***DEATH PENALTY CASE***
Executions Scheduled for April 17, 20, 24, and 27, 2017

UNITED STATES COURT OF APPEALS FOR THE EIGHTH CIRCUIT
______________________________________________________________________________________________________________________

No. 17-1804
______________________________________________________________________________________________________________________

JASON McGEHEE, et al.,
Plaintiffs-Appellees,
v.
ASA HUTCHINSON, et al.,
Defendants-Appellants.
______________________________________________________________________________________________________________________________________________________________________________________________________________________________

On Appeal from the United States District Court for the Eastern District of
Arkansas (Hon. Kristine Baker)
________________________________________________________________________________________________________________________________________________________________________________________________________________________________

APPELLANTS’ EMERGENCY MOTION TO VACATE STAY OF EXECUTION /
PRELIMINARY INJUNCTION

LESLIE RUTLEDGE
Attorney General
LEE RUDOFSKY
Solicitor General
NICHOLAS J. BRONNI
Deputy Solicitor General
COLIN JORGENSEN
Senior Assistant Attorney General
JENNIFER MERRITT
Assistant Attorney General
OFFICE OF THE ARKANSAS
ATTORNEY GENERAL
323 Center St., Suite 200
Little Rock, AR 72201
(501) 682-2007
lee.rudofsky@arkansasag.gov

 INTRODUCTION
Appellees have a long history of filing and dismissing claims to manipulate
the judicial process and prevent Arkansas from carrying out their executions. This
case—which is identical to Appellees’ multiple previous challenges to Arkansas’s
three-drug (midazolam) lethal-injection protocol—is just the latest iteration.
Moreover, despite repeatedly bringing (and then nonsuiting or losing on the merits)
that same claim in both state and federal court, Appellees nevertheless sought a
preliminary injunction staying their immediate executions so they could relitigate
that same claim.
In granting a preliminary injunction that stayed Appellees’ imminent
executions, the district court ignored those basic facts. It also disregarded the fact
that delaying Appellees’ executions by even a few days—until after Arkansas’s
supply of midazolam expires—will make it impossible for Arkansas to carry out
Appellees’ just and lawful sentences. Immediate reversal is warranted.
BACKGROUND
A. Appellees
Appellees brutally murdered young mothers, children, and men who were
unfortunate enough to cross their paths. Their guilt—and the justice of their
sentences—is beyond dispute.

1

 B. Prior Litigation
1. Arkansas Lethal-Injection Protocol.
In 2015, Arkansas adopted Act 1096, which: 1) codifies the three-drug
lethal-injection protocol upheld in Glossip v. Gross, 135 S. Ct. 2726 (2015), as an
alternative to a single-drug protocol previously upheld; 2) permits the Arkansas
Correction Department to use compounded drugs; and 3) ensures supplier
confidentiality. Thereafter, Arkansas adopted its current three-drug midazolam
protocol.
2. Williams v. Kelley
In April 2015, Appellees filed Williams v. Kelley, a state court action
alleging that Arkansas’s execution protocol violates the Eighth Amendment and
Arkansas’s corresponding provision. After Arkansas removed that complaint to
federal court, Appellees nonsuited. See No. 4:15-CV-206-JM (DE 4). Appellees
then filed amended complaints in state court raising only state-law claims. See
Kelley v. Johnson, 496 S.W.3d 346, 352 (Ark. 2016). Following motion to dismiss
briefing, Appellees nonsuited their claims a second time. (DE 28-17.)1
3. Johnson v. Kelley
While the Williams motion to dismiss was pending, Appellees filed
Johnson v. Kelley, No. 60CV-15-2921, in state court, alleging that Arkansas’s

1

“DE” refers to the docket entry in the district court.
2

 three-drug protocol violated the Arkansas Constitution’s ban on cruel-or-unusual
punishment.
Because Arkansas Governor Asa Hutchinson had set Appellees’ execution
dates for September 2015, Appellees sought a stay pending resolution of their new
case. The Arkansas Supreme Court stayed the executions “pending the resolution
of the litigation.” DE 28-19. Appellees then filed an amended complaint claiming
that Arkansas’s lethal-injection procedure violated Arkansas’s ban on cruel or
unusual punishment because, inter alia, midazolam would not sufficiently
anesthetize and Appellees would experience severe pain. Appellees also alleged
that the use of compounded drugs would impose cruel and unusual punishment and
suggested as alternative execution methods, the firing squad, barbiturates,
anesthetic gas, and opioid overdoses. (DE 28-18.)
Arkansas moved to dismiss, and the state trial court denied that motion.
Both parties also moved for summary judgment on the midazolam claim, and the
trial court denied Arkansas’s motion. The Arkansas Supreme Court reversed “in
toto and dismiss[ed]” Appellees’ claims because under Baze v. Rees, 553 U.S. 35
(2008), and Glossip v. Gross, 135 S. Ct. 2726 (2015), Appellees had failed to meet
their burdens of both pleading and providing evidence that their alternative
execution methods are available and readily implementable. Kelley, 496 S.W.3d at

3

 357-60. The United States Supreme Court declined review. Johnson v. Kelley, 137
S.Ct. 1067 (2017).
Thus, Appellees’ claims were dismissed on the merits, the case ended, and
the stay dissolved. Nevertheless, Appellees attempted to file an amended state
court complaint restating the same claims. The Arkansas Supreme Court responded
by making clear that Appellees’ case had ended. Arkansas then moved to dismiss
Appellees’ amended complaint. The trial court granted that motion and held that
the Arkansas Supreme Court had “dismissed the litigation, with prejudice[.]” (DE
28-23, at 5.)
C. Procedural History
On February 24, 2017, Governor Hutchinson set execution dates for eight of
Appellees. Appellees then launched a legal avalanche, filing cases in multiple
forums, including this case seeking a preliminary injunction.
LEGAL STANDARDS
Execution stays are reviewed “for an abuse of discretion.” Nooner v. Norris,
491 F.3d 804 (2007). A preliminary injunction staying an execution “is an
equitable remedy, and an inmate challenging a state’s lethal injection protocol
through a § 1983 action is not entitled to a stay of execution as a matter of course.”
Id.; Hill v. McDonough, 547 U.S. 573, 584 (2006). Instead, Appellees “must
satisfy all of the requirements for a stay, including a showing of a significant
4

 possibility of success on the merits.” Id.; see also Glossip, 135 S.Ct. at 2736 (“A
plaintiff seeking a preliminary injunction must establish that he is likely to succeed
on the merits.”).
Moreover, in imminent execution cases, “[a] court considering a stay must
also apply ‘a strong equitable presumption against the grant of a stay where a
claim could have been brought at such a time as to allow consideration of the
merits without requiring entry of a stay.” Hill, 547 U.S. at 584 (quoting Nelson v.
Campbell, 541 U.S. 637, 650 (2004)) (emphasis added). Likewise, courts “must be
sensitive to the State’s strong interest in enforcing its criminal judgments without
undue interference from the federal courts.” Hill, 547 U.S. at 584. Reversal is
required because the district court did not apply that standard.
ARGUMENT
I. The district court abused its discretion in concluding that Appellees had
demonstrated a substantial likelihood of success on their midazolam claim.
A. Appellees’ claim is barred by res judicata.
Appellees cannot prevail on their claim that using midazolam as the first
drug in a three-drug protocol constitutes cruel and unusual punishment because
their claim is bared by res judicata. Federal courts cannot adjudicate “claims that
have already been fully adjudicated in state court.” Sparkman Learning Center v.
Ark. Dep’t of Human Servs., 775 F.3d 993, 998 (8th Cir. 2014); see also 28 U.S.C.

5

 1738. State law determines whether a claim has been adjudicated. Sparkman
Learning Ctr., 775 F.3d at 998.
In Arkansas, res judicata applies where the first suit was contested,
jurisdiction existed, the prior case involved the same parties, and there was a final
determination on the merits. Jayel Corp. v. Cochran, 234 S.W.3d 278, 281 (2006);
see also id. (doctrine likewise bars claims that could have been litigated). Applying
that standard, Appellees’ midazolam claim is barred.
First, the Arkansas Supreme Court issued a final judgment resolving the
merits of Appellees’ midazolam claim. Indeed, contrary to the district court’s
conclusion, in Kelley, the Arkansas Supreme Court did not simply dismiss
Appellees’ complaint. Rather, the Arkansas Supreme Court reversed the lower
court’s decision in toto, including its denial of summary judgment on the grounds
that Appellees had failed to substantiate their midazolam claim. See Johnson, 496
S.W.3d at 355-59 (discussing evidence); id. at 359 (no evidence “that the proposed
alternative drugs are available to ADC for use in an execution”). And that
conclusion, as the state trial court later concluded in granting a motion to dismiss
an amended complaint that Appellees filed after the Arkansas Supreme Court’s

6

 decision fully resolved the case on the merits and “dismissed the litigation, with
prejudice[.]” (DE 28-23, 5.)2
Additionally, to the extent that there might be any doubt that Appellees’
claims have been fully resolved on the merits, under Arkansas law, the dismissal of
Appellees’ claim in Kelley operated as a final adjudication on the merits because
that dismissal was actually the third dismissal of Appellees’ midazolam claim. See,
e.g., Ark. R. Civ. P. 41(b) (“dismissal operates as an adjudication on the merits”
where “action has been previously dismissed, whether voluntarily or
involuntarily); Ballard Group, Inc. v. BP Lubricants USA, Inc., 436 S.W.3d 445,
456-57 (Ark. 2014); Brown v. Tucker, 954 S.W.2d 262 (Ark. 1997). Indeed, as
explained above, Appellees previously nonsuited their federal midazolam
challenge in Williams and their subsequent filing of an amended complaint under
that same case heading in state court. (DE 28-17.)
Second, the Arkansas Supreme Court (and the federal district court upon
removal in Williams) had jurisdiction to consider Appellees’ prior midazolam
claims. The district court reached a contrary conclusion by reasoning that the
Arkansas Supreme Court had lacked jurisdiction to resolve Appellees’ midazolam
claim because that court ultimately determined that Appellants were entitled to
sovereign immunity. (DE 53, 32.) Yet that approach ignores the fact that the
2

The district court’s conclusion that “there has been no final judgment” likewise
conflicts with those decisions.
7

 Arkansas Supreme Court had to resolve whether Appellees had provided any
evidence on summary judgment that Arkansas’s use of midazolam violated
Arkansas’s ban on cruel or unusual punishment to determine whether sovereign
immunity applied. Arkansas State Claims Comm’n v. Duit Constr. Co., 445
S.W.3d 496, 502 (Ark. 2014) (no immunity where state officers violate the law).
Third, as summarized above, the prior suits were fully contested in good
faith, including the filing of dismissal motions in, preliminary injunction litigation,
and cross-motions for summary judgment.
Forth, this case likewise involves the same claims as Kelley. Indeed, while
the causes of action might be different and some evidence that “was not relevant
in” the first case might be introduced, Lane v. Peterson, 899 F.2d 737, 743-44 (8th
Cir. 1990), Appellees’ current midazolam claim is based on the same facts as
Kelley and Williams. In fact, the same legal standard governs both the federal and
state claims. See Johnson, 496 S.W.3d at 357 (adopting Baze and Glossip
standards).
Fifth, privity is satisfied because this case involves the same parties asserting
the same rights. See Sparkman, 775 F.3d at 999. Therefore, res judicata applies
and the injunction must be vacated.

8

 B. Collateral estoppel bars Appellees’ claim.
Collateral estoppel bars parties from relitigating previously resolved issues.
In re Scarborough, 171 F.3d 638 (8th Cir. 1999). State law determines whether
that doctrine applies, and in Arkansas, it applies to previously litigated issues that
were essential to—and resolved by—the previous court’s final judgment. State
Office of Child Support Enforcement v. Willis, 59 S.W.3d 438, 444 (Ark. 2001);
see Baker Elec. Coop., Inc. v. Chaske, 28 F.3d 1466, 1475 (8th Cir. 1994). That
doctrine applies here because in Kelley, Appellees challenged midazolam’s use and
argued—as here—that its use in executions is “sure or very likely to cause serious
illness and needless suffering” and that there are alternative execution methods that
are “feasible, readily implemented, and in fact significantly reduce a substantial
risk of severe pain.” Glossip, 135 S. Ct. at 2737. Indeed, those claims were
addressed in dismissal briefing and summary judgment motions, and, as discussed
above, they were resolved by the Arkansas Supreme Court’s conclusion that
Appellees had not adduced evidence on summary judgment demonstrating that
Arkansas has access to a known, feasible, readily implemented, and available
alternative method of execution.
Further, Appellees cannot avoid collateral estoppel simply by pointing to
new factual allegations because collateral estoppel bars relitigation of an ultimate
fact issue—i.e., whether the use of midazolam in lethal injection is cruel or
9

 unusual—when that issue was determined by a final judgment in a prior
proceeding. E.g., Lundquist v. Rice Memorial Hospital, 121 Fed. Appx. 664, 66869 (8th Cir. 2005). Thus, collateral estoppel bars Appellees’ attempt to relitigate
their midazolam claim.
B. Appellees failed to show that midazolam is sure or very likely to cause
needless suffering.
Baze and Glossip require Appellees to show that Arkansas’s protocol gives
rise to a “substantial risk of severe pain.” Jones v. Hobbs, 604 F.3d 580, 581 (8th
Cir. 2010). The district court held that Appellees had shown that the use of
midazolam “qualifies as an objectively intolerable risk” of severe pain and that the
“risk is exacerbated” by Arkansas’s compressed schedule. This Court reviews the
district court’s ultimate legal conclusion that midazolam presents an “objectively
intolerable risk” of pain de novo. See Heartland Academy Comm. Church v.
Waddle, 335 F.3d 684 (8th Cir. 2003); Harbison v. Little, 571 F.3d 531, 535 (6th
Cir. 2009).
The district court’s order ignores the following undisputed evidence:
 According to the manufacturer-prepared, FDA-approved drug information
for midazolam, the drug can be used to induce general anesthesia, without
narcotic premedication or other sedative premedication, at a dose of 0.3 to
0.35 mg/kg. (DE 28-6, 6-7).

10

  Appellees’ anesthesiology expert concedes that midazolam can be used for
induction of anesthesia and used as the sole anesthetic for very painful
medical procedures, such as endotracheal intubation and C-section surgeries.
(Tr. 942-43).
 A 20 to 30 mg dose of midazolam will induce general anesthesia in a 200-lb
man sufficient to endure painful medical procedures, including
laryngoscopies and endotracheal intubations. A 400-pound man would
require a 40 to 60 mg dose. (Tr. 998-1000, 1057-58).
 A 500-mg dose of midazolam is far more than you would need to anesthetize
a person and render him insensate to severe pain, even using the Appellees’
expert data, and the effects of such a large dose would last far longer than
necessary to carry out an execution. (Tr. 1003-04).
 Appellees’ expert, Craig Stevens, co-authored a pharmacology textbook
listing “Anesthesia” as the major clinical use of midazolam and explaining
that “[i]ntravenous administration of benzodiazepines [like midazolam] can
produce anesthesia and mild respiratory depression.” Stevens also explains
that, “[m]idazolam is used intravenously as an anesthetic for patients
undergoing endoscopy, other diagnostic procedures, or minor surgery.” (DE
28-5, 3, 4, & 6).

11

  Evidence established that regular clinical doses of midazolam (12 or 18 mg)
given to 24 patients without premedication to induce general anesthesia
caused them to lose consciousness and be unresponsive to stimuli. (Tr.
1089).
 Some patients who received doses of midazolam between 4.5 and 20 mgs
have demonstrated BIS scores under 60, which is the level that Appellees’
experts suggest represents a general anesthesia level of unconsciousness.
(DE 54, 58-59; DE 28-12; Tr. 95.). Moreover, to the extent patient scores
were above 60, Appellees’ expert concedes that these monitors are often not
reliable. (Tr. 24).
The district court erred by relying on testimony and evidence regarding the
so-called “ceiling affect” dose of midazolam (DE 54 at 59-61), which has never
been subject to peer review. Even Appellees’ expert, Stevens, who calculated the
ceiling-effect dose at 228 mg after cherry-picking data from animal and in vitro
studies, admitted that his methodology was not sufficiently “robust” to be
published in a peer-review journal. (Tr. 332). This type of untested analysis has
no place in the courtroom under Daubert v. Merrell Dow Pharmaceuticals, Inc.,
509 U.S. 579 (1993), and the district court clearly erred in relying upon it.
Additionally, the district court did not and could not determine at what dosage the
ceiling effect, if it even exists, occurs in humans. (DE 54, 60). Without being able
12

 to determine at what dose the ceiling effect occurs, the alleged existence of a
ceiling effect is irrelevant to the analysis. The more important point is that
everyone appears to agree that whatever the ceiling effect, it occurs after the level
at which general anesthesia can be induced by midazolam.
Even assuming that the district court correctly concluded that the expert
evidence cut 50-50, Glossip and Baze require Appellees to be able to show a
significant possibility that the execution methodology is sure or very likely to
cause severe pain. Thus, Appellees cannot possibly meet their burden.
Lacking sufficient scientific evidence, the district court resorted to basing its
decision on a few anecdotal accounts of executions in other states—most of which
used completely different drug protocols or experienced problems unrelated to
midazolam. A perfect example of the district court’s selective use is its fixation on
the Lockett execution. (DE 54, 14-17). Yet the Lockett execution failed not
because of the midazolam, but because the IV line dislodged unbeknownst to state
officials. Arkansas takes precautions to ensure what occurred in the Lockett
execution cannot happen here. (DE 28-1, 30 & 44) (unlike Oklahoma, Arkansas
uses two IV sites instead of one, continuously monitors the IV sites, and ensures
constant real-time communication between the executioner and medical staff in the
execution chamber). Perhaps most importantly, the district court fails to
acknowledge that after Lockett, Oklahoma had a more recent execution with the
13

 500 mg midazolam protocol. And witnesses to that execution reported that it
“went very smoothly” and the drug dosage worked. (Tr. 423-24).
Glossip upheld a finding that a 500 mg dose of midazolam—the same dose
in Arkansas’s protocol—reliably produces anesthesia and renders an inmate
insensate to noxious stimuli such as pain. 135 S. Ct. at 2742. It also explained that,
despite the dueling experts, the only real-world evidence of the effects of a 500 mg
dose of midazolam involve the “12 other executions [that] have been conducted
using the three-drug protocol at issue here . . . without any significant problems.”
Id. at 2746. Since Glossip, even more executions have been carried out using the
same protocol as Arkansas, and there is no evidence in this case or otherwise that
any of those executions, when carried out pursuant to plan, suffered any significant
problems. And when repeatedly pressed for evidence that Arkansas’s protocol is
sure or very likely to cause needless suffering, Appellees’ experts had none. The
claims are based on nothing more than speculation, conjecture, and assumptions of
future negligence on the part of ADC’s employees, which cannot support a valid
Eighth Amendment claim.
C. Appellees failed to show a readily implementable alternative execution
method.
Appellees were also required to show a significant likelihood of proving that
“any risk posed by the challenged method is substantial when compared to known
and available alternative methods of execution.” Glossip, 135 S. Ct. at 2737-38.
14

 Accordingly, they are required to point to a known alternative “that is ‘feasible,
readily implemented, and in fact significantly reduce[s] a substantial risk of severe
pain.’” Id. at 2737. It is only where “a State refuses to adopt such an alternative in
the face of these documented advantages, without a legitimate penological
justification for adhering to its current method of execution,” that the “refusal to
change its method can be viewed as ‘cruel and unusual’ under the Eighth
Amendment.” Baze, 553 U.S. at 52.
In an unfortunately prescient manner, the Supreme Court warned against the
watering down Glossip’s second prong. See id. at 52 n.3 (noting that “articulated
standards” are important to prevent courts from becoming “boards of inquiry
determining best [execution] practices”); see also id. at 63-64 (Alito, J.,
concurring) (“Properly understood, this standard will not, as Justice Thomas
predicts, lead to litigation that enables those seeking to abolish the death penalty to
embroil the States in never-ending litigation concerning the adequacy of their
execution procedures.). The district court’s decision—including its adoption of the
Sixth Circuit’s loose interpretation of “available” over the Eleventh Circuit’s
stricter interpretation—does precisely what the Supreme Court warned against.
The district court erroneously assumed that an alternative method is
“available” if there is any remote possibility that Arkansas can obtain drugs or
tools necessary for an alternative method. See DE 54, at 78-80. But, as the
15

 Eleventh Circuit concluded in Arthur v. Commissioner of Alabama Department of
Corrections, 840 F.3d 1268 (11th Cir. 2016) and Brooks v. Warden, 810 F.3d 812
(11th Cir. 2016), the Baze and Glossip standard is much more demanding. If the
drugs or tools for an alternative method of execution are not actually available to
Arkansas despite normal efforts to obtain them, such an alternative cannot be
considered “feasible and readily implemented” and Arkansas cannot be considered
to be acting wantonly in foregoing that hypothetically potential alternative. The
Eleventh Circuit’s test for “availability” is more faithful to the words and purpose
of the second prong set out in Baze and Glossip. It is also more faithful to the
actual outcome in Glossip—that that pentobarbital and sodium thiopental were not
“available” because Oklahoma had been “unable to procure [them] despite a goodfaith effort to do so.” Glossip, 135 S. Ct. at 2738.3
This Court should adopt the Eleventh Circuit’s approach to “availability.”
This Court’s post-Baze cases indicate an understanding that the second prong of
Glossip must be strictly observed and requires that an alternative be actually, as
opposed to hypothetically, available. See, e.g., Johnson v. Lombardi, 809 F.3d 388
(8th Cir. 2015) (“Johnson’s threadbare assertion that lethal gas is legally available
in Missouri is not the same as showing that the method is a feasible or readily
implementable alternative method of execution. Indeed, nowhere in Johnson’s
3

This is precisely why the Arkansas Supreme Court ruled against Appellees. See
496 S.W.3d at 359.
16

 complaint does he plead that Missouri could readily implement the lethal-gas
method.”)
The district court justified its adoption of the looser Sixth Circuit standard
on two grounds. First, it concluded that the Eleventh Circuit’s test “places an
almost impossible burden on plaintiffs . . . .” (DE 54, at 79.) But the district
court’s conclusion is belied by the record since Appellees hired an investigator to
determine whether any manufacturers or suppliers would sell sevoflurane gas to
Arkansas. Tr. at 212-21. And, indeed, Appellees found a willing supplier. Id.
There is no reason Appellees could not have taken that same approach in regard to
other potential alternative methods. They simply chose not to do so. Id. at 217-21.
Second, the district court was concerned that the stricter interpretation of
Glossip “may put plaintiff’s counsel in the position of offering a state the drugs
that would be used to end their client’s lives.” (DE 54, at 80.) But that makes no
sense in the context of a claim where the client is not fighting the propriety of his
conviction or sentence. The inmate is not contesting his guilt or his sentence, but
rather seeking a constitutionally proper execution. In this context, no ethical
quandary is created by an inmate’s attorney providing the state with information
about an alternative drug that he or she believes would significantly reduce a
substantial risk of severe pain. In any event, whether or not it might be
discomforting to an inmate’s attorney, identifying an alternative that is actually
17

 available is critical to the test. Thus, unless Appellees can show that Arkansas is
consciously disregarding an alternative that is actually available to it, Appellee’s
cannot show a constitutional violation.
The district court also misinterpreted Baze’s and Glossip’s requirement that
the known and available alternative “would in fact substantially reduce a risk of
severe pain.” Glossip, 135 S. Ct. at 52 (emphasis added). This burden is not met
“by showing a slightly or marginally safer alternative.” Id. And it is not met by
some controverted evidence or theory. Cf id. at 67 (Alito, J., concurring) (“[A]n
inmate should be required to do more than simply offer the testimony of a few
experts or a few studies. Instead, an inmate challenging a method of execution
should point to a well-established scientific consensus. Only if a State refused to
change its method in the face of such evidence would the State's conduct be
comparable to circumstances that the Court has previously held to be in violation
of the Eighth Amendment.”). The district court’s cursory analysis of whether the
suggested potential alternatives would meet this standard fails to enforce the heavy
burden on the Plaintiffs who seek a stay. (DE 54, at 80-86) (discussing six
potential alternatives). The district court’s analysis rather allows Appellees to
obtain a stay based on little more than speculation and uncertainty. Id.
Furthermore, Appellees suggested, and the district court accepted, that FDAapproved, bulk manufactured pentobarbital and compounded pentobarbital were
18

 “available for use in executions.” (DE 54, at 81.) The district court did not
suggest either was actually available to Arkansas. Instead, the district court
suggested that Missouri, Texas, and Georgia have on at least one occasion used
either FDA manufactured or compounded pentobarbital. Id. As explained above,
this does not show that Arkansas can obtain such drugs.
The district court suggests that Arkansas’s confidentiality statute (shielding
in most instances the identity of a manufacturer and supplier of drugs) should aid
ADC in obtaining drugs. (DE 54, at 82.) But that Arkansas could not obtain a
particular drug prior to the confidentiality statute being put in place does not
automatically mean that Arkansas can obtain that drug simply because the statute
exists. Indeed, the district court fails to wrestle with the fact that even after the
statute was enacted, Director Kelley tried to but could not obtain FDA-approved
and bulk-manufactured pentobarbital. (See Tr. at 1265; DE 28-1.) The district
court appears to discount the Director’s efforts because they occurred in 2015, not
2017. (DE 54, at 75.) But a constitutional rule that would require ADC to
continuously search for an alternative drug is unworkable and has not been
required by any court to date. Moreover, in 2016, Rory Griffin attempted to obtain
FDA-approved and bulk manufactured pentobarbital and could not obtain it. (DE
28-13, 3-8.) While the district court found that Griffin’s search was in anticipation
of litigation, it did not find that search lacked good-faith. And in light of Director
19

 Kelly’s inability to locate manufactured pentobarbital in 2015, this Court should
conclude that it is not “available” to Arkansas.
With respect to the compounded pentobarbital, it is impossible to
understand how the district court determined that such an alternative would in fact
substantially reduce a risk of severe pain. In their state court complaint in 2016,
Appellees spent pages detailing the “grave risks” of compounded drugs, including
compounded pentobarbital. (DE 28-18.) Appellees pled that compounded drugs
carried such grave risk of causing pain and suffering that their use would be cruel
or unusual. Id. They supported this pleading with an affidavit from Larry Sasich.
Id. Given Appellees’ previous pleadings, it cannot be said with any certainty that
compounded pentobarbital would “in fact substantially reduce a risk of severe
pain.”
Appellees next posit that a gas overdose of sevoflurane or nitrogen hypoxia
as available alternative methods of execution. The district court acknowledged
that neither of these alternatives have ever been used in the United States. (DE 54,
at 82.) Nonetheless the district court found them to be “known and available
alternatives.” With respect to nitrogen hypoxia, the Court relied on the fact that
Louisiana had studied it as a potential method of execution and that Oklahoma has
adopted it (but not used it). With respect to sevoflurane, the Court merely relied on
Dr. Steven’s testimony that he believed it could be done. (DE 54, at 83.)
20

 Methods that have never been used in the United States cannot logically
qualify as “known and available” alternatives. If they did, they would render the
second prong of Baze and Glossip into nothing more than a test of imagination.
Indeed, it would undermine the entire purpose of the test, which would be to
determine whether or not the State was unreasonably refusing to move to a tried
and true alternative. A fortiori, a state would have a justifiable reason for not
moving to an execution method that had never been used before. Arkansas thus
cannot be considered to be acting wantonly by not adopting such a novel execution
method.
Appellees’ last alternative is the firing squad. Even assuming that the
necessities for a firing squad (facility, plan, training, etc.) were available, the
district court erroneously concluded that the Appellees showed a likelihood of
success in proving that this method of execution would in fact substantially reduce
a risk of severe pain. The district court primarily relies on Groner’s testimony
regarding the firing squad. (DE 54, at 83-85.) Groner is a pediatric trauma surgeon
whose patients are often child gunshot victims, and he conceded that he was not an
expert in marksmanship. Id. He is also not a firearms expert. (Tr. at 602.)
Groner opined that there is less risk of pain associated with the firing squad
than midazolam. His opinion was based on his view that if shot in the left ventricle
with multiple bullets, an inmate would lose consciousness (from blood loss to the
21

 brain) within a few seconds. He did not base this opinion on any studies or
scientific articles. Instead, he based it on his experience as a trauma surgeon. But
he conceded that his trauma patients do not arrive at the hospital until at least
several minutes (sometimes much longer) after they have been shot. Id. at 612-13.
Thus, it is unclear what the basis for his opinion is.
Groner conceded that if it takes more than a few second to lose
consciousness, getting shot with five bullets would amount to significant pain.
Groner also conceded that he could not provide any opinion as to how often the
firing squad would hit their target—the left ventricle. He could not say whether it
would be 100, 75, 50, or 25 percent of the time. Id. at 605. And he conceded that
the Utah firing squad protocol contemplates the potential need for a second volley
of shots when the first volley leaves inmates conscious. Id. at 606-07.
The district court never expressly concludes that the firing squad will in fact
substantially reduce a risk of severe pain. But to the extent that can be implied
from the district court’s decision concluding that the firing squad is a reasonable
alternative, its decision is based on nothing more than speculation. Indeed, in this
vein, the district court pointed to the fact that some prison officers are qualified to
and can shoot prisoners from guard towers. (DE 54, at 85.) But that officials can
shoot a prisoner from a guard tower if an emergency occurs does not mean that
they could reliably shoot the left ventricle in the perfect spot to ensure quick loss of
22

 consciousness. It is precisely this type of non-scientific speculation that would
render the Baze and Glossip test meaningless.
II. The district court failed to properly consider the public interest or weigh
that interest against Appellees’ dilatory conduct.
“Both the State and the victims of crime have an important interest in the
timely enforcement of a sentence,” Hill, 547 U.S. at 584, and that interest weighed
heavily against issuing an injunction in a case that does not concern guilt or
innocence. See Nooner, 491 F.3d at 807-08. Indeed, Appellees have had multiple
opportunities to challenge their convictions, sentences, and—critically—their
method of execution. Their guilt is beyond dispute, and Arkansas is entitled to see
that their victims receive justice decades after Appellees’ horrific crimes. The
district court’s bare passing reference to those facts demonstrates that it abused its
discretion in failing to consider those facts. (DE 54, at 86.)
Further, the district court failed to consider the fact that the preliminary
injunction in this case will make it impossible for Arkansas to perform lawful
executions because Arkansas’s supply of midazolam (the critical component of
Arkansas’s lethal injection protocol) expires in two weeks. Nor did the district
court consider the fact that Appellees—knowing that fact—waited until the very
eve of their executions to bring this action. Indeed, as Appellees’ repeated
challenges (and countless voluntary dismissals) of their midazolam claim
demonstrate, Appellees did not suddenly discover their midazolam claim. And
23

 equity should not permit them to deliberately manipulate the judicial process to
evade justice.
CONCLUSION
For the foregoing reasons, the preliminary injunction should be vacated.
Respectfully submitted,
LESLIE RUTLEDGE
Attorney General
LEE RUDOFSKY
Solicitor General
NICHOLAS J. BRONNI
Deputy Solicitor General
COLIN JORGENSEN
Senior Assistant Attorney General
JENNIFER MERRITT
Assistant Attorney General
/s/ Lee Rudofsky
OFFICE OF THE ARKANSAS
ATTORNEY GENERAL
323 Center St., Suite 200
Little Rock, AR 72201
(501) 682-6302
Lee.Rudofsky@arkansasag.gov
April 15, 2017

24

 CERTIFICATE OF SERVICE

I, Lee Rudofsky, hereby certify that on April 15, 2017, I filed the foregoing with
the Clerk of the Court via email and also provided the foregoing via email to the following:
Scott Braden
scott_braden@fg.org
Julie Vandiver
julie_vandiver@fd.org
Jeff Rosenzweig
jrosenzweig@att.net
Lee Short
leedshort@gmail.com
Attorneys for Appellees
s/ Lee Rudofsky
Lee Rudofsky

25

 CERTIFICATE OF COMPLIANCE
Pursuant to Rules 27(d)(2) and 32(c)(1) of the Federal Rules of Appellate
Procedure, the undersigned hereby states that the applicable portions of this motion
contain 5,183 words in proportionally sized 14-point Times New Roman font. The
brief was prepared in Microsoft Office Word 2010. The electronic version of the
brief has been scanned for viruses and is virus-free.

/s/ Nicholas J. Bronni
Nicholas J. Bronni

26