essentially
the same as the foreseeability analysis under Texas law,
see Wisniewski v. Great Atl. & Pac. Tea Co., 323 A.2d
29
744, 748 (Pa. Super. Ct. 1974) (citing Majors v.
Brodhead Hotel, 205 A.2d 873 (Pa. 1965)), so KBR has
adduced sufficient evidence to satisfy this element for the
same reasons it can show foreseeability under Texas law.

As to the first element, whether the military’s
decisions were a “substantial factor” depends on three
factors:
(a) the number of other factors which
contribute in producing the harm and the
extent of the effect which they have in
producing it;
(b) whether the actor’s conduct has created a
force or series of forces which are in
continuous and active operation up to the
time of the harm, or has created a situation
harmless unless acted upon by other forces
for which the actor is not responsible; [and]

(c) lapse of time.
Lux, 887 A.2d at 1287. KBR’s evidence supports a
finding that these factors show that the military was a
substantial factor in Staff Sergeant Maseth’s death. The
first factor, which is based on the Second Restatement of
Torts, asks whether there is one event that had such a
“predominant effect” that it should foreclose liability for
other events that contributed to the harm. Restatement

30
(Second) of Torts § 433 cmt. d. The evidence
demonstrates that there are at least two events that
contributed to the staff sergeant’s death: the military’s
maintenance decisions despite the known electrical
problems and KBR’s alleged negligent response to the
work orders. Yet even if KBR’s negligence caused the
harm, it is difficult to see why the negligence is so
predominant that it should foreclose any fault that is
plausibly attributable to the military for knowingly
placing service members in buildings with dangerous
electrical systems.
KBR has also presented sufficient evidence on the
second and third factors. As to the second factor, the
military’s decisions were a “continuous and active” force
“up to the time of the harm,” Lux, 887 A.2d at 1287,
because they created the environment for the harm to
occur and made electrocution likely to occur by using the
barracks with substandard electrical wiring. As to the
third factor, KBR has shown no “lapse of time,” Lux, 887
A.2d at 1287, because these military decisions were
essentially ongoing, as evidenced by the military’s
continual inaction regarding a technical inspection report
from KBR two months before Staff Sergeant Maseth
died. See J.A. at 525–27.

All of this is to say that KBR has adduced
sufficient evidence to present its defenses that the
military’s housing and maintenance decisions were at
least a proximate cause of the death and that they were
31
the proximate cause. Left unanswered, however, is
whether either of these defenses present a nonjusticiable
issue because they require evaluating unreviewable
military decisions.

KBR’s defense that the military was the sole cause
of Staff Sergeant Maseth’s death does not require such an
evaluation. As discussed above, KBR can successfully
use this defense if it proves any of the following: that the
military (rather than KBR) installed or performed faulty
maintenance on the pump, that fixing the electrified
showers was beyond the scope of KBR’s contract, or that
no work order was ever submitted that would have
required grounding or bonding of the pump or given
KBR reason to notice that it should be. Unsurprisingly,
several of these possibilities are related to existing factual
disputes between the parties. They disagree over whether
KBR installed the pump, could have fixed the problem
within the scope of the second contract, or responded to
work orders that would have required work on the
specific pump that caused Staff Sergeant Maseth’s death.
Resolving these disputes—and thus whether the cause of
the death was the sole fault of the military—does not
require evaluating military decisions. All of these
disputes are simply about who did what, and whether
KBR could have performed the work it failed to do under
the contract.

To be sure, resolving these disputes will require
submission of evidence that the military could have
32
installed or maintained the pump. Such evidence might
include the military’s shared responsibility for
maintaining life-support systems on the base and its
occasional performance of maintenance that contradicted
KBR’s recommendations. But the submission of
evidence related to strategic military decisions that are
necessary background facts for resolving a case involving
a defense contractor is not sufficient to conclude that a
case involves an issue textually committed to the
executive. Instead, the case must require evaluation of
those decisions such that the fact finder is asked to
reexamine their wisdom. See McMahon, 502 F.3d
at 1359–61 (explaining that a claim must require
“reexamination” of a military decision before holding
that the claim at issue did not implicate the political-
question doctrine even though military decisions were
relevant to the case). KBR’s defense that the military was
the sole cause of Staff Sergeant Maseth’s death does not
require such an evaluation because the disputes are
entirely factual: KBR did or did not install or maintain
the pump, did or did not have authority under the contract
to fix the showers, and did or did not receive a work
order that would have required it to fix the pump. The
District Court thus erred when it concluded that resolving
this defense would require determining whether the
military was negligent.

The other variation of KBR’s proximate-cause
defense—that the military was a proximate cause of Staff
33
Sergeant Maseth’s death—is another matter. It may
require evaluation of strategic military decisions, and
those questions turn on state law. If a jurisdiction uses a
proportional-liability system which assigns liability by
the degree of fault, then a proximate-cause defense
introduces a nonjusticiable issue. In such a system, there
is simply no way to determine damages without
evaluating military decisions. The fact finder cannot
decide the respective degrees of fault as between a
military contractor like KBR and the military without
evaluating the decisions made by each—particularly, the
military’s decisions to house troops in unsafe barracks
that would not be repaired. See Fisher v. Halliburton,
667 F.3d 602, 621–22 (5th Cir. 2012) (explaining that
Texas’s proportional-liability system could introduce a
political question but resolving the case on other
grounds).

Tennessee and Texas use proportional-liability
systems. McIntyre v. Balentine, 833 S.W.2d 52, 56
(Tenn. 1992); Tex Civ. Prac. & Rem. Code § 33.004. So
if Tennessee or Texas law applies, then damages cannot
be estimated without evaluating unreviewable military
decisions.11 Under Pennsylvania law, however, joint-and-

11
This conclusion depends on the ability of fact finders
to assign fault to immune parties, such as the
government. Both states permit this. The Tennessee
Supreme Court appears to have never dealt with the
34
several liability would apply. See 42 Pa.C.S.A. § 7102(b)
(West 2004); Act No. 2011-17, 195th Pa. Gen. Assemb.
(2011) (eliminating joint-and-several liability for actions
that accrue after the law’s enactment).12 So if
Pennsylvania law controls, then calculation of damages
does not require evaluating strategic military decisions
because the plaintiffs are free to obtain the entirety of

assignment of fault to the government but has stated
frequently that “a jury may generally apportion fault to
immune nonparties.” Carroll v. Whitney, 29 S.W.3d 14,
19 (Tenn. 2000). The Texas Supreme Court does not
appear to have dealt with this question, but one
intermediate appellate court has stated that the relevant
Texas statute allows assignment of fault to immune
nonparties. In re Unitec Elevator Servs. Co., 178 S.W.3d
53, 56 n.5 (Tex. App. 2005).
12
The liability rule could differ for other cases governed
by Pennsylvania law because the state only recently
eliminated joint-and-several liability for many torts. Act
No. 2011-17, 195th Pa. Gen. Assemb. § 1 (2011). But
this change only “appl[ies] to causes of action which
accrue on or after the effective date of this section [June
28, 2011].” Id. at § 3. Staff Sergeant Maseth was killed
on January 2, 2008. So the causes of action in this case
accrued before, rather than “on or after,” June 28, 2011.
Pennsylvania’s old rule of joint-and-several liability
would apply.
35
their relief from KBR. See Maloney v. Valley Med.
Facilities, Inc., 984 A.2d 478, 488–89 (Pa. 2009).

Whether KBR’s proximate-cause defense
implicates a nonjusticiable issue thus depends on which
state law controls. If the District Court decides that
Pennsylvania law applies, then the defense does not
introduce any nonjusticiable issues. But if the Court
decides that either Tennessee or Texas law applies, then
the defense will introduce such an issue. Even if
Tennessee or Texas law applies, though, only the fact
finder’s calculation of damages would be nonjusticiable.
This means that we can extract the nonjusticiable issue in
a manner that possibly preserves some of the plaintiffs’
claims by dismissing only the damages claims that rely
on proportional liability. See Baker, 369 U.S. at 217;
Powell v. McCormack, 395 U.S. 486, 517–18 (1969).
Accordingly, if the District Court determines that
Tennessee or Texas law applies, then it should not
dismiss the case. Instead, it should foreclose the plaintiffs
from obtaining the types of damages that are assigned
using proportional liability but allow the plaintiffs to
proceed on any damages claim that does not implicate
proportional liability (such as nominal damages, if
available).

Eliminating the plaintiffs’ claims for these
damages is the appropriate solution to the introduction of
a political question by KBR’s defense because remedies,
unlike breaches of a duty owed, can be extricated from a
36
case. We are mindful that the test from Baker is that one
of the listed factors must be “inextricable from the case.”
369 U.S. at 217. This suggests that if an issue can be
extracted from the case, then the case should be
permitted to proceed with that issue removed—which is
exactly what the District Court is directed to do if
Tennessee or Texas law applies.

Powell v. McCormack also suggests that this is the
correct approach. There, the Supreme Court analyzed
federal courts’ ability to “mold effective relief”
separately from “whether the duty asserted can be
judicially identified and its breach judicially determined.”
395 U.S. at 517–18 (quoting Baker, 369 U.S. at 198).
Importantly, when discussing the ability to provide relief,
the Court avoided deciding whether the request for
injunctive relief introduced a nonjusticiable issue.
Instead, it determined that the plaintiff’s request for
declaratory relief was justiciable. Id. This suggests that
when the request for one type of remedy is foreclosed by
the political-question doctrine, plaintiffs may proceed if
they are seeking other damages that do not implicate the
doctrine. Accordingly, because KBR’s argument that the
military was a proximate cause implicates unreviewable
strategic military decisions only because of the necessity
of apportioning fault, the plaintiffs may still proceed if
they seek any relief that does not implicate the
proportional-liability system.

d) KBR’s Contributory-Negligence Defenses
37
 Whether KBR’s contributory-negligence defense
presents a nonjusticiable issue also turns on the
applicable state law. KBR argues that it is not liable
because Staff Sergeant Maseth acted negligently when he
decided to take a shower in his barracks despite allegedly
knowing of the risk. Contributory negligence allows
defendants to avoid liability if they can show that the
injured party’s own negligence caused more than
50 percent of the harm.13 42 Pa. Cons. Stat. § 7102
(providing that a plaintiff’s negligence is not a bar to
recovery if “such negligence was not greater than the
causal negligence of the defendant”); McIntyre, 833
S.W.2d at 57 (holding that a plaintiff’s negligence bars
recovery only if it is not “less than . . . the defendant’s
negligence”); Tex. Civ. Prac. & Rem. § 33.001
(providing that “a claimant may not recover damages if

13
The denomination of this defense is confusing because
of state variations. While several states refer to it as
“contributory negligence,” see, e.g., Boyle v. Indep. Lift
Truck, Inc., 6 A.3d 492, 496 (Pa. 2010), others refer to it
as “modified-comparative negligence,” see, e.g.,
McIntyre v. Balentine, 833 S.W2d 52, 57 (Tenn. 1992).
We use the term “contributory negligence” primarily
because that is how the parties refer to it.
Although KBR does not currently make a comparative-
negligence argument, our analysis of its contributory-
negligence defense would apply to it with equal force.
38
his percentage of responsibility is greater than 50
percent”). KBR has presented sufficient evidence from
which a reasonable jury could find that Staff Sergeant
Maseth was aware of the electrocution risk and that safe
alternative showering facilities were available. Such
evidence could lead a fact finder to conclude that the
staff sergeant was negligent in using the barracks shower.

This defense might require evaluation of strategic
military decisions. To determine whether Staff Sergeant
Maseth’s alleged negligence caused more than 50 percent
of the harm, the degree of causation that can be assigned
as between the military’s alleged negligence and KBR’s
alleged negligence must also be determined. That is, the
proportion of the injured party’s fault cannot be decided
without also effectively deciding the extent to which the
negligence of other parties caused the harm. For
example, the relevant Pennsylvania Suggested Standard
Civil Jury Instruction requires that the jury determine if
the plaintiff’s negligence is greater than 50 percent by
assigning fault to each defendant and then to the plaintiff.
Pa. Bar Inst. Bd. of Dirs., Pennsylvania Suggested
Standard Civil Jury Instructions § 13.230 (4th ed. 2008).
That means that for the fact finder to find that Staff
Sergeant Maseth was, say, 60 percent at fault, the fact
finder would have to assign fault to KBR and the military
individually that summed to 40 percent. This assignment
of fault to the military inevitably would require
evaluating the wisdom of the strategic military decisions
39
that caused the death.14 This defense therefore might
require evaluation of strategic military decisions and
make this case nonjusticiable.

Whether it does, however, depends on whether
state law allows the fact finder to assign fault to
nonparties and whether Staff Sergeant Maseth was
negligent. The military is not a party to this suit and, as
explained, the source of the nonjusticiable issue in
KBR’s contributory-negligence defense is the need to
assign fault to the military to determine whether Staff
Sergeant Maseth was more than 50 percent responsible
for the harm suffered. So if state law does not permit the
assignment of fault to nonparties, then KBR’s defense
does not require assigning fault to the military or
evaluating strategic military decisions. As mentioned
above, it is yet to be determined if Pennsylvania,

14
Because this defense introduces a nonjusticiable issue
through the assignment of fault to the military, for KBR
to rely on this as a basis for dismissing this case on
political-question grounds, it must first present sufficient
evidence from which a reasonable jury could assign some
fault to the military for Staff Sergeant Maseth’s death.
For the reasons explained in the analysis of KBR’s
proximate-cause defense, KBR has met that standard by
presenting evidence that the military’s strategic decisions
were negligent and a proximate cause of the death. See
supra Part II.A.1.c.
40
Tennessee, or Texas law applies in this case. Harris,
878 F. Supp 2d at 567. These states differ on whether a
nonparty can be assigned fault by a fact finder deciding if
a plaintiff’s fault is greater than other tortfeasors’. As a
result, the District Court must determine which state’s
law applies before it can resolve whether KBR’s defense
introduces a nonjusticiable issue.

Pennsylvania does not permit assigning fault to
nonparties for the purpose of contributory-negligence
defenses. See 42 Pa. Cons. Stat. § 7102 (providing that a
plaintiff’s negligence bars recovery if it is “greater than
the causal negligence of the defendant or defendants
against whom recovery is sought” (emphasis added));
Kelly v. Carborundum Co., 453 A.2d 624, 627 (Pa.
Super. Ct. 1982) (rejecting the defendant’s argument that
§ 7102 permits “apportionment among all tortfeasors
causally responsible for an injury” and explaining that
the statute “merely provides for apportionment among
those defendants against whom recovery is allowed”).15

15
See also Thornton v. Philadelphia Hous. Auth., 4 A.3d
1143, 1153 (Pa. Commw. Ct. 2010) (“A plaintiff’s
recovery is barred only if his contributory negligence is
greater than the causal negligence of the defendants
against whom recovery is sought.”); Heckendorn v.
Consol. Rail Corp., 465 A.2d 609, 612 (Pa. 1983) (“[I]t
is clear that in the Comparative Negligence Act the
legislature did not contemplate an apportionment of
41
So if Pennsylvania law applies, then KBR’s contributory-
negligence defense—like its proximate-cause defense—
does not introduce a nonjusticiable issue.

Tennessee and Texas, however, are another matter.
These states permit fault to be assigned to nonparties for
the purposes of contributory negligence. See Mullins v.
State, 294 S.W.3d 529, 536 (Tenn. 2009) (explaining two
principles of Tennessee tort law, which “are that all
tortfeasors must be joined in the suit unless joinder is
specifically prohibited by law . . . and that parties may
assert, as an affirmative defense, that another party or
even a non-party is responsible for the plaintiff’s
injuries”); Tex. Civ. Prac. & Rem. § 33.001 (providing
that “a claimant may not recover damages if his
percentage of responsibility is greater than 50 percent”);
id. § 33.003 (providing that fact finders must assign
responsibility to “each claimant; each defendant; each
settling person; and each responsible third party who has
been designated under § 33.004 [which contains several
procedural requirements]”); Martin K. Eby Const. Co. v.
LAN/STV, 350 S.W.3d 675, 680 (Tex. App. 2011)

liability between one or more third party tortfeasors
(against whom recovery may be had) and the plaintiff’s
employer (against whom recovery may neither be sought
nor allowed).”); Morris v. Lenihan, 192 F.R.D. 484, 492
(E.D. Pa. 2000) (“[A]pportionment may only take place
among parties that are properly in the case.”).
42
(“Section 33.003 requires the trier of fact to determine
the percentage of responsibility for each claimant,
defendant, settling person, and responsible third party
who “caus[ed] or contribut[ed] to cause in any way the
harm for which recovery of damages is sought . . . .”). So
if Tennessee or Texas law applies, then KBR’s
contributory-negligence defense introduces a
nonjusticiable issue as long as KBR can show that Staff
Sergeant Maseth acted negligently.
KBR must be able to show that Staff Sergeant
Maseth acted negligently for its contributory-negligence
defense to introduce a nonjusticiable issue into this case
under Tennessee or Texas law. If he was not negligent,
then there is no need to determine the degree of fault for
which the military is responsible. As explained, only the
comparison of Staff Sergeant Maseth’s negligence to that
of KBR’s and the military’s implicates nonjusticiable
issues. Deciding whether the staff sergeant was negligent
does not. This, like the assumption-of-risk defense,
depends entirely on factual questions regarding his
knowledge of the risk and the availability of alternative
showers. Unlike in Taylor, where the injured party’s
alleged negligence was that party’s decision to ignore a
strategic military decision about the number of generators
a tank ramp needed, 658 F.3d at 410–11, there is no
evidence that Staff Sergeant Maseth was second-guessing
a military decision about showering by using the shower
in his barracks. So if the District Court concludes that
43
Tennessee or Texas law applies, then the fact finder must
first determine whether Staff Sergeant Maseth was
negligent. If he was, then the case must be dismissed as
nonjusticiable. If he was not, then the case will proceed
to the merits.
2. The Remaining Political-Question Factors

Resolution of the remaining political-question
factors—whether this case presents issues that lack
judicially manageable standards or that cannot be
resolved without affording respect to the coordinate
branches of government—turns on the same analysis.
Both of these bases for nonjusticiability are inextricable
from this case if the fact finder must evaluate the wisdom
of the military’s housing and maintenance decisions. And
regarding the lack of a judicially manageable standard,
“it is difficult to conceive of an area of governmental
activity in which the courts have less competence. The
complex, subtle, and professional decisions as to the
composition, training, equipping, and control of a
military force are essentially professional military
judgments.” Gilligan v. Morgan, 413 U.S. 1, 10 (1973).
Housing and maintenance decisions on a battlefield are
exactly this type of decision—complex, subtle, and
professional decisions within the military’s professional
judgment and beyond courts’ competence. For this same
reason, resolving a case requiring evaluation of these
decisions would also fail to express the respect due to the
coordinate branches of government. See Aktepe v. United
44
States, 105 F.3d 1400, 1404 (11th Cir. 1997) (explaining
that such respect is not shown when courts “subject[] [the
political branches’] discretionary military and foreign
policy decisions to judicial scrutiny, notwithstanding the
judiciary’s relative lack of expertise in these areas”).
Whether this case includes an issue whose
resolution would express a lack of respect or that lacks a
manageable standard thus turns on whether a strategic
military decision must be reviewed. This is the same
question that controlled our earlier analysis of whether
this case contains an issue textually committed to another
branch. Consequently, the remaining political-question
factors will be inextricable from this case only if the case
presents an issue textually committed to another branch.
As a result, if Pennsylvania law controls, then this case
lacks any nonjusticiable issues. But if either Tennessee or
Texas law controls, then the case contains nonjusticiable
issues that require eliminating any damages based on
proportional liability. In such instance, if Staff Sergeant
Maseth is found contributorily negligent, the case should
be dismissed.
B. Section 2860(j) Combatant-Activities Preemption

The District Court alternatively held that the
plaintiffs’ claims are preempted by the combatant-
activities exception to the Federal Tort Claims Act. The
Federal Tort Claims Act waives the United States’
sovereign immunity for many tort claims against it. 28
45
U.S.C. § 2674. But that waiver contains numerous
exceptions, one of which—the combatant-activities
exception—is raised here. Under the combatant-activities
exception, the United States remains immune from
“[a]ny claim arising out of the combatant activities of the
military or naval forces, or the Coast Guard, during time
of war.” 28 U.S.C. § 2680(j). Of course, defense
contractors are not part of the government, so concepts
like sovereign immunity, waiver of sovereign immunity,
and exceptions to waiver do not apply directly to defense
contractors. In fact, the Federal Tort Claims Act says as
much. 28 U.S.C. § 2671 (stating that “Federal agency”
“does not include any contractor with the United States”).
But the Supreme Court has held that the Act’s
exceptions sometimes express federal policies that
impliedly preempt state claims against defense
contractors providing services to the military. In Boyle v.
United Technologies, the Court held that another
exception—§ 2860(a)’s discretionary-function
exception—provides a federal policy that preempts state
tort law interfering with it. 487 U.S. at 511–12. The
question before the Court was whether a claim for
defective design against a helicopter manufacturer was
preempted. Id. at 503. The Court first recognized that
there is a federal interest in federal-government contracts
with private parties that is implicated in suits by private
parties against a government contractor for conduct
resulting from the government contract. Id. at 504–07.
46
 To determine whether the plaintiff’s state claim
conflicted with this federal interest, the Court relied on
the discretionary-function exception to establish the
scope of the preempting policy. Id. at 510–11. This
exception prevents suits against the United States for
“[a]ny claim . . . based upon the exercise or performance
or the failure to exercise or perform a discretionary
function or duty on the part of a federal agency or an
employee of the Government, whether or not the
discretion involved be abused.” 28 U.S.C. § 2680(a).
From this exception, the Court derived a federal policy
for avoiding second-guessing government decisions that
“often involve[] not merely engineering analysis but
judgment as to the balancing of many technical, military,
and even social considerations.” Boyle, 487 U.S. at 511.
And because state design-defect claims against
“contractors would produce the same effect sought to be
avoided by the FTCA exception,” id., these claims must
also be preempted.
Importantly, the Court did not determine whether
the state laws in question were preempted by simply
applying the statute as if the contractor were the federal
government. Instead, it created a three-part test designed
to protect the federal policy underlying § 2680(a). Id.
at 512 (holding that state claims against procurement
contractors are preempted if “(1) the United States
approved reasonably precise specifications; (2) the
equipment conformed to those specifications; and (3) the
47
supplier warned the United States about the dangers in
the use of the equipment that were known to the supplier
but not to the United States”).

To decide how Boyle applies to § 2680(j), we must
undertake the same analytic process. Boyle’s analysis
involved three steps: (1) identify a unique federal interest
that is associated with a FTCA exception, (2) determine
the scope of the policy that underlies the exception, and
(3) derive a test that ensures preemption of state laws that
frustrate this policy. The two circuits that have
confronted this agree that § 2680(j) represents a unique
federal interest in the management of wars. Saleh v. Titan
Corp., 580 F.3d 1, 5–7 (D.C. Cir. 2009); Koohi, 976 F.2d
at 1336–37. But they disagree over the scope of the
federal policy underlying the exception and, as a
consequence, what test should follow.

The Ninth Circuit, in Koohi, held that “one
purpose of the combatant activities exception is to
recognize that during wartime encounters no duty of
reasonable care is owed to those against whom force is
directed as a result of authorized military action.” 976
F.2d at 1337. By contrast, in Saleh, the D.C. Circuit held
that “the policy embodied by the combatant activities
exception is simply the elimination of tort from the
battlefield, both to preempt state or foreign regulation of
federal wartime conduct and to free military commanders
from the doubts and uncertainty inherent in potential
subjection to civil suit.” 580 F.3d at 7. This latter, more
48
expansive, policy is partially based on § 2680(j)’s use of
“arising out of,” which we know from “workmen’s
compensation statutes to denote any causal connection,”
id. at 6.

There is very little authority for us to rely on to
resolve this disagreement. The Federal Tort Claims Act
does not explicitly state the purpose of the exception, nor
does legislative history exist to shed light on it. Johnson
v. United States, 170 F.2d 767, 769 (9th Cir. 1948) (“An
examination of the record fails to produce clear evidence
of Congressional intent or policy which might guide us
toward a proper interpretation of [§ 2680(j)].”). We agree
with the D.C. Circuit that the phrase “arising out of”
suggests that this immunity is quite broad. As a result,
the Ninth Circuit’s statement of purpose, limiting the
policy to foreclosing any “duty of reasonable care . . . to
those against whom force is directed,” is too narrow16—
which is well demonstrated by the fact that the plain
language would prevent suits against the military for
harm it causes through friendly fire.
This leaves the D.C. Circuit’s articulation of the
purpose, which we find persuasive in some respects. We

16
In fairness, the Ninth Circuit describes its articulated
purpose as “one purpose,” Koohi, 976 F.2d at 1337,
which means that the court may recognize that there are
other, broader purposes as well.
49
agree that the statute represents a federal policy to
prevent state regulation of the military’s battlefield
conduct and decisions. See Saleh, 580 F.3d at 7
(explaining that § 2680(j) reveals Congress’ intent to
“preempt state or foreign regulation of federal wartime
conduct and to free military commanders from the doubts
and uncertainty inherent in potential subjection to civil
suit”). But we do not go as far as the D.C. Circuit’s
holding that § 2680(j) reveals a policy of “the elimination
of tort from the battlefield.” Id. at 7; see also id. (“The
very purposes of tort law are in conflict with the pursuit
of warfare.”). This broader statement loses sight of the
fact that § 2680(j), as a part of the Federal Tort Claims
Act, does not provide immunity to nongovernmental
actors. So to say that Congress intended to eliminate all
tort law is too much, which the D.C. Circuit itself
implicitly recognizes by crafting a test that does not
preempt state tort claims challenging contractors’
performance of certain kinds of contracts. Id. at 9–10.
The purpose underlying § 2680(j) therefore is to
foreclose state regulation of the military’s battlefield
conduct and decisions. With this policy in mind, we turn
to the last step of the Boyle framework: deriving a test to
decide which state claims are preempted. The D.C.
Circuit articulates one test: “During wartime, where a
private service contractor is integrated into combatant
activities over which the military retains command
authority, a tort claim arising out of the contractor’s
50
engagement in such activities shall be preempted.” Id.
at 9. KBR urges us to adopt the Solicitor General’s two-
part test: (1) “whether a claim against the United States
alleging similar conduct would be within the FTCA’s
exception for combatant activities,” and (2) “whether the
contractor was acting within the scope of its contractual
relationship with the federal government at the time of
the incident out of which the claim arose.” Br. for the
United States as Amicus Curiae, Al Shimari v. CACI
Int’l, Inc., Nos. 09-1335, 10-1891, 10-1921, at 17–19
(4th Cir. Jan. 14, 2012)).
We adopt the D.C. Circuit’s combatant-activities,
command-authority test because it best suits the purpose
of § 2680(j). The Solicitor General’s test is overinclusive.
The latter test, by preempting combatant-activity-related
contractor conduct so long as the conduct is within the
“scope of [the contractor’s] contractual relationship,”
would insulate contractors from liability even when their
conduct does not result from military decisions or orders.
The Solicitor General makes this clear by explaining that
under his approach, “federal preemption would generally
apply even if an employee of a contractor allegedly
violated the terms of the contract . . . as long as the
alleged conduct at issue was within the scope of the
contractual relationship.” Id. at 20 (defining scope by
analogy to the Westfall Act and Barr v. Matteo, 360 U.S.
564 (1959) (plurality opinion)). A scope of preemption
that includes contractors’ contractual violations is too
51
broad to fit § 2680(j)’s purpose because the conduct
underlying these violations is necessarily made
independently of the military’s battlefield conduct and
decisions. After all, if the contractors’ conduct did follow
from the military’s decisions or orders, then the conduct
would presumably not be in violation of the contract.
State regulation of these violations thus does not
constitute the regulation of the military’s battlefield
conduct or decisions that § 2680(j) is meant to prevent.
The combatant-activities, command-authority test,
in contrast, is well-tailored to the purpose underlying
§ 2680(j): The first prong—whether the contractor is
integrated into the military’s combatant activities—
ensures that preemption occurs only when battlefield
decisions are at issue. And the second prong—whether
the contractor’s actions were the result of the military’s
retention of command authority—properly differentiates
between the need to insulate the military’s battlefield
decisions from state regulation and the permissible
regulation of harm resulting solely from contractors’
actions.
Under the combatant-activities, command-
authority test we adopt, the plaintiffs’ claims are not
preempted. As to the combatant-activities prong, KBR’s
maintenance of electrical systems at a barracks in an
active war zone qualifies as integration into the military’s
combatant activities. The plaintiffs contend otherwise,
arguing that this maintenance is not a combatant activity
52
because it does not include actual combat such that it
“arises from combatant activities of the military or naval
forces.” This takes too narrow a view of the phrase
“combatant activities.” As the Ninth Circuit explained,
combatant activities “include not only physical violence,
but activities both necessary to and in direct connection
with actual hostilities.” Johnson, 170 F.2d at 770. As an
example, the Court explained that “[t]he act of supplying
ammunition to fighting vessels in a combat area during
war is undoubtedly a ‘combatant activity.’” Id.
Maintaining the electrical systems for a barracks in an
active war zone is analogous to supplying ammunition to
fighting vessels in a combat area and is certainly
“necessary to and in direct connection” to the hostilities
engaged in by the troops living in those barracks. The
plaintiffs’ argument is thus unpersuasive and the first
prong of the test is satisfied.

This case is ultimately not preempted, however,
because the second prong is not satisfied. The military
did not retain command authority over KBR’s installation
and maintenance of the pump because, as explained
above, the relevant contracts and work orders did not
prescribe how KBR was to perform the work required of
it. Instead, the contracts and the work orders provided for
general requirements or objectives and then gave KBR
considerable discretion in deciding how to satisfy them.
See supra text accompanying note 4. As the D.C. Circuit
explained, these types of contracts are “performance-
53
based” contracts that “‘describe the work in terms of the
required results rather than either “how” the work is to be
accomplished or the number of hours to be provided.’”
Saleh, 580 F.3d at 10 (quoting 48 C.F.R. § 37.602(b)(1)).
“[B]y definition, the military [cannot] retain command
authority nor operational control over contractors
working on [this] basis and thus tort suits against such
contractors [are] not [ ] preempted” under the combatant-
activities, command-authority test. Id. The considerable
discretion KBR had in deciding how to complete the
maintenance at issue here thus prevents the plaintiffs’
suit from being preempted because the military did not
retain command authority over KBR’s actions.
III
We will remand to the District Court for
proceedings consistent with this opinion. The plaintiffs’
claims are not preempted by the combatant-activities
exception, and it is possible that those claims are not
foreclosed by the political-question doctrine. To decide
the latter issue, the District Court will first need to decide
which state’s law applies. If Pennsylvania law applies,
then this case lacks any nonjusticiable issues and may
proceed. But if either Tennessee or Texas law applies,
then the case contains nonjusticiable issues. At the least,
in that situation, the District Court will need to eliminate
any damages that are based on proportional liability but
allow the case to move forward to provide such other
remedies as may exist. At most, the case will be
54
dismissed if Staff Sergeant Maseth is first found
contributorily negligent.

55