Rhonda Kemper v. Deutsche Bank Ag

In the
United States Court of Appeals
For the Seventh Circuit
____________________
No. 18-1031
R HONDA K EMPER ,
Plaintiff-Appellant,
v.
D EUTSCHE BANK AG,
Defendant-Appellee.
____________________
Appeal from the United States District Court for the
Southern District of Illinois.
No. 16-CV-0497-MJR-SCW — Michael J. Reagan, Chief Judge.
____________________
A RGUED SEPTEMBER 7, 2018 — D ECIDED D ECEMBER 12, 2018
____________________
Before WOOD, Chief Judge, and R OVNER and BRENNAN, Cir-
cuit Judges.
WOOD, Chief Judge. On May 16, 2009, U.S. Army Specialist
David Schaefer, Jr., was killed by a roadside bomb while serv-
ing a tour of duty in Iraq. This litigation represents his
mother’s attempt to hold someone responsible for the sense-
less loss of her son. It is easy to understand why she wishes to
do so. But not everything is redressable in a court. Terrorist
attacks such as the one that took Spc. Schaefer’s life often

-- 1 of 23 --

2 No. 18-1031
elude the conventional judicial system. Those directly respon-
sible may be beyond the reach of the court, because they are
unidentifiable, or because they are beyond the reach of the
court’s personal jurisdiction, or because they themselves have
come to a violent end. Secondary actors, such as the organiza-
tions that fund the terrorists, are often amorphous and diffi-
cult to hale into court. Finally, despite Congress’s effort to
make state sponsors of terrorism accountable in U.S. courts,
see 28 U.S.C. § 1605A, any resulting judgment may be uncol-
lectible. See, e.g., Rubin v. Islamic Republic of Iran, 138 S. Ct. 816
(2018).
Rhonda Kemper attempted to get around these formida-
ble obstacles by alleging that the bomb that killed her son was
a signature Iranian weapon that traveled from the Iranian
Revolutionary Guard Corps (“the Guard”) to Hezbollah to
Iraqi militias, who then placed it in the ground in Basra, Iraq,
where it killed Spc. Schaefer. Kemper asserts that Deutsche
Bank AG, a German entity with U.S. affiliates, is responsible
for her son’s death under the Anti-Terrorism Act (“ATA”), 18
U.S.C. § 2333. She ties Deutsche Bank to the fatal bomb
through the Bank’s alleged membership in an Iranian conspir-
acy to commit acts of terror. It joined that conspiracy, she con-
tends, when it instituted procedures to evade U.S. sanctions
and facilitate Iranian banking transactions.
The district court found that Kemper failed to plead facts
that plausibly indicated that Deutsche Bank’s actions caused
her son’s death. It thus dismissed her complaint for failure to
state a claim. We affirm.

-- 2 of 23 --

No. 18-1031 3
I
A
In presenting the following account of the facts, we take
Kemper’s account as true, understanding that it would be
contestable if this case were to move forward. The May 2009
attack that killed Spc. Schaefer was typical of Iran’s long and
sordid history of supporting terrorism. The United States des-
ignated Iran a State Sponsor of Terrorism in 1984, and that
designation continues to this day. Although Iran plays no of-
ficial role in the ongoing hostilities in Iraq, it maintains a large
presence in that country through proxies in Iraqi Shi’a mili-
tias. One of those proxies planted the bomb that killed Spc.
Schaefer.
The allegation that Iran had a role in Spc. Schaefer’s death
is plausible because of the distinctive explosive used in the
attack. Iran supplies Explosively Formed Penetrators
(“EFPs”) to its Iraqi agents. These EFPs have precisely crafted
copper linings and military-grade explosives that are capable
of piercing American armored vehicles. The Iranian EFPs
leave a distinctive fingerprint on bomb debris. But for Iran,
the technical sophistication and explosive power found in
EFPs would be unavailable to Iraqi militias.
As a result of Iran’s involvement in terrorism, in 1995 the
United States imposed broad-ranging sanctions prohibiting
almost all trade between the two countries. See Exec. Order
No. 12959. Until November 2008, however, there were excep-
tions to the sanctions. The one that interests us allowed Iran
some access to the U.S. financial system through a regulation
known as the U-Turn exemption. 31 C.F.R. § 560.516 (1995),
amended 73 Fed. Reg. 66,541 (Nov. 10, 2008). This exemption

-- 3 of 23 --

4 No. 18-1031
allowed Iranian entities to use a non-Iranian bank with a cor-
respondent account in the United States to process transac-
tions. For example, an Iranian bank could keep an account
with a non-U.S. bank such as Deutsche Bank. Deutsche Bank
could then use its own correspondent account with a U.S.
bank to process the Iranian bank’s transaction through the
United States. The exemption was a practical necessity for
Iran because much of Iran’s economy is oil-based, and the oil
market is conducted in U.S. dollars. Without access to the U.S.
financial system, Iran’s ability to use its oil-based earnings
would be severely restricted. The U-Turn exemption thus
gave Iran the ability to use its earnings from oil sales for legit-
imate, non-terroristic, purposes. Critical to the U-Turn ex-
emption’s functioning was the transparent identification of
the various counterparties to the transactions conducted pur-
suant to it. This transparency allowed U.S. banks to ensure
that no transactions would benefit any sanctioned entities,
thereby preventing Iran from using any of its assets to pro-
mote its terrorism goals.
Deutsche Bank found a way to subvert this regulatory
scheme. By strategically removing names or otherwise hiding
the existence of potentially sanctioned counterparties to U.S.-
dollar-clearing transactions, Deutsche Bank avoided the
U-Turn exemption’s transparency requirements and thus the
additional regulatory scrutiny called for by the U.S. sanctions.
For example, on some transactions Deutsche Bank employees
would include notes stating “PLS DON’T MENTION THE
NAME OF BANK SADERAT IRAN OR IRAN IN USA,” or
“THE NAME BANK MELLI OR MARKAZI SHOULD NOT
BE MENTIONED … IMPORTANT: NO IRANIAN NAMES
TO BE MENTIONED WHEN MAKING PAYMENT TO NEW
YORK” (capitalization in original). These notes, while far

-- 4 of 23 --

No. 18-1031 5
from subtle, were effective. The names of sanctioned or po-
tentially sanctioned parties were removed from thousands of
transactions that later passed through U.S. banks’ sanctions-
detection systems without triggering any inquiry. For its trou-
bles, Deutsche Bank charged a premium for providing these
sanctions-avoidance services. It put these practices in place
around 1999 and maintained them until at least 2006. Some
illicit transactions persisted even after the bank formally un-
dertook to abolish them.
The use of these non-transparent services was not spo-
radic. Deutsche Bank offered them to Iranian, Libyan, Syrian,
Burmese, and Sudanese financial institutions as well as to
other entities subject to U.S. sanctions. Over seven years,
Deutsche Bank processed more than 27,200 transactions val-
ued at approximately $10.86 billion for these customers. Of
these, Deutsche Bank estimated that approximately 600 trans-
actions valued at over $38 million were illegal under one of
the U.S. sanctions programs. Although the existence of
Deutsche Bank’s sanctions-avoidance program was hidden,
Deutsche Bank considered it an important part of its business.
At least one member of the Bank’s management board was
kept aware of its functioning; some non-U.S. employees were
considered experts in the practices and trained other employ-
ees; and the Bank issued formal and informal written instruc-
tions and training manuals for dealing with sanctioned cus-
tomers.
In November 2015, the discovery of these practices led to
Deutsche Bank’s agreeing to a consent order with the New
York State Department of Financial Services. This consent or-
der required Deutsche Bank to pay a $258 million civil pen-
alty, to agree to an independent monitor’s oversight, and to

-- 5 of 23 --

6 No. 18-1031
fire numerous employees. The consent order contains an ex-
haustive factual description of how Deutsche Bank avoided
U.S. sanctions; it is the source of many of the facts that appear
in Kemper’s complaint. But the consent order does not state
or suggest that any of the transactions Deutsche Bank pro-
cessed were used to fund terrorism. And neither the consent
order nor Kemper’s complaint specifies what percentage of
Deutsche Bank’s sanctions-avoidance business came from Ira-
nian institutions, as opposed to its other sanctioned custom-
ers.
B
Kemper asserts that Deutsche Bank’s sanctions-avoidance
business was part of a much larger ongoing conspiracy to fur-
ther Iran’s terroristic goals. The alleged co-conspirators in-
cluded the following:
 Iran;
 the Guard;
 Iranian banks, including Bank Saderat, Bank Melli,
and the Central Bank of Iran;
 the Islamic Republic of Iran Shipping Lines (“Ship-
ping Lines”), Iran’s national maritime carrier,
which in 2008 the United States labeled a Specially
Designated National for its role in perpetuating
Iran’s terrorism;
 various Western financial institutions, including
Deutsche Bank.
These co-conspirators worked to funnel millions of dollars to
Hezbollah, a U.S.-designated Foreign Terrorist Organization,
and enabled certain co-conspirators directly to perpetrate

-- 6 of 23 --

No. 18-1031 7
homicides, bombings, and other acts of international terror-
ism.
Plaintiffs allege that Deutsche Bank was actively involved
in this conspiracy from 1999 until 2011 (well after the formal
abandonment of the evasive measures). The Bank’s role was
to use the sanctions-avoiding practices described above to fa-
cilitate Iran’s and its co-conspirators’ access to funds. In turn,
the Iranian banks would provide financial services to the
Guard, Hezbollah, and others who facilitated Iran’s terror-
ism-related goals. Deutsche Bank also facilitated payments to-
taling more than $60 million to Shipping Lines, and those pay-
ments enabled Shipping Lines to assist in terrorist attacks.
Deutsche Bank allegedly knew of or was deliberately in-
different to the terroristic aims of this conspiracy. Popular
news sources such as the BBC and CNN repeatedly reported
on Iran’s role in the killing of American troops in Iraq as far
back as 2006. Furthermore, the Bank knew that the financial
oversight systems imposed as part of the U.S. sanctions on
Iran relied on financial institutions as the first line of defense
to prevent Iran from using the U.S. financial system for terror-
ist funding.
C
Deutsche Bank responded to Kemper’s complaint with a
motion to dismiss pursuant to Federal Rule of Civil Procedure
12(b)(6) for failure to state a claim upon which relief could be
granted. The district court held that Kemper failed to allege
facts showing that Deutsche Bank proximately caused her
son’s death in Iraq, and that this was a dispositive omission.
It found that there were too many steps in the hypothesized
causal chain between Deutsche Bank’s actions and the

-- 7 of 23 --

8 No. 18-1031
terrorists that attacked Spc. Schaefer to support liability. It
also found that, to the extent Deutsche Bank joined any con-
spiracy, it joined only a conspiracy to avoid sanctions. The lat-
ter conspiracy was distinct from any of Iran’s terrorism-re-
lated goals. For that reason, the complaint did not satisfy the
ATA’s requirement that to establish liability, the pleaded con-
spiracy must be one to provide material support for terrorism.
The district court thus dismissed Kemper’s complaint, and
this timely appeal followed.
We consider the district court’s dismissal of Kemper’s
complaint de novo, assuming that all of her well-pleaded alle-
gations are true and construing reasonable inferences in her
favor. Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009); Bell Atlantic
Corp. v. Twombly, 550 U.S. 544 (2007); see also Manistee Apart-
ments, LLC v. City of Chicago, 844 F.3d 630, 633 (7th Cir. 2016).
II
In order to state a claim under the ATA, Kemper must
plead facts showing that a United States national was injured
“by reason of an act of international terrorism.” 18 U.S.C.
§ 2333(a). At first glance, her complaint seems to meet that re-
quirement. Neither party disputes that the bombing that
killed Spc. Schaefer was an act of international terrorism, and
Schaefer and Kemper are both United States nationals who
were injured by reason of that act. But the statute does not
specify the set of possible defendants it contemplates. We can
presume that Kemper could sue the person who placed the
bomb, if she could find that person in the United States. But
what about Deutsche Bank? Is the Bank sufficiently connected
to that act of international terrorism to hold it liable for Kem-
per’s injuries? If not, then do any of Deutsche Bank’s other
actions fit the statutory definition of international terrorism?

-- 8 of 23 --

No. 18-1031 9
The ATA defines the term “international terrorism”
through three requirements. First, the act must “involve vio-
lent acts or acts dangerous to human life” that violate U.S. or
state criminal laws. 18 U.S.C. § 2331(1)(A). Kemper alleges
that Deutsche Bank violated the federal criminal prohibitions
on providing material support to terrorism and conspiring to
provide material support. See 18 U.S.C. §§ 2339A, 2339B. Sec-
ond, the act must “appear to be intended to intimidate or co-
erce a civilian population; to influence the policy of a govern-
ment by intimidation or coercion; or to affect the conduct of a
government by mass destruction, assassination, or kidnap-
ping.” 18 U.S.C. § 2331(1)(B). And third, the act must “occur
primarily outside the territorial jurisdiction of the United
States, or transcend national boundaries … .” 18 U.S.C.
§ 2331(1)(C).
In addition to these statutory criteria, we have held that
the ordinary tort requirements of “fault, state of mind, causa-
tion, and foreseeability must be satisfied” in an ATA action.
Boim v. Holy Land Found. For Relief & Dev., 549 F.3d 685, 692
(7th Cir. 2008) (en banc) (hereinafter “Boim III”). Finally, be-
cause the ATA provides for treble damages and cost shifting,
18 U.S.C. § 2333(a), a plaintiff must prove intentional miscon-
duct by the defendant. Boim III, 549 F.3d at 692–93. Although
the district court focused on Kemper’s failure to plead causa-
tion adequately, we begin with the question whether
Deutsche Bank’s actions met the definition of international
terrorism, and then move to causation.
A
Kemper’s complaint fails plausibly to allege that Deutsche
Bank’s actions satisfy the statutory definition of international
terrorism and thus fall within the ambit of the ATA. The

-- 9 of 23 --

10 No. 18-1031
Bank’s conduct was not “violent” or “dangerous to human
life” as 18 U.S.C. § 2331(1)(A) requires, nor did it display the
terroristic intent required by 18 U.S.C. § 2331(1)(B).
The facts in our case are different from those in Boim III,
which dealt with direct donations to a known terrorist organ-
ization. While giving fungible dollars to a terrorist organiza-
tion may be “dangerous to human life,” doing business with
companies and countries that have significant legitimate op-
erations is not necessarily so. That these business dealings
may violate U.S. sanctions does not convert them into terrorist
acts. This is especially so when, as here, the great majority of
these business interactions did not actually violate any sanc-
tions, and the sanctions at issue cover more than terrorism-
related transactions.
Deutsche Bank’s actions also do not appear intended to in-
timidate or coerce any civilian population or government. 18
U.S.C. § 2331(1)(B)(i), (ii). To the objective observer, its inter-
actions with Iranian entities were motivated by economics,
not by a desire to “intimidate or coerce.” This view is bol-
stered by the consent decree between Deutsche Bank and the
New York State Department of Financial Services. While that
decree describes Deutsche Bank’s many wrongful actions
taken in attempts to evade the U.S. sanctions apparatus, it
never once mentions terrorism. Indeed, that decree repeat-
edly states that Deutsche Bank built its sanctions-evading
business because it was “lucrative.”
B
Even if we were to overlook the fundamental question
whether Deutsche Bank’s actions count as “international ter-
rorism,” Kemper would still need to show that her allegations

-- 10 of 23 --

No. 18-1031 11
would support civil liability against the Bank. Terrorism is an
emotionally and politically charged issue, but the ATA ulti-
mately is a tort statute. That means that causation must be
proven before liability is established. Traditionally, the causa-
tion inquiry is broken down into two subparts: but-for or fac-
tual causation, i.e., whether the harm could have occurred
without the alleged tortious act; and proximate or legal cause,
which attempts to determine the scope of liability for a de-
fendant’s actions. See, e.g., Bridge v. Phoenix Bond & Indem. Co.,
553 U.S. 639, 653–54 (2008).
We have recognized that strict but-for causation is not nec-
essary to prove ATA liability. See Boim III, 549 F.3d at 695–99.
In Boim III, we found that providing material support to ter-
rorists is a wrongful act that, like spilling toxic waste into
groundwater, creates liability regardless of literal but-for cau-
sation because it is done in the context of others committing
similar wrongful acts. Id. at 696–97; see also Restatement
(Third) of the Law of Torts: Liability for Phys. & Emot. Harm
§ 27, cmts. f, g (2010) (“Restatement (Third) of Torts”). We
held that making a monetary donation to Hamas, knowing
that organization’s active participation in terrorism, fits into
this exception to the general rule of but-for causation. 549 F.3d
at 697–99. Even if Hamas would have committed the terrorist
attack at issue without the Boim III defendants’ donations, the
funders could still be held liable because they added to Ha-
mas’s ability to engage in terrorism. Id. Deutsche Bank argues
that Boim III is no longer good law on this point because later
Supreme Court decisions have held that when a statute such
as the ATA uses the phrase “by reason of,” liability under that
statute requires but-for causation. See Burrage v. United States,
571 U.S. 204, 213 (2014) (Controlled Substances Act). We need
not address this issue, however, because the failure of

-- 11 of 23 --

12 No. 18-1031
Kemper’s complaint plausibly to allege proximate causation
independently requires dismissal.
We first take this opportunity to clarify some language in
Boim III that might be read to suggest that something less than
proximate cause might suffice to prove ATA liability. See, e.g.,
549 F.3d at 698, 700 (noting that “terrorism is sui generis” and
that ATA liability might attach to “someone who … contrib-
uted to a terrorist organization in 1995 that killed an Ameri-
can abroad in 2045”). Such a reading would not be compatible
with the opinion as a whole. Boim III expressly held that “the
ordinary tort requirements relating to … causation … must be
satisfied” to establish ATA liability. Id. at 692. Indeed, the core
of the logic underlying Boim III comes directly from the ca-
nonical case of Summers v. Tice, 33 Cal. 2d 80 (1948), cited at
549 F.3d at 696. The analogy between multiple sources
providing material support to terrorists, and multiple hunters
firing in the same direction, while imperfect, is helpful. In
both cases there is an undoubtedly wrongful act. And in both
cases a rigid insistence on singular but-for causation would
perversely result in the tortfeasor’s escaping liability not be-
cause of her own actions, but because she was lucky enough
that other people were committing similarly wrongful acts at
the same time. The causation inquiry was not meant to pre-
vent liability in these situations.
Kemper does not argue that the ATA is devoid of a prox-
imate-cause requirement, and for good reason. The Supreme
Court has told us that “[i]t is a well established principle of
[the common] law that in all cases of loss, we are to attribute
it to the proximate cause, and not to any remote cause. … [The
Court] assume[s] Congress is familiar with the common-law
rule and does not mean to displace it sub silentio in federal

-- 12 of 23 --

No. 18-1031 13
causes of action.” Bank of Am. Corp v. City of Miami, Fla., 137 S.
Ct. 1296, 1305 (2017) (internal citations and quotations omit-
ted). Furthermore, the Supreme Court has repeatedly and ex-
plicitly held that when Congress uses the phrase “by reason
of” in a statute, it intends to require a showing of proximate
cause. See Holmes v. Sec. Investor Prot. Corp., 503 U.S. 258, 267–
68 (1992) (“by reason of” in RICO statute requires proximate
cause); Assoc. Gen. Contractors of Cal., Inc. v. Cal. State Council
of Carpenters, 459 U.S. 519, 532–35 (1983) (“by reason of” in
section four of the Clayton Act requires proximate cause); see
also Hemi Group, LLC v. City of New York, N.Y., 559 U.S. 1
(2010). Three of our sister circuits have expressly extended
this reasoning to the ATA. See Owens v. BNP Paribas, S.A., 897
F.3d 266, 273 (D.C. Cir. 2018); Fields v. Twitter, Inc., 881 F.3d
739, 744–45 (9th Cir. 2018); Rothstein v. UBS AG, 708 F.3d 82,
95–96 (2d Cir. 2013). We see no reason to depart from this
logic, and so we join them in holding that proximate cause is
necessary for ATA liability.
That said, we know that simply stating that the ATA re-
quires proof of “proximate cause” does not help much. A firm
definition for the term “proximate cause” has escaped judges,
lawyers, and legal scholars for centuries. See, e.g., Waters v.
Merchants’ Louisville Ins. Co., 36 U.S. 213, 221 (1837) (attempt-
ing to determine the proximate cause of the destruction of a
steamboat). The term has so bedeviled the authors of the Re-
statements of Torts that they omitted it entirely from the First
and Second Restatements and expressed a wish for its exter-
mination in the Third. See Restatement (Third) of Torts, Spec.
Note on Proximate Cause (2010).
The parties have jumped into the fray and, to no one’s sur-
prise, have reached different conclusions. Drawing on

-- 13 of 23 --

14 No. 18-1031
conspiracy law, Kemper directs us to the foreseeability of an
outcome, see Halberstam v. Welch, 705 F.2d 472 (D.C. Cir. 1983),
while Deutsche Bank implores us to focus on the number of
steps that occur between an action and its consequence. See
Nichols v. Mich. City Plant Planning Dep’t, 755 F.3d 594, 604 (7th
Cir. 2014) (“Proximate cause requires only some direct rela-
tion between injury asserted and injurious conduct alleged,
and excludes only those links that are too remote, purely con-
tingent, or indirect.” (quoting Staub v. Proctor Hosp., 562 U.S.
411, 419 (2011))). This is, to a degree, a pointless debate: di-
rectness and foreseeability are logically linked. In most cases
the more directly related an outcome is to an underlying ac-
tion, the more likely that the outcome will have been foresee-
able, and vice versa. In addition to directness and foreseeabil-
ity, other courts attempting to define proximate cause under
the ATA have also considered whether a defendant’s actions
were “a substantial factor in the sequence of events” leading
to a plaintiff’s injuries. Owens, 897 F.3d at 273 (quoting Roth-
stein, 708 F.3d at 91) (internal quotations omitted).
Fortunately, we do not need to resolve once and for all the
centuries-old debate about what exactly proximate cause re-
quires. It is enough to note that all of the factors identified by
the parties and our sister circuits—foreseeability, directness,
and the substantiality of the defendant’s conduct—are rele-
vant to the inquiry. Moreover, as Boim III displays, other es-
tablished principles of tort causation will often be useful as
well. This catch-all approach has been endorsed by the Su-
preme Court and is a recognition of the fact that “we use
‘proximate cause’ to label generically the judicial tools used
to” perform an inquiry that ultimately “‘reflects ideas of what
justice demands, or of what is administratively possible and
convenient.’” Holmes, 503 U.S. at 268 (quoting W. Keeton, et

-- 14 of 23 --

No. 18-1031 15
al., P ROSSER AND KEETON ON LAW OF T ORTS § 41, p. 264 (5th ed.
1984)); see also Bridge, 553 U.S. at 654 (“Proximate cause … is
a flexible concept that does not lend itself to a black-letter rule
that will dictate the result in every case.” (internal quotation
marks omitted)).
We can now return to the facts of this case. Kemper alleges
that Deutsche Bank provided financial services to Iranian fi-
nancial institutions and to other Iranian businesses such as
Shipping Lines, and those Iranian entities went on to provide
services to terrorist groups. Kemper does not allege that
Deutsche Bank ever serviced a terrorist group directly. Nor
does she allege that these various Iranian institutions exist
solely to perform terrorist acts. Indeed, although Kemper al-
leges that Shipping Lines transported weapons for Iran, she
also specifically asserts that it “provides a variety of maritime
transport services” and is “a global operator with a world-
wide network of subsidiaries, branch offices and agent rela-
tionships.” The Iranian financial institutions similarly have
branches and subsidiaries spread across Europe and Asia.
These facts take Kemper’s allegations out of the purview of
Boim III. Even if Hamas’s non-terroristic endeavors were
closely tied to its terroristic mission, it strains credulity to sug-
gest that a worldwide network of banks or shipping lines is
similarly dedicated to terrorism.
There is an additional hurdle that Kemper has not over-
come: the Islamic Republic of Iran. The central theory tying
all of Kemper’s allegations together is that Iran is responsible
for the terrorist attack that killed her son, and that Deutsche
Bank facilitated that attack by providing services to Iran and
state-owned Iranian businesses. But this theory ignores that,
as the D.C. and Second Circuits have noted, Iran “is a

-- 15 of 23 --

16 No. 18-1031
sovereign state with ‘many legitimate agencies, operations,
and programs to fund.’” Owens, 897 F.3d at 276 (quoting Roth-
stein, 708 F.3d at 97). When one of the links on a causal chain
is a sovereign state, the need for facts specifically connecting
a defendant’s actions to the ultimate terrorist attack is espe-
cially acute. See id. That Iran is involved in substantial non-
terrorist activities is clear from Kemper’s complaint. Kemper
alleges that Deutsche Bank facilitated or conspired to facilitate
Iran’s providing approximately $210 million of transactions
for various sanctioned entities and terrorist groups over sev-
eral years. By contrast, Kemper states that over seven years,
Iran’s oil and gas revenues totaled $972.9 billion. Even accord-
ing to Kemper, then, Iran had over $972 billion in revenues
that it applied to non-terroristic undertakings.
Kemper’s complaint also fails to suggest how her theory
might overcome the traditional tort doctrine of superseding
or intervening cause. A cause is superseding when “it is a
‘cause of independent origin that was not foreseeable.’” Staub,
562 U.S. at 420 (quoting Exxon Co., U.S.A. v. Sofec, Inc., 517 U.S.
830, 837 (1996)). We have recognized, consistently with the
Restatement (Third) of Torts, that although “criminal acts are
not superseding causes per se[,] … acts that are either criminal
or intentionally tortious … are more likely to be adjudged su-
perseding causes.” Scottsdale Ins. Co. v. Subscription Plus, Inc.,
299 F.3d 618, 620 (2002); see also Hibma v. Odegaard, 769 F.2d
1147, 1155–56 (7th Cir. 1985); Restatement (Third) of Torts §
34, cmt. e. Here, there are numerous criminal intervening acts
separating Deutsche Bank from the terrorist attack that killed
Kemper’s son. After Deutsche Bank did business with Iranian
institutions, all of the following actors had to commit an inde-
pendent criminal act before the offending EFP was set in
Basra: the Iraqi militias, Hezbollah, the Guard, the Iranian

-- 16 of 23 --

No. 18-1031 17
financial and shipping institutions, and most importantly, the
Islamic Republic itself.
Although Owens and Rothstein do not use the language of
superseding cause, both recognize that a sovereign’s affirma-
tive choice to engage in a wrongful act will usually supersede
a third party’s choice to do business with that sovereign. See
Owens, 897 F.3d at 276; Rothstein, 708 F.3d at 97. This makes
sense. Finding that a sovereign state’s actions supersede other
more tangential causes prevents the scope of liability from ex-
tending to the many individual persons, businesses, and other
sovereigns that interact with that state. For example, the
United Kingdom has authorized both Bank Saderat and Ship-
ping Lines to operate subsidiaries in its territory. The exist-
ence of those subsidiaries surely furthers Iran’s ability to en-
gage in terrorism, but we cannot imagine that U.K. regulators
could be held responsible for Iran’s terrorist acts. Similarly,
purchasers of Iranian oil and natural gas contribute funds to
Iran that Iran might use to support terrorism, but those pur-
chasers are not liable for the attacks that Iran may facilitate
with those funds.
Kemper counters these arguments with three of her own.
First, she reminds us that Iran is a designated state sponsor of
terrorism. Second, she finds it suspicious that the Iranian en-
tities used Deutsche Bank’s non-transparent services even
though the U-Turn exemption was available for any legiti-
mate business. And third, she reasons that the Iranian sanc-
tions are anti-terrorism measures and so any attempt to sub-
vert them necessarily furthers Iran’s terrorism capabilities. All
of that may be true, but we do not see how these points suffi-
ciently connect Deutsche Bank to terrorism supported by the
Iranian state. The United States has regularly differentiated

-- 17 of 23 --

18 No. 18-1031
between providing support to state sponsors of terror and
providing support to terrorist organizations. See, e.g., Owens,
897 F.3d at 276 (noting that Congress found a total prohibition
on financial support to terrorist organizations justified while
explicitly permitting some financial transactions with state
sponsors of terrorism). Indeed, the ATA itself makes this dis-
tinction. See 18 U.S.C. § 2333(d)(2) (limiting liability to injuries
caused by “an organization that had been designated as a for-
eign terrorist organization”).
The Iranian institutions’ decision to avail themselves of
Deutsche Bank’s sanctions-avoiding services similarly does
not sufficiently connect Deutsche Bank to terrorism. As we al-
ready have noted, most of the transactions processed by
Deutsche Bank did not violate any U.S. sanctions (though
they did violate processing rules). Beyond this, Deutsche
Bank provided its services to sanctioned entities, such as
those from Burma, whose disfavored status is unrelated to
terrorism. These facts weaken the inference that Deutsche
Bank’s provision of sanctions-avoiding services was neces-
sarily tied to terrorism. Finally, although the Iran sanctions
may have begun because of Iran’s support for terrorism, the
actual sanctions extend far beyond commerce that could sup-
port terroristic endeavors. Instead, the sanctions prohibit vir-
tually all trade between the United States and Iran, regardless
of whether that trade could conceivably facilitate terrorism.
Even if Deutsche Bank had facilitated Iran’s purchase of a
crate of oranges, that deal could violate U.S. sanctions. With-
out some other fact suggesting either an intent to support ter-
rorism or a direct provision of services to terrorists, the viola-
tion of such a broad prohibition does not create a sufficient
link to establish liability for terrorism-related torts under any
traditional notion of proximate cause. The implication from

-- 18 of 23 --

No. 18-1031 19
these facts is that Deutsche Bank’s sanctions-avoiding actions,
while wrongful, were designed to increase its profits by
providing a service to willing customers at premium prices.
In other words, Kemper has not alleged facts plausibly sug-
gesting that it was foreseeable that Deutsche Bank’s actions
would fund terrorism, or that it was a substantial factor in her
son’s death. And she has not alleged any facts directly con-
necting Deutsche Bank’s business dealings, as shady as those
dealings might have been, to Iranian terrorist acts.
III
Unlike previous ATA plaintiffs, who have focused on aid-
ing and abetting theories of liability, Kemper alleges that
Deutsche Bank joined a conspiracy with the Iranian financial
institutions, Shipping Lines, Iran, and the various terrorist or-
ganizations that committed the attack that killed her son. She
argues that terrorist attacks, including the one that killed Spc.
Schaefer, were reasonably foreseeable acts done pursuant to
or in furtherance of this conspiracy, and so Deutsche Bank is
liable for those acts. See Pinkerton v. United States, 328 U.S. 640
(1946) (co-conspirators liable for crimes committed in further-
ance of conspiracy); Halberstam, 705 F.2d at 487 (co-conspira-
tors in civil conspiracy liable for acts that are “a reasonably
foreseeable consequence of the scheme”). But Kemper’s con-
spiracy allegations are factually implausible, and her legal
theory underlying those allegations is problematic.
The crux of any conspiracy is an agreement between the
co-conspirators. Ocasio v. United States, 136 S. Ct. 1423, 1429
(2016). Although “a conspirator [need] not agree to commit or
facilitate each and every part of the offense,” she must “reach
an agreement with the specific intent that” the conspiratorial
goal be completed. Id. (alteration in original) (emphasis

-- 19 of 23 --

20 No. 18-1031
added) (internal quotation marks omitted). Here, Kemper has
not alleged facts that give rise to a plausible inference that
Deutsche Bank agreed to provide material support for terror-
ism. Cf. Bell Atlantic Corp. v. Twombly, 550 U.S. 544, 556–57
(2007). None of the allegations suggest that Deutsche Bank
cared how its Iranian customers obtained or spent the funds
that it processed for them. “A person who is indifferent to the
goals of an ongoing conspiracy does not become a party to
this conspiracy merely because that person knows that his or
her actions might somehow be furthering that conspiracy.”
United States v. Collins, 966 F.2d 1214, 1219–20 (7th Cir. 1992).
This principle—that one cannot join a conspiracy through ap-
athy—is especially important in business dealings. Because
“[b]y definition market transactions—whether in legal or ille-
gal markets—benefit both parties, [ ] we do not assume, ab in-
itio, that they carry with them the excess baggage of conspir-
acy.” United States v. Townsend, 924 F.2d 1385, 1392 (7th Cir.
1991). The facts here suggest only that Deutsche Bank may
have engaged in business dealings that incidentally assisted a
separate terrorism-related conspiracy involving Iran; they do
not suggest that Deutsche Bank ever agreed to join that con-
spiracy.
We thus agree with the district court’s reading of the com-
plaint to allege, at most, that Deutsche Bank joined a conspir-
acy to evade sanctions. Deutsche Bank and the Iranian finan-
cial institutions agreed to omit any reference to Iranian coun-
terparties so as to avoid detection by U.S. banks and the pos-
sible sanctions that detection might trigger. But no facts sug-
gest that Deutsche Bank agreed to facilitate any wrongful con-
duct beyond this.

-- 20 of 23 --

No. 18-1031 21
Beyond this fundamental factual defect, the ATA’s text
creates two other hurdles for Kemper’s conspiracy allega-
tions. If Kemper’s conspiracy liability theory could succeed
on these facts, it would undermine the causation analysis re-
quired by the ATA’s use of the phrase “by reason of.” Kem-
per’s continuous invocation of conspiracy law in her discus-
sions of causation makes this fact obvious. Because co-con-
spirators are liable for each other’s acts done in furtherance of
the conspiracy, pleading the type of wide-ranging conspiracy
that Kemper has would allow plaintiffs to hold defendants li-
able who were not the proximate cause of the plaintiff’s in-
jury. This is not to say that a plaintiff could never adequately
plead a plausible ATA conspiracy. But the further down the
causal chain a defendant sits, the more diligent a plaintiff
must be to plead facts that plausibly suggest that the defend-
ant entered into an agreement to support terrorism.
The second problem with Kemper’s theory of conspiracy
liability is that she premises it on the primary liability found
in 18 U.S.C. § 2333(a). We held in Boim III that even though
section 2333(a) does not encompass secondary theories such
as conspiracy liability, through a chain of incorporations Con-
gress effectively created “[p]rimary liability … [that] has the
character of secondary liability.” 549 F.3d at 691. Conspiracy
and other secondary liability principles could thus inform an
analysis of section 2333(a) primary liability. Id. at 691–92. But
after Boim III was decided, Congress passed the Justice
Against Sponsors of Terrorism Act, Pub. L. No. 114-222, § 4,
130 Stat. 852, 854 (2016), which explicitly added a textual basis
for secondary liability to the ATA. See 18 U.S.C. § 2333(d). The
amendment created liability “for an injury arising from an act
of international terrorism committed, planned, or authorized
by an organization that had been designated as a foreign terrorist

-- 21 of 23 --

22 No. 18-1031
organization … as to any person … who conspires with the
person who committed such an act of international terror-
ism.” Id. (emphasis added). The district court held that section
2333(d)’s requirement that the terrorist attack be committed
by a designated foreign terrorist organization excluded liabil-
ity for attacks committed by a state sponsor of terrorism, i.e.,
Iran.
The district court thus concluded that because Kemper al-
leged the attack at issue here was orchestrated by Iran, her
claim could not be sustained under section 2333(d). Kemper
has not contested this holding on appeal, and so any argu-
ment she may have regarding it is waived. Bernard v. Sessions,
881 F.3d 1042, 1048 (7th Cir. 2018). We note, however, that as-
suming the district court’s interpretation of the statute is cor-
rect, Kemper’s theory of conspiracy liability under section
2333(a) would render section 2333(d)’s express allowance for
conspiracy liability superfluous. If section 2333(a)’s allowance
for primary liability with the “character” of conspiracy liabil-
ity can support a conspiracy including a state sponsor of ter-
ror, a designated terrorist organization, or anyone else (per-
haps a lone wolf), then litigants would never need to use the
more limited conspiracy liability authorized by section
2333(d). That result is one we must avoid. See Marx v. Gen.
Revenue Corp., 568 U.S. 371, 392–93 (2013) (“[A] statute should
be constructed so that effect is given to all of its provisions, so
that no part will be inoperative or superfluous, void or insig-
nificant.” (quoting Corley v. United States, 556 U.S. 303, 314
(2009)) (internal quotations omitted)).
IV
The events leading to this litigation are heartbreaking. The
United States lost a soldier, and Rhonda Kemper lost her son.

-- 22 of 23 --

No. 18-1031 23
The desire to hold someone responsible for this grievous loss
is eminently understandable. But Deutsche Bank is the wrong
defendant. Its actions are too attenuated from the terrorist at-
tack that killed Spc. Schaefer to hold it liable for his death. We
A FFIRM the district court’s judgment.

-- 23 of 23 --