Workman v. Bredesen

• Court: Court of Appeals for the Sixth Circuit
• Decided: May 7, 2007
• Docket: 07-5562
• Status: Published

Opinion

RECOMMENDED FOR FULL-TEXT PUBLICATION Pursuant to Sixth Circuit Rule 206 File Name: 07a0162p.06

UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT _________________

X Petitioner-Appellee, - PHILIP WORKMAN, - - - No. 07-5562 v. , > GOVERNOR PHIL BREDESEN, et al., - Respondents-Appellants. - N Appeal from the United States District Court for the Middle District of Tennessee at Nashville. No. 07-00490—Todd J. Campbell, District Judge. Decided and Filed: May 7, 2007 Before: SILER, COLE, and SUTTON, Circuit Judges. _________________ OPINION _________________ SUTTON, J., delivered the opinion of the court, in which SILER, J., joined. COLE, J. (pp. 29-37), delivered a separate dissenting opinion. SUTTON, Circuit Judge. Philip Ray Workman is scheduled to be executed by the State of Tennessee on May 9, 2007, at 1:00 a.m., for the murder of Lieutenant Ronald Oliver. On May 4, 2007, Workman filed a motion for a temporary restraining order in federal district court, claiming that the State’s three-drug protocol for implementing the death penalty violates the Eighth (and Fourteenth) Amendment, and later that day the court granted the motion. Still later that same day, the Governor of Tennessee and the other defendants filed an appeal from that order. Early today, May 7, 2007, the Governor and others filed a 19-page motion in this court to vacate the district court’s order. A little later this morning, Workman filed a 45-page brief in response. This dispute arises from a 25-year-old capital sentence, and the district court’s order, if upheld, would be Workman’s sixth stay of an execution date set by the State over the last seven years. At no point until last Friday, May 4, 2007, did Workman challenge the State’s method of execution, even though the components of the procedure that Workman challenges today have been in existence in the main since 1998. He thus cannot escape the Supreme Court’s and this court’s limitations on dilatory challenges to an execution procedure. Workman’s prospects for success on the merits also are dim. The Supreme Court has never invalidated a State’s chosen method of execution. No court has invalidated the three-drug protocol used by Tennessee (and 29 other jurisdictions). Several state and federal courts have upheld this same three-drug protocol (including the Tennessee Supreme Court in 2005). Our court vacated a similar stay decision in 2006 with respect to a similar challenge and permitted the State to execute the inmate under the protocol. Notwithstanding the decision of the Tennessee Supreme Court in

1 No. 07-5562 Workman v. Bredesen, et al. Page 2

2005 and the decision of this court in 2006, the State undertook an effort in 2007 to review and improve the procedure. Workman acknowledges that the new procedure is only slightly different from the old procedure, and he offers no explanation how Tennessee has done anything more than make the new procedure less prone to implementation errors. Everything, indeed, the State has done in reviewing and revising the procedure shows that it is trying to prevent Workman from suffering any pain during his execution, not that it is trying or willing to allow a procedure that imposes unnecessary and wanton pain. For these reasons and those elaborated below, we vacate the district court’s temporary restraining order. I. A. On August 5, 1981, at 10:00 p.m., Workman walked into a Wendy’s restaurant in Memphis, Tennessee, held the employees and a customer at gunpoint, herded them into the manager’s office and ordered the manager to empty the day’s receipts into a bag. State v. Workman, 667 S.W.2d 44, 46 (Tenn. 1984). He instructed the employees to remain in the office, stole an employee’s car keys, locked the door and started to leave the restaurant. Id. Responding to a silent alarm that one of the employees had triggered, Lt. Oliver stopped the defendant as he was exiting. Id. At some point, Workman broke away from Lt. Oliver and fled. Additional officers at the scene grabbed Workman, who broke free of their grasp, then shot Lt. Oliver in the chest and a second officer in the arm, fired a second shot at the second officer, then ran to a business next door, pausing mid-flight to fire another bullet at a third officer. Id. Lt. Oliver died from the gun shot. Id. at 47. Police cordoned off the crime scene area and after an extensive search found Workman hiding in the underbrush with a .45 caliber pistol nearby. Id. In 1982, a jury found Workman guilty of first-degree murder and imposed a capital sentence. Id. Direct review of Workman’s conviction and sentence ended without success in 1984. See State v. Workman, 667 S.W.2d 44 (Tenn.), cert. denied, Workman v. Tennessee, 469 U.S. 873 (1984). In 1986, the Shelby County Criminal Court denied Workman’s first petition for post-conviction relief, which the Tennessee Court of Criminal Appeals affirmed. Workman v. State, C.C.A. No. 111, 1987 WL 6724 (Tenn. Ct. Crim. App. 1987). The Tennessee Supreme Court denied leave to appeal, and so did the United States Supreme Court. Workman v. Tennessee, 484 U.S. 873 (1987). In 1988, Workman filed a second petition for post-conviction relief, which also was unsuccessful. See Workman v. State, 868 S.W.2d 705 (Tenn. Ct. Crim. App. 1993), cert. denied, Workman v. Tennessee, 510 U.S. 1171 (1994). In 1994, Workman filed a petition for a writ of federal habeas corpus in the United States District Court for the Western District of Tennessee. When that proved unsuccessful, see Workman v. Bell, 178 F.3d 759 (6th Cir. 1998), cert. denied, 528 U.S. 913 (1999), he filed several other petitions in federal court, which also proved unsuccessful. See Workman v. Bell, 227 F.3d 331 (6th Cir. 2000) (en banc), cert. denied, 531 U.S. 1193 (2001) (unsuccessful request to reopen habeas proceedings); Workman v. Bell, 245 F.3d 849 (6th Cir. 2001), cert. denied, 532 U.S. 955 and In re Workman, 532 U.S. 954 (2001) (second attempt to reopen habeas proceedings denied). In March 2001, he collaterally attacked his conviction in state court, filing a petition for a writ of coram nobis. See Workman v. Tennessee, No. 81239 (Tenn. Crim. Ct. March 28, 2001); Tenn. Code § 40-26-105. The state courts rejected the claim. See Tennessee v. Workman, 111 S.W.3d 10 (Tenn. Ct. Crim. App. 2002). In 2003, Workman returned to federal court. He filed a motion for relief from the district court’s denial of his first federal habeas petition, see Fed. R. Civ. P. 60(b), claiming that the No. 07-5562 Workman v. Bredesen, et al. Page 3

Tennessee Attorney General perpetrated a fraud upon the district court during Workman’s habeas proceedings. The district court denied the motion on October 17, 2006. On January 17, 2007, the Tennessee Supreme Court scheduled Workman’s execution for May 9. On February 1, Governor Bredesen issued an executive order suspending Tennessee’s lethal- injection protocol and asked the Commissioner of Corrections to review the State’s capital punishment administration procedures and to develop a new protocol by May 2. See State of Tennessee, Executive Order by the Governor No. 43 (Feb. 1, 2007). In late April (April 30), the Governor announced the new lethal-injection procedure for the State, which left the prior procedure unchanged in the main, though it formalized some components of the procedure and improved others. On April 27, the district court granted Workman a certificate of appealability to seek review of the district court’s denial of his Rule 60(b) motion but denied Workman’s request for a stay pending appeal. On May 1, Workman filed an appeal from the Rule 60(b) motion and sought a stay of his execution in this court. On May 4, 2007, we denied the motion for a stay, concluding that Workman had not shown a likelihood of success in reversing the district court’s Rule 60(b) decision. Workman v. Bell, Nos. 06-645, 07-5031, Order at 4–5 (6th Cir. May 4, 2007). That same day, Workman filed a new complaint in a different federal court—the Middle District of Tennessee. In an 82-page complaint, he challenged the constitutionality of the State’s lethal-injection protocol and sought a temporary retraining order suspending his May 9 execution date. Later that same day (still May 4), the district court granted a temporary restraining order, set a preliminary injunction hearing for May 14 regarding the constitutionality of the procedure and effectively stayed Workman’s May 9 execution date. B. Given the nature of Workman’s challenge, a brief review of the history of Tennessee’s execution procedures is in order. In 1796, Tennessee became a State, and its first constitution mentioned “capital offenses.” The common law permitted the death penalty, which the State generally carried out by hanging. Tennessee Dep’t of Corr., Capital Punishment Chronology, available at http://www.state.tn.us/correction/newsreleases/pdf/chronology-rev0905.pdf (last visited May 7, 2007). The State adopted a criminal code in 1829, which codified hanging as the sole method of imposing the death penalty. Id. Death by hanging remained the method of execution in Tennessee until 1913, when the State replaced the gallows with the electric chair. Tennessee’s last execution by electrocution took place in 1960. In the aftermath of Furman v. Georgia, 408 U.S. 238 (1972), the Tennessee General Assembly passed a new death penalty statute in 1974, which the State’s Supreme Court later declared unconstitutional. The State enacted another death-penalty law in 1977, which still authorized electrocution as the State’s method of execution. Tennessee Dep’t of Corr., Capital Punishment Chronology. In 1998, Tennessee legislators made lethal injection an option for capital inmates though it continued to allow inmates to choose electrocution. Abdur’Rahman v. Bredesen, No. M2003-01767- COA-R3-CV, 2004 WL 2246227, at *3 (Tenn. Ct. App. Oct. 6, 2004); see also Tenn. Code. Ann. § 40-23-114(a). The State’s General Assembly vested the Tennessee Department of Corrections with the authority “to promulgate necessary rules and regulations to facilitate the implementation” of death by lethal injection. Tenn. Code § 40-23-114(c). The Department of Corrections decided to use a three-drug protocol modeled after the one that Texas adopted because “Texas had the most experience with carrying out executions by lethal injection.” Abdur’Rahman, 2004 WL 2246227, at *3. No. 07-5562 Workman v. Bredesen, et al. Page 4

In 2003, an inmate challenged the constitutionality of the State’s lethal-injection protocol under the Eighth Amendment. In 2005, the Tennessee Supreme Court rejected the challenge. See Abdur’Rahman v. Bredesen, 181 S.W.3d 292, 314 (Tenn. 2005). On February 1, 2007, the Governor directed the Department of Corrections to engage in a “comprehensive review” of Tennessee’s death penalty protocol and procedures, both for lethal injection and for electrocution. See Executive Order No. 43; Report on Administration of Death Sentences in Tennessee (Report) at 3 (Apr. 30, 2007) (see Appendix A). The Commissioner of the Department formed a committee, which conducted research, obtained input from experts, sought information from other jurisdictions and updated its execution manual for carrying out the death penalty. Report at 3. The updated execution manual includes: • Detailed descriptions of each step of the electrocution and lethal- injection processes. • Detailed descriptions of the qualifications, selection processes, and training requirements for execution team members. • A detailed description of the services provided to family members of the condemned inmate’s victims. • Enhanced requirements for contemporaneous documentation of each significant stage of an execution as it is carried out. • Enhanced accountability in connection with the procurement, storage, and disposition of the legal-injection chemicals. Report at 1. In reviewing the State’s lethal-injection protocol, the committee focused on (1) what chemicals should be used in the procedure and in what quantities, (2) what training and qualifications the executioner and the IV team should possess, (3) what method the IV team should use to administer drugs if unable to establish “peripheral venous access,” and (4) what documentation should be required regarding the administration of the chemicals. Report at 4. The committee met 19 times from February through April 2007 and held a public hearing in April before issuing its findings. Report at 5. Chemical selection and amounts. Tennessee decided to retain the three-drug protocol it had adopted in 1998, a protocol that 29 other jurisdictions, including the Federal Government, employ. Report at 6. The protocol involves the injection of 5 grams of sodium thiopental followed by 100 milligrams of pancuronium bromide (Pavulon) followed by 200 millequivalents of potassium chloride, all delivered intravenously. Report at 6. The dose of sodium thiopental, a barbiturate that “reduces oxygen flow to the brain and causes respiratory depression,” Execution Procedures for Lethal Injection at 35, quickly anesthetizes the inmate and is sufficient to cause death in the absence of the two additional chemicals in the protocol. Pancuronium bromide is a “muscle paralytic” that “assist[s] in the suppression of breathing and ensure[s] death.” Id. The amount of pancuronium bromide the State administers also proves fatal on its own, and the State selected the drug because it hastens death and “prevents involuntary muscular movement that may interfere with the proper functioning of the IV equipment,” thus “contribut[ing] to the dignity of the death process.” Report at 6, 7. Potassium chloride, a salt, interferes with heart function, causing “cardiac arrest and rapid death.” Report at 6; Execution Procedures for Lethal Injection at 35. If administered properly, the sodium thiopental anesthetizes inmates before they receive the remaining two drugs. Report at 7. No. 07-5562 Workman v. Bredesen, et al. Page 5

Before deciding to continue using the three-drug protocol, the committee considered two variations on this procedure—a two-drug option and a one-drug option. The two-drug option would have eliminated pancuronium bromide from the protocol. The State rejected this approach after concluding that “the administration of potassium chloride without a preceding dose of pancuronium bromide would typically result in involuntary movement which might be misinterpreted as a seizure or an indication of consciousness.” Report at 8. The State rejected a one-drug protocol, which would have used sodium thiopental alone, because it would slow down the death process and because “the effect of the required dosage of sodium thiopental would be less predictable and more variable when it [was] used as the sole mechanism for producing death.” Id. On top of these concerns, the commission noted that no State had used a one- or two-drug protocol to carry out an execution, leaving Tennessee with no data or case studies to indicate that these options would work better and would operate more humanely than the three-drug protocol. Id. The State also considered the benefits of the three-drug protocol—that the protocol had worked well in the past, that all courts that had reviewed the protocol had deemed it constitutional and that dozens of States had used it and thus could provide information, data and expertise about their experiences with it and refinements to it. The State balanced these considerations against the risks associated with the three-drug protocol—that using three different chemicals is the “most complicated” of the three options, that “there is a remote chance of an error in implementation that may cause the inmate to incur brief pain” and that pancuronium bromide requires special attention because it must be refrigerated. Report at 7–8. Balancing these costs and benefits, Tennessee chose to continue using the three-drug protocol. Training and qualifications of IV team and executioner. Tennessee’s IV teams have always consisted of “Emergency Medical Technicians with IV certification or certified paramedics,” Report at 9, but the previous execution manual did not contain language requiring team members to possess this training and these qualifications. The State updated the manual to make these requirements express. Id.; see also Execution Procedures for Lethal Injection at 32–33. Despite the fact the State has mandated that IV team members adhere to continuing education requirements to keep their licences and certification current, as well as required that team members “regularly practice establishing IV access during execution training exercises,” these requirements had not been contained in the execution manual. The State also made these requirements express. Report at 9; Execution Procedures for Lethal Injection at 33, 50. Tennessee, finally, formalized its previously unwritten rule that executors must be trained in IV therapy so that they can recognize when they have failed to establish IV access. Report at 9; Execution Procedures for Lethal Injection at 32–33. Cut-down procedure. If the IV team cannot locate a usable vein during an execution (due for example to drug use by the inmate), see Execution Procedures for Lethal Injection at 41, Tennessee uses a cut-down procedure—which means that a physician makes an incision in order to obtain IV access. After reviewing the cut-down procedure and its alternatives “with several experts,” the State concluded that “cut-down procedures are not particularly difficult for physicians to perform,” and therefore decided to keep the procedure as its contingency plan during the lethal- injection process. Report at 9; Execution Procedures for Lethal Injection at 20, 41, 67. Documentation. Tennessee revised its procedures to require “enhanced documentation” for procurement and storage of the chemicals used for lethal injection. Report at 9. The State also amended its execution manual to include a “contemporaneous documentation” requirement—meaning that a member of the IV team must document the administration of the three- drug protocol during an execution. Id. at 10; Execution Procedures for Lethal Injection at 21, 65, 82–86. No. 07-5562 Workman v. Bredesen, et al. Page 6

II. We have jurisdiction, as an initial matter, to review the district court’s temporary restraining order. See N.E. Ohio Coal. for Homeless & Serv. Employees Int’l Union, Local 1199 v. Blackwell, 467 F.3d 999, 1005 (6th Cir. 2006). While we generally do not have jurisdiction to review temporary restraining orders, our jurisdiction is not controlled by the name that a claimant attaches to a motion or the name that a district court attaches to an order. Rather than looking to “the label attached by the trial court, we “look[] to the nature of the order and the substance of the proceeding below to determine whether the rationale for denying appeal applies.” Id. Were the shoe on the other foot, we suspect, Workman would agree. Name aside, the salient question is whether the order effectively operates as an “injunction”—which is what 28 U.S.C. § 1292(a)(1) permits us to review on an interlocutory basis. The Tennessee Supreme Court set Workman’s execution date for May 9; the temporary restraining order runs through May 14; and once May 9 passes Tennessee may not execute its judgment until a new execution date is set by order of the Tennessee Supreme Court. See Tenn. Sup. Ct. R. 12.4(E). To suggest, as Workman does, that the district court’s order does not enjoin Tennessee from executing Workman on May 9, that the State has meaningful appellate options for imposing this 25- year-old sentence other than through interlocutory review and that the trial court’s order does not affect an important interest of the State is untenable. The order “has the practical effect of an injunction,” which simultaneously operates to stay Workman’s long-delayed execution and to give us authority to review it. N.E. Ohio Coal., 467 F.3d at 1005; 28 U.S.C. § 1292(a)(1); see Boltz v. Jones, 182 F. App’x 824, 825 (10th Cir. June 1, 2006) (vacating temporary restraining order that stayed execution, reasoning that “[t]hough the order is denominated a TRO rather than an injunction, we have jurisdiction pursuant to 28 U.S.C. § 1292(a)(1)”); see also Carson v. Am. Brands, Inc., 450 U.S. 79, 84 (1981) (holding that an interlocutory order is immediately appealable when appellant can show that the order might have a “serious, perhaps irreparable consequence, and that the order can be effectually challenged only by immediate appeal”) (internal quotation marks omitted); cf. Hill v. McDonough, 126 S. Ct. 2096, 2104 (2006) (noting “the State’s strong interest in enforcing its criminal judgments without undue interference from the federal courts”). III. We review a district court’s decision to issue a temporary restraining order for abuse of discretion. See Bowersox v. Williams, 517 U.S. 345, 346 (1996) (per curiam) (reviewing order involving stay of execution); N.E. Ohio Coal., 467 F.3d at 1009; see also Alley v. Little, 181 F. App’x 509, 512 (6th Cir. May 12, 2006) (explaining that we “weigh the merits of the district court’s stay [of an execution] . . . based on the reason furnished in its opinion.”). We will find an abuse of discretion only when left with “a definite and firm conviction that the trial court committed a clear error of judgment.” Logan v. Dayton Hudson Corp., 865 F.2d 789, 790 (6th Cir. 1989). “A district court abuses its discretion when it relies on clearly erroneous findings of fact, or when it improperly applies the law or uses an erroneous legal standard.” Tompkin v. Philip Morris USA, Inc., 362 F.3d 882, 891 (6th Cir. 2004). In reviewing the district court’s decision, we consider the same four factors that the district court considered: (1) whether the claimant has demonstrated a strong likelihood of success on the merits, (2) whether the claimant will suffer irreparable injury in the absence of a stay, (3) whether granting the stay will cause substantial harm to others, and (4) whether the public interest is best served by granting the stay. N.E. Ohio Coal., 467 F.3d at 1009. We have two objections to the district court’s order—one that the district court briefly considered in its four-page order (whether Tennessee’s three-drug protocol likely violates the Eighth Amendment) and one that it did not (whether Workman has waited too long in bringing this No. 07-5562 Workman v. Bredesen, et al. Page 7

challenge). Before we reach these issues, however, we must address Workman’s contention that the State has waived any right to challenge the district court’s decision on these two grounds. A. After challenging our power to review the district court’s order, Workman challenges the State’s power to contest the bases for it. Last Friday morning, on May 4, 2007, Workman served a motion for a temporary restraining order on the State, and the court held a hearing on the motion. At the hearing, the State took the legitimate position that the court had no authority to grant the order because Workman had not yet filed his complaint and that it would not respond to a complaint that it had not seen. After the hearing, Workman filed his 82-page complaint, and an hour later the district court (without holding a new hearing) granted the motion. The State did not waive any challenges to the motion under these circumstances, and Workman offers no relevant authority to suggest otherwise. The State also did not waive these arguments in the motion that it filed in our court this morning. As to undue delay, the State makes that consideration one of the key points of the motion (if not the key point of the motion). As to the likelihood that a court would strike down this procedure on constitutional grounds, the State throughout its motion relies heavily on our decision in Alley and the Tennessee Supreme Court’s decision in Abdur’ Rahman—one of which found that this same claim had little likelihood of success on the merits, and the other of which rejected the claim on the merits. The State also specifically critiques the district court’s reliance on the opinions of Dr. Heath as grounds for establishing a “likelihood of success,” objecting that the district court failed to address the Alley and Abdur’Rahman decisions, even though they addressed the same issue and even though they found unpersuasive the reports of this same doctor about the same procedure. Motion at 14 n.12. Again, no waiver occurred. B. Like a prior panel of this court that considered a challenge to Tennessee’s lethal-injection procedure a year ago, we think that Workman has “a small likelihood” of success with respect to this challenge because his contention is “unsupported by current law, which offers no basis for finding lethal injection protocols unconstitutional.” Alley, 181 F. App’x at 513. And like that panel, we think that the district court abused its discretion in concluding otherwise. Id. at 511. In contending that Tennessee’s lethal-injection protocol likely violates the Eighth Amendment’s prohibition on imposing “cruel and unusual punishments,” U.S. Const. amend. VIII, Workman faces several obstacles. First, the Supreme Court has never invalidated a State’s (or the Federal Government’s) chosen method of execution. As best we can tell, the Court has considered three such challenges under the Eighth Amendment, only one of which reached the merits. See Wilkerson v. Utah, 99 U.S. 130, 134–35 (1878) (holding that the use of a firing squad was not cruel and unusual); In re Kemmler, 136 U.S. 436, 446–49 (1890) (declining to apply the Eighth Amendment to the States); State of La. ex rel. Francis v. Resweber, 329 U.S. 459, 464 (1947) (declining to apply the Eighth Amendment to the States). Since these decisions, the Court has had ample opportunities to constrain methods of execution that seem to raise far greater risk of cruel and unusual punishment than lethal injection, but it has declined to do so. See, e.g., Campbell v. Wood, 511 U.S. 1119, 1119 (1994) (Blackmun, J., dissenting from denial of certiorari) (hanging); Gomez v. U.S. Dist. Court for the N. Dist. of Cal., 503 U.S. 653, 657–58 (1992) (Stevens, J., dissenting from vacation of stay of execution) (gas chamber) (critiquing the majority for sanctioning the use of the gas chamber when most States had moved away from this method of execution because it caused unnecessary infliction of pain); Glass v. Louisiana, 471 U.S. 1080, 1080 (1985) (Brennan, J., dissenting from denial of certiorari) (electrocution); Gray v. Lucas, 463 U.S. 1237, 1240 (1983) (Marshall, J., dissenting from denial of certiorari) (gas chamber). No. 07-5562 Workman v. Bredesen, et al. Page 8

Second, the experience of the lower state and federal courts is similar. No court to our knowledge has issued a final decision declaring a State’s lethal-injection protocol unconstitutional. And several lower courts have upheld this specific three-drug, lethal-injection protocol. See, e.g., Abdur’Rahman, 181 S.W.3d at 297–98; Aldrich v. Johnson, 388 F.3d 159, 161 (5th Cir. 2004) (lethal injection in Texas); People v. Snow, 65 P.3d 749, 800–01 (Cal. 2003); Sims v. State, 754 So. 2d 657, 668 (Fla. 2000); State v. Webb, 750 A.2d 448, 453–57 (Conn. 2000); LaGrand v. Stewart, 133 F.3d 1253, 1265 (9th Cir. 1998) (lethal injection in Arizona); cf. Hill v. Lockhart, 791 F.Supp. 1388, 1394 (E.D. Ark. 1992) (Arkansas lethal injection protocol); State v. Hinchey, 181 Ariz. 307, 315 (1995); State v. Deputy, 644 A.2d 411, 421 (Del. Super. Ct. 1994); State v. Moen, 309 Or. 45, 98–99 (1990); Hopkinson v. State, 798 P.2d 1186, 1187 (Wyo. 1990). One cannot credibly establish a likelihood of success in attacking a death-penalty procedure when the theory of success has yet to succeed in a considerable number of cases over a considerable number of years. Third, one of the benchmarks the Court uses to identify Eighth Amendment violations—a consideration of the “evolving standards of decency that mark the progress of a maturing society,” McCleskey v. Kemp, 481 U.S. 279, 300 (1987) (internal quotation marks omitted)—does not help Workman. The “clearest and most reliable objective evidence of contemporary values is the legislation enacted by the country’s legislatures.” Atkins v. Virginia, 536 U.S. 304, 312 (2002) (internal quotation marks omitted). As modern sensibilities have moved away from hanging, the firing squad, the gas chamber and electrocution as methods of carrying out a death sentence, so too have the death-penalty procedures of the States and the Federal Government. While the Supreme Court has tolerated continuity in this area, the democratic processes have demanded change. The method of execution in 37 of the 38 States that authorize capital sentences has evolved to make lethal injection the preferred method of carrying out a death sentence with only Nebraska clinging to electrocution. The Federal Government now uses lethal injection as well. See Death Penalty Information Center, Lethal Injections, available at http://www.deathpenaltyinfo.org/article.php?did=1686&scid=64#drugs (last visited May 7, 2007); see also Jones v. Allen, ___ F.3d ___, 2007 WL 1225393, at *3 n.3 (11th Cir. April 27, 2007) (noting that Alabama also uses 3-drug protocol). Not only is there a consensus among the States and the Federal Government that lethal injection is the most humane method of execution, but there also is a consensus among these jurisdictions that the three-drug protocol used by Tennessee is the preferred form of lethal injection. Twenty-nine States (and the Federal Government) use the same three drugs, while one State (New Jersey) uses a combination of the first two drugs in the protocol, and the remaining seven States—Kansas, Kentucky, Nevada, New Hampshire, Pennsylvania, South Carolina and Virginia—apparently do not specify what drugs they use. Death Penalty Information Center, Lethal Injections; see also Report at 6 (“At least 30 jurisdictions . . . have a three-chemical lethal injection protocol consisting of sodium thiopental, pancuronium bromide, and potassium chloride in varying amounts.”). If “evolving standards of decency” are the measure of constitutionality, Tennessee has satisfied the requirement. Fourth, the other benchmark the Court uses to identify Eighth Amendment violations—whether the punishment involves the “unnecessary and wanton infliction of pain,” Nelson, 541 U.S. at 645 (internal quotation marks omitted); cf. id. at 644 (referring to the “deliberate indifference” standard)—does not help Workman either. The whole point of the Tennessee lethal- injection protocol is to avoid the needless infliction of pain, not to cause it. The idea is to anesthetize the individual with one drug before the State administers the remaining two drugs, so that the serial combination of drugs causes a quick and pain-free death. See Abdur’Rahman, 181 S.W.3d at 307–08 (noting “that a dosage of five grams of sodium Pentothal as required under Tennessee’s lethal injection protocol causes nearly immediate unconsciousness and eventually death[,] . . . that such a dose would cause an inmate to be unconscious in about five seconds and that the inmate would never regain consciousness and would feel no pain prior to dying”). No. 07-5562 Workman v. Bredesen, et al. Page 9

No one debates these premises, or this purpose, of Tennessee’s lethal-injection procedure. We thus do not have a situation where the State has any intent (or anything approaching intent) to inflict unnecessary pain; the complaint is that the State’s pain-avoidance procedure may fail because the executioners may make a mistake in implementing it. But no one has demonstrated that this problem has occurred in Tennessee in the past, and as we have shown the State has extensive procedures in place to prevent this very thing from happening. The risk of negligence in implementing a death-penalty procedure, particularly when the risk has not come to pass in the State, does not establish a cognizable Eighth Amendment claim. At some level, every execution procedure ever used contains risk that the individual’s death will not be entirely pain free. See Beardslee v. Woodford, 395 F.3d 1064, 1075 (9th Cir. 2005) (“Obviously there are risks involved in virtually every method of execution. However, the Supreme Court has rejected Eighth Amendment challenges based on an ‘unforeseeable accident,’ and has presumed that state officials have acted ‘in a careful and humane manner.’”) (quoting Louisiana ex rel. Francis v. Resweber, 329 U.S. 459, 462, 464 (1947)); Campbell v. Wood, 18 F.3d 662, 687 (9th Cir. 1994) (en banc) (“[T]he risk of accident cannot and need not be eliminated from the execution process in order to survive constitutional review.”). Fifth, even though Tennessee has adopted a method of execution designed to eliminate rather than cause pain, even though the vast majority of executions carried out in modern times have moved progressively to this procedure, even though no state or federal court has issued a final decision invalidating the three-drug protocol under the Eighth Amendment, even though the Tennessee Supreme Court in 2005 held that the three-drug protocol did not inflict “unnecessary physical or psychological pain and suffering,” Abdur’Rahman, 181 S.W.3d at 307, and even though our court in 2006 permitted the State to execute Sedley Alley under the three-drug protocol, Alley, 181 F. App’x at 514, the State has not left it at that. In 2007, the Tennessee Governor initiated a review of the procedure to ensure “that death sentences are administered in a constitutional and appropriate manner.” Executive Order No. 43, at 1. To that end, Governor Bredesen directed the corrections commissioner to undertake “a comprehensive review of the manner in which death sentences are administered in Tennessee” and asked the commissioner to establish a revised protocol by May 2. Id. at 1–2. In response, the corrections commissioner appointed a committee to prepare a report on the administration of death sentences in Tennessee. Among other things, the committee reviewed available court decisions addressing lethal injection protocols; spoke with officials in sister States regarding their experiences in implementing lethal injection; attended an on-site presentation by the Federal Bureau of Prisons’ execution team; consulted with anesthesiologists; and held a public hearing, inviting representatives from the Tennessee Medical Association, the Tennessee Bar Association, the University College of Medicine, and the Southeastern Pharmacology Society, among others. See Report at 4–5. “The most significant issue” the committee faced was “the selection of the chemicals and dosage to be used in lethal injection executions in Tennessee.” Id. at 6. “After considerable research and consultation with medical experts,” it “retained a three-chemical protocol.” Id. While several considerations drove the committee’s decision (including weaknesses in the one-drug and two-drug options), the paramount consideration was the reality that “the three-chemical protocol has been used in almost all of the lethal injection executions that have taken place in this country, allowing Tennessee to draw upon the considerable experience of other jurisdictions in implementing the protocol.” Id. at 7. The committee also sought to improve Tennessee’s practices by “develop[ing] updated execution manuals for lethal injection . . . that incorporate best practices” from other jurisdictions. Id. at 1. Call the requirements of the Eighth Amendment what you will—avoiding the “unnecessary and wanton infliction of pain,” refraining from “deliberate indifference” to the needs of inmates or No. 07-5562 Workman v. Bredesen, et al. Page 10

keeping up with “evolving standards of decency”—they do not prohibit the adoption, implementation and refinement of a lethal-injection procedure in as comprehensive a manner as this. The efforts of the Governor and the corrections department suggest a State intent not just on satisfying the requirements of the Eighth Amendment but on far exceeding them. Attempting to fend off this conclusion, Workman maintains that the use of pancuronium bromide (the second of the three drugs used by Tennessee) must be cruel and unusual because even veterinarians refuse to use it in euthanizing animals. On reflection, however, this contention is more of a debater’s point than a legitimate attack on the three-drug protocol. In euthanizing animals, veterinarians use just one drug—a barbiturate not unlike sodium thiopental (the first drug used by Tennessee), except that the barbiturate used on animals acts more slowly. The problem with using a barbiturate alone, Tennessee determined, is that it takes too long, and no other State uses a barbiturate by itself. Report at 8 (noting that a multi-drug protocol “likely result[s] in a more rapid death” and that “to date no other state has used a [one-drug] protocol”). In its recent review of the protocol, Tennessee also considered the other option for eliminating pancuronium bromide from the procedure—use the first and third drugs. As the State explained, however, a two-drug protocol would lead to convulsions, a phenomenon the State understandably wished to avoid out of respect for the dignity of the individual and presumably out of respect for anyone, including the inmate’s family, watching the execution. Id. at 8 (Lethal injection without pancuronium bromide “would typically result in involuntary movement which might be misinterpreted as a seizure or an indication of consciousness.”). Like the vast majority of States and the Federal Government, as a result, Tennessee uses pancuronium bromide to prevent this from happening. Id. at 7 (“[P]ancuronium bromide . . . speeds the death process, prevents involuntary muscular movement that may interfere with the proper functioning of the IV equipment, and contributes to the dignity of the death process.”). That employees of the corrections department, who are not physicians, perform the execution procedure in a typical case does not change matters. For one, no one alleges that running an IV requires a doctor. For another, the State has a doctor on hand to address any problems if the trained employee cannot start the IV. Execution Procedures for Lethal Injection at 20 (A physician must “be present at the precise time of execution” and “perform the cut-down procedure should the IV Team be unable to find a vein adequate to insert the catheter”). As the State’s recent review of its procedures confirms, moreover, it requires its employees to engage in considerable training to handle executions. Report at 9. While Workman claims that the employees and doctor should be next to him during the administration of the procedure (rather than in an adjacent room), there are countervailing interests at work. As the state trial court observed in upholding this procedure: A paramount concern in an execution is security. The condemned has committed a violent act, and he is facing termination of life. Under these circumstances it is necessary to deviate from the surgical norm of physical proximity. It is necessary, for security reasons, to assure that the executioner is securely removed from the condemned. The separateness of the executioner and the syringes containing the lethal dosages, while it does decrease the executioner’s ability to monitor intake of the Pentothal, is for good reason. To make up for the separateness of the executioner, the Tennessee lethal injection method has a TV monitor in the execution room, a camera above the gurney, and the warden is located in the execution room within a foot of the condemned’s head. The warden has been trained on detecting problems such as crimping of the IV line, or failure of the injection to go into the vein. Abdur’Rahman v. Sundquist, No. 02-2236-III, Order at 9 (Tenn. Ch. Ct. 20th Dist. June 2, 2003), available at http://www.tsc.state.tn.us/OPINIONS/TSC/CapCases/Rahman/Rahman.htm (last visited No. 07-5562 Workman v. Bredesen, et al. Page 11

May 7, 2007). For exceedingly practical reasons, no State can carry out an execution in the same manner that a hospital monitors an operation. Nor, relatedly, does the Eighth Amendment require an anesthesiologist to be on hand to monitor the inmate’s consciousness during every execution. While it may well be a good practice for a State to hire an anesthesiologist for each execution (assuming one is willing to handle the task), the Constitution does not require it. The risks of pain that Workman complains about remain remote (and do not occur when the procedure is properly implemented), and no one has shown that they have occurred in Tennessee in the past. Under its lethal-injection protocol, Tennessee administers 5 grams of sodium thiopental to anesthesize the inmate. See Execution Procedures for Lethal Injection at 35. That lethal dosage represents the highest level that other States use, and it renders the inmate unconscious “nearly immediate[ly],” Abdur’Rahman, 181 S.W.3d at 308. This 5-gram dose thus reduces, if not completely eliminates, any risk that Workman would “incur constitutionally excessive pain and suffering when he is executed.” See id. at 308 (“Dr. Heath [Workman’s expert] . . . testified that a lesser dosage of two grams of sodium Pentothal would cause unconsciousness in all but ‘very rare’ cases and that a dosage of five grams would ‘almost certainly cause death.’”). The protocol also calls for certification and training requirements that reduce the risk of error in administering the drugs. Although the protocol does not contain an explicit instruction to monitor Workman’s consciousness, it does require the participation of a certified IV team and the presence of a doctor. This combination of factors suggests that there is ample recourse if the 5-gram dosage of sodium thiopental—14 times the dosage used to anesthetize hospital patients—somehow fails to render Workman unconscious. Under these circumstances, we cannot accept the district court’s conclusion that Tennessee’s lethal-injection protocol “creates a foreseeable and likely unnecessary risk that the Plaintiff will incur constitutionally excessive pain and suffering when he is executed.” Temporary Restraining Order at 3. Only one thing has changed since a panel of this court convincingly demonstrated that this challenge to the Tennessee procedure has a “small likelihood of . . . success,” Alley, 181 F. App’x at 513, and accordingly vacated a similar stay order—the State has reevaluated and improved its procedure. Then, as now, it remains the case that no state or federal court has issued a final decision invalidating the three-drug protocol, and several decisions (including the Tennessee Supreme Court’s decision in 2005) have upheld this procedure. In its defense, the district court of course had just a half day to consider all of this, which brings us to our second objection to Workman’s action—undue delay. C. While the absence of any meaningful chance of success on the merits suffices to resolve this matter, Workman faces a second problem: He waited far too long to bring this challenge—just five days before what is now the sixth execution date the Tennessee has set for him. The district court did not consider this reason for denying the motion (perhaps because it had too little time to do so and perhaps because it had only one party’s—Workman’s—brief ), even though the Supreme Court has indicated that in considering the equitable remedy of staying an execution, “a district court must consider . . . the extent to which the inmate has delayed unnecessarily in bringing the claim.” Nelson v. Campbell, 541 U.S. 637, 649–50 (2004) (emphasis added); see Tompkin, 362 F.3d at 891 (abuse of discretion occurs when district court makes a mistake of law). Workman responds that he filed this lawsuit within four days of receiving the revised lethal- injection protocol. See TRO Motion at 52–53. But as Workman acknowledges, the new protocol is only “slightly different” from the old protocol, Complaint at 3, and he does not point to any revision that makes the protocol worse, only to revisions that could have been made but were not. Workman’s problem, then, is not just that he waited until five days before his execution; his challenge would have been late even had he filed it immediately before or after the Governor set his No. 07-5562 Workman v. Bredesen, et al. Page 12

most recent execution date on January 17, 2007. Having refused to challenge the old procedure on a timely basis, he gets no purchase in claiming a right to challenge a better procedure on the eve of his execution. There is “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.” Nelson, 541 U.S. at 650. Even had Workman filed this challenge on January 17, 2007, that still would have been “too late in the day,” Hill v. McDonough, 126 S. Ct. 2096, 2104 (2006), to allow the trial and appellate courts to reach the merits of any subsequent challenge. See Jones, 2007 WL 1225393, at *3 n.2 (“[A]djudicating Jones’s [lethal-injection-protocol] claim would take much more than three months and . . . a subsequent appeal would add months, if not years, to this litigation.”) (internal quotation marks omitted); Harris v. Johnson, 376 F.3d 414, 417 (5th Cir. 2004) (“The brief window of time between the denial of certiorari and the state’s chosen execution date—in this case, four months—is an insufficient period in which to serve a complaint, conduct discovery, depose experts, and litigate the issue on the merits.”). He thus cannot revive a dilatory action when the only concrete challenges to the new procedure were features of the old procedure. Workman’s opportunities to avoid this scenario—where the only way to litigate the validity of the three-drug protocol is to stay his execution—were many. The Tennessee Supreme Court affirmed his death sentence in 1984, see State v. Workman, 667 S.W.2d 44 (Tenn. 1984), the state courts denied his petition for post-conviction relief in 1993, see Workman v. State, 868 S.W.2d 705 (Tenn. Ct. Crim. App. 1993), and we denied his initial federal habeas petition in 1998, see Workman v. Bell, 178 F.3d 759 (6th Cir. 1998); see also Neville v. Johnson, 440 F.3d 221, 222 (5th Cir. 2006) (“A challenge to a method of execution may be filed at any time after the plaintiff’s conviction has become final on direct review.”). In 1998, Tennessee prescribed lethal injection as a lawful means of execution, see Tenn. Code § 40-23-114; 1998 Tenn. Pub. Acts ch. 982, and the State adopted “a lethal injection protocol that included the use of three drugs,” the same three drugs the State currently uses, Abdur’Rahman v. Bredesen, 181 S.W.3d at 300. In 2000, the legislature deemed lethal injection the presumptive method for all executions in the State—in other words, the State will use that method of execution unless the individual affirmatively opts for electrocution. See Tenn. Code § 40-23-114; 2000 Tenn. Pub. Acts ch. 614. By 2000, Workman had completed his state and federal direct and (initial) collateral attacks on his sentence, and he faced the prospect of imminent execution by lethal injection. Nonetheless, from that year to the present, he chose not to challenge the procedure, whether in federal or state court. In April 2000, Workman came within two days of being executed by lethal injection but he did not challenge the procedure before (or after) a stay was issued. Workman v. Bell, 209 F.3d 940 (6th Cir. 2000) (en banc). In January 2001, he came within five days of execution but did not challenge the procedure before (or after) a stay was issued. Workman v. Bell, Nos. 96-6652, 00- 5367 (6th Cir. Jan. 26, 2001) (en banc). In March 2001, he came within two hours of execution but did not challenge the procedure before (or after) a stay was issued. Workman v. State, 41 S.W.3d 100, 101 (Tenn. 2001). In September 2003, he came within nine days of execution but did not challenge the procedure before (or after) an executive reprieve was granted. See Workman v. Bell, No. 03-2660 (W.D. Tenn. Sept. 15, 2003). And in September 2004, he came within 20 days of execution but did not challenge the procedure before (or after) a stay was issued. Workman v. Bell, Nos. 94-2577, 03-2660 (W.D. Tenn. Sept. 2, 2004). By any measurable standard, Workman has had ample time to challenge the procedure. While Workman may be correct that his other litigation efforts during these seven years—his state coram nobis petition and his federal Rule 60(b)(6) motion, among others—did not require a challenge to the procedure in these actions, his suggestion that he had no way to challenge the procedure simultaneously in a separate action in federal or state court is simply mistaken. Throughout this time, other Tennessee death-row inmates, though not Workman, have challenged No. 07-5562 Workman v. Bredesen, et al. Page 13

the constitutionality of the State’s lethal injection protocol. See Alley, 181 F. App’x 509; Abdur’Rahman, 181 S.W.3d 292. Nothing prevented Workman from intervening in the Abdur’Rahman case. See Tenn. R. Civ. P. 24.02. And our court told Alley in 2006 that he had been dilatory in bringing his challenge. See Alley v. Little, 186 F. App’x 604, 607 (6th Cir. 2006) (vacating district court stay and noting that challenge “was very late in coming”). Having waited until the eve of his sixth execution date to bring this challenge and having chosen to challenge the improved (but not the inferior) method of execution, Workman must face the reality that this is the kind of “dilatory” suit from which “federal courts can and should protect [the] States.” Hill, 126 S. Ct. at 2104. See Jones, 2007 WL 1225393, at *3 n.3 (noting that delay in filing lethal-injection challenge could not be justified on ground that inmate knew little about the procedure because the thrust of the challenge went to the impermissibility of the same three-drug procedure that most States use). Any balancing of hardships on this record strongly favors the State with respect to this 25- year-old sentence. “A State’s interests in finality are compelling when a federal court of appeals issues a mandate denying federal habeas relief,” Calderon v. Thompson, 523 U.S. 538, 556 (1998)—an event which in Workman’s case occurred in 1999, see Workman v. Bell, No. 96-6652 (6th Cir. Oct. 12, 1999), or 2000 at the latest. At that point, the State’s interest in finality acquired “an added moral dimension. . . . Only with real finality can the victims of crime move forward knowing the moral judgment will be carried out.” Calderon, 523 U.S. at 556. Indeed, while we need not rely on the point, it bears adding that, whether one looks to the test for determining the timeliness of a § 1983 death-penalty-procedure action adopted by the majority or the dissent in our recent decision in Cooey v. Strickland, 479 F.3d 412 (6th Cir. 2007), a challenge by Workman to the three-drug protocol in January 2007 seemingly would have been time barred under either approach. It is true, as Workman points out and as the district court pointed out, that a few courts have granted stays to litigate this question. But we have not seen a case, and Workman has not pointed us to a case, where the inmate passed on as many opportunities to challenge the three-drug protocol as Workman has and the inmate has waited so long after the completion of his first federal habeas petition to bring the action. At some point in time, the State has a right to impose a sentence—not just because the “State’s interests in finality are compelling,” but also because there is a “powerful and legitimate interest in punishing the guilty,” which attaches to “the State and the victims of crime alike.” Calderon, 523 U.S. at 556 (internal quotation marks omitted). Twenty-five years after the imposition of this sentence, that time, it seems to us, has come. IV. For these reasons, we vacate the district court’s temporary restraining order. No. 07-5562 Workman v. Bredesen, et al. Page 14

APPENDIX A No. 07-5562 Workman v. Bredesen, et al. Page 15 No. 07-5562 Workman v. Bredesen, et al. Page 16 No. 07-5562 Workman v. Bredesen, et al. Page 17 No. 07-5562 Workman v. Bredesen, et al. Page 18 No. 07-5562 Workman v. Bredesen, et al. Page 19 No. 07-5562 Workman v. Bredesen, et al. Page 20 No. 07-5562 Workman v. Bredesen, et al. Page 21 No. 07-5562 Workman v. Bredesen, et al. Page 22 No. 07-5562 Workman v. Bredesen, et al. Page 23 No. 07-5562 Workman v. Bredesen, et al. Page 24 No. 07-5562 Workman v. Bredesen, et al. Page 25 No. 07-5562 Workman v. Bredesen, et al. Page 26 No. 07-5562 Workman v. Bredesen, et al. Page 27 No. 07-5562 Workman v. Bredesen, et al. Page 28 No. 07-5562 Workman v. Bredesen, et al. Page 29

_________________ DISSENT _________________ R. GUY COLE, JR., Circuit Judge, dissenting. For the first time in a death-penalty case, to my knowledge, this Court vacates a temporary restraining order—an order that the Court is incompetent to review because it is not appealable—and in so doing clears the way for Philip Workman’s execution on May 9, 2007. Just as troubling, despite the extensive and detailed allegations Workman raises tending to show that Tennessee’s new lethal-injection protocol will subject him to pain and suffering in violation of the Eighth Amendment; despite that Workman supports his allegations with testimony from physicians familiar with lethal-injection protocols, medical studies, and evidence from recent botched executions; despite the statements from federal courts across the United States expressing deep skepticism with similar lethal-injection protocols adopted by other states; and despite the deference that an appellate court owes to the judgment of a district court, the majority concludes that Workman’s concerns are insufficiently compelling to warrant a brief five-day preservation of the status quo to determine whether his claims have merit. In the end, I simply cannot conclude that in the face of Workman’s disturbing allegations, the State’s legitimate interest in “finality” and giving effect to its criminal judgments will be irretrievably impaired by the TRO here. Indeed, the State’s interest in executing Workman “will, at worst, simply be delayed but not denied” if this Court denies the State’s motion to vacate the TRO. Skillern v. Procunier, 469 U.S. 1182, 1185 (1985) (Brennan, J. dissenting). And if Workman is ultimately successful in proving the constitutional infirmity of Tennessee’s new lethal-injection protocol, “then [the TRO] will have prevented a harm the legality of which will be open to serious question under federal law.” Id. Accordingly, I respectfully dissent. I. Appealability The majority’s opinion rests on a profound jurisdictional defect: There is no appealable order before this Court. The district court issued a temporary restraining order, not a preliminary injunction. It is well established that “[a]n order granting, denying, or dissolving a temporary restraining order is generally not appealable.” Moore’s Federal Practice § 65.41 (3d ed. 2005). TRO’s have the modest purpose of preserving the status quo to give the court time to determine whether a preliminary injunction should issue. First Tech. Safety Sys., Inc. v. Depinet, 11 F.3d 641, 650 (6th Cir. 1993). The short duration of a TRO—no more than 10 days under Rule 65(b)—is one of its chief distinctions from a preliminary injunction. Indeed, as this Court recently acknowledged, “[t]he rationale for this rule [(i.e., the non-appealability of TRO’s)] is that TRO’s are of short duration and usually terminate with a prompt ruling on a preliminary injunction, from which the losing party has an immediate right of appeal.” Ne. Ohio Coal. for the Homeless v. Blackwell, 467 F.3d 999, 1005 (6th Cir. 2006). The district court’s TRO cannot be magically transformed into a preliminary injunction, which is an appealable order, even though the State and a majority of this Court may wish it. This makes the majority’s heavy reliance on the unpublished decision in Alley v. Little, 181 F. App’x 509 (6th Cir. May 12, 2006)—which involved a preliminary injunction—entirely inapposite. True, in certain situations not applicable here, courts will treat TRO’s as appealable preliminary injunctions. For instance, if a TRO is extended beyond the 10-day limit provided for in Rule 65(b), then it may be treated as a preliminary injunction. See, e.g., Nordin v. Nutri/Sys., Inc., 897 F.2d 339 (8th Cir. 1990) (treating TRO as preliminary injunction because it had no expiration date and exceeded Rule 65(b)’s 10-day limit); Chicago United Indus., Ltd. v. City of Chicago, 445 F.3d 940 (7th Cir. 2006) (treating TRO as preliminary injunction because 20 days since its issuance had elapsed). This is not No. 07-5562 Workman v. Bredesen, et al. Page 30

an issue here because the district court’s order sets a preliminary-injunction hearing date of May 14, 2007, and specifies that the TRO will dissolve on that date. In addition, courts sometimes say that even though an order is styled a TRO, it is in substance a preliminary injunction. Ne. Ohio Coal. for the Homeless, 467 F.3d at 1005. The State contends that this is precisely the situation here because the TRO will prevent the State from carrying out its May 9 scheduled execution of Workman, and therefore cause “serious, perhaps irreparable” harm to the State’s interest in giving effect to its criminal judgments. (Defs.’ Motion To Vacate Temporary Restraining Order at 5 (quoting Ne. Ohio Coal. for the Homeless, 467 F.3d at 1005)). But that conclusion is illogical. The State’s interest is in no way irreparably harmed, or even seriously undermined. The TRO here does not interfere with the State’s conviction of Workman; it does not interfere with the State’s ultimate imposition of the death sentence; and it does not indefinitely preclude the State from executing Workman. The TRO does no more than prohibit Workman’s execution on May 9, so that the district court may determine—a mere five days later—whether a preliminary injunction should issue. I cannot conclude that the State’s interest—whether described as avoiding delay or achieving finality—is so compelling as to necessitate treating what is manifestly a TRO as a preliminary injunction. Indeed, such a conclusion cannot be right for two reasons. First, as a legal matter, the State’s contention provides grounds for converting every TRO into a preliminary injunction, thereby eviscerating the distinctions between these two procedural devices. Parties subjected to TRO’s invariably have an interest in proceeding on schedule with whatever activity they intended to undertake prior to being restrained. The essence of a TRO is to preserve the status quo; by definition, then, its purpose is to prevent from happening an event that would otherwise happen. If an interest in avoiding any delay, no matter how brief, is a legitimate consideration in determining whether a TRO is substantively a preliminary injunction, then TRO’s could always be characterized as preliminary injunctions. This is exactly the majority’s mistake: in its view, any TRO that prevents an act from taking place as scheduled “effectively operates” as a preliminary injunction and can therefore be reviewed. Second, as a practical matter, the State’s interest in “finality” is simply not the type of interest that can be irreparably or even seriously harmed by the one-week delay imposed by the district court’s TRO. Although the State’s interest in giving effect to Workman’s death sentence is certainly recognized to be a strong one under applicable case law, it is not so strong as to amount to an inviolable interest in executing Workman on May 9. As described above, the TRO does no more than give the district court five additional days in which to decide whether a preliminary injunction should issue. And if the district court denies the preliminary injunction, the State can hardly complain that the minimal delay entailed in the issuance of the TRO put it in a materially worse position. The State will be free to proceed with Workman’s execution and, under Tennessee law, it will not be required to take any action to do so, such as applying to the state supreme court for a new death warrant. The Tennessee Supreme Court will automatically set a new execution date, which could be fixed as little as seven days from the date of the court’s new execution order. Tenn. Sup. Ct. R. 12.4(E). If the district court grants Workman’s motion for a preliminary injunction, that order will be immediately appealable to this Court, and nothing would stand in the way of the State requesting expedited review of its appeal. Because I believe that there is no doubt that the district court issued a TRO and not a preliminary injunction, I would deny the State’s motion for lack of jurisdiction. II. Standard of Review: Abuse of Discretion “The district court’s decision to grant a temporary restraining order, when appealable, is reviewed by this Court for abuse of discretion.” Ne. Ohio Coalition for the Homeless, 467 F.3d at No. 07-5562 Workman v. Bredesen, et al. Page 31

1009. This Court has stated many times that the abuse-of-discretion standard is “highly deferential” to the judgment of the district court. See, e.g., United States v. Owens, 426 F.3d 800, 805 (6th Cir. 2006); Smith v. Botsford Gen. Hosp., 419 F.3d 513, 516 (6th Cir. 2005). Indeed, “[w]e will find an abuse of discretion only when we have a definite and firm conviction that the trial court committed a clear error of judgment.” Mitchell v. Boelcke, 440 F.3d 300, 303 (6th Cir. 2006). Thus, so long as the district court acted within its sound discretion, we may not reverse its judgment even if we would have decided the matter differently. EEOC v. Ky. State Police Dep’t, 80 F.3d 1086, 1100 (6th Cir. 1996). Whatever else may be said about the majority’s review of the district court’s order, it is not for an abuse of discretion. Workman put before the district court an impressive record, particularly when considering that he was only at the TRO stage, not the preliminary-injunction stage. As discussed in more detail below, Workman filed an 82-page complaint detailing extensive allegations with respect to the constitutional infirmity of Tennessee’s Revised Protocol. He further filed a 55- page memorandum in support of his motion for a TRO, supported by 48 exhibits, including, among other things, the Revised Protocol, affidavit testimony from two physicians familiar with lethal- injection protocols and their inherent risks, a recent medical study critical of lethal injection, and execution logs from two botched executions. Besides the record, the district court had before it something that this Court has not had the benefit of, namely, the parties themselves. Even though it was operating under difficult time constraints, the district court heard oral argument from the parties and was able to test their respective positions. The transcript of the oral argument shows that the district judge was engaged with the difficult questions presented in Workman’s motion and thoughtful in his assessment. (Tr. of Dist. Ct. Proceedings, Motion for TRO, May 4, 2007.) As discussed in more detail below, I simply cannot say that the district court abused its discretion in issuing the TRO. More importantly, the clear import of the majority’s holding is that virtually no defendant in Workman’s shoes—facing an imminent execution under a lethal-injunction protocol that may be unconstitutional—can make a sufficient showing to satisfy this Court that a brief five-day delay is warranted to determine whether a preliminary injunction should issue. The majority sets the bar extraordinarily high, and unnecessarily so given the limited nature of a TRO. III. The Four Traditional TRO Factors Each of the four traditional factors for equitable relief weighs in Workman’s favor. The district judge was well within his discretion to so conclude. 1. Success on the Merits The Eighth Amendment bars executions that “involve the unnecessary and wanton infliction of pain,” Gregg v. Georgia, 428 U.S. 153, 173 (1976), or that “involve torture or a lingering death,” In re Kemmler, 136 U.S. 436, 447 (1890). Whether a particular execution procedure will inflict unnecessary pain is fundamentally an inquiry regarding whether the inmate is “subject to an unnecessary risk of unconstitutional pain or suffering.” Cooper v. Rimmer, 379 F.3d 1029, 1033 (9th Cir. 2006) (emphasis added). Workman has shown the likelihood of that risk here. a. Risk of Unnecessary Pain There is no dispute that the drugs used to execute Tennessee inmates can inflict excruciating pain if not properly administered. Yet Tennessee’s Revised Protocol fails to provide for properly credentialed personnel to ensure that the drugs are properly administered. But grave problems arise even before the drugs enter the inmate. For example, if the EMT cannot access the inmate’s vein—something particularly relevant here because of Workman’s past intravenous drug use—the Revised Protocol instructs that a doctor will conduct what is known as No. 07-5562 Workman v. Bredesen, et al. Page 32

a “cut-down procedure.” (Workman’s Mem. In Support of Motion for TRO, Ex. 1, at 41, 67.) As Columbia University anesthesiologist Dr. Mark Heath explains, this is “an outdated method of achieving venous access for the administration of anesthetic drugs”; it has been “virtually completely supplanted by the ‘percutaneous’ technique,” which is “less invasive, less painful, less mutilating, faster, safer, and less expensive.” (Id., Ex. 23 (Heath Decl. at 22).) Using the cut-down method “would defy contemporary medical standards and would be in violation of any modern standard of decency.” (Id.) Tennessee’s adherence to this outdated method “would represent the gratuitous infliction of pain and mutilation to the condemned prisoner.” (Id. (emphasis added).) This explains why many states have abandoned this procedure for executions. (Id.) Once a vein is accessed—perhaps through this unnecessary, mutilating cut-down procedure—the first of the three drugs, sodium thiopental, is provided to anesthetize the inmate. Properly providing this anesthetic is, of course, crucial to ensure lack of consciousness; everyone recognizes that the next two drugs would otherwise inflict unbearable pain. But there is a risk that Workman will be conscious. Sodium thiopental is an ultrashort-acting anesthetic that is extremely sensitive to errors in administration. (Id. 14–15.) Thus, if the intended amount of sodium thiopental fails to reach the inmate’s brain, the duration of narcosis will be only brief and the inmate will reawaken during the execution process. (Id.) Yet the Revised Protocol does not require medically trained personnel to supervise or assist in the medical tasks necessary to prepare for the execution or during the execution. These critical tasks include mixing the sodium- thiopental solution; setting up the IV line and associated equipment to ensure fluids do not leak and are not misdirected (not an immaterial possibility considering that IV-line extensions extend into another room—something that would not be permitted in a medical setting) (Id. 16); finding a usable vein and properly inserting the IV line in the proper direction in the vein; and avoiding the risk that the inmate’s vein will rupture and the drugs will flow into the surrounding tissue. (Id. 17.) The President of the American Society of Anesthesiologists, writing about lethal injection, recently stated that “the only way to assure [a proper level of anesthesia] would be to have an anesthesiologist prepare and administer the drugs, carefully observe the inmate and all pertinent monitors, and finally to integrate all this information.” (Id.19 (quoting Orin F. Guidry, M.D., Message from the President. Observations Regarding Lethal Injection (June 30, 2006)).) There is no evidence that anyone on the Tennessee injection team has any training in administering anesthesia, or, if there is training, what that training might be. (Id. 20.) Most disturbingly, the Revised Protocol makes no mention of the need for effective monitoring of the inmate’s condition or whether he is anesthetized and unconscious after the IV lines are inserted. (Id.) No medical personnel are permitted to be in the execution chamber during the administration of the drugs. (Id. 20.) In other words, there is no way to determine if something went wrong and the inmate is awake. Monitoring consciousness is a regular part of the standard of care in many states for euthanizing dogs and cats. (Id.) It should also be the standard for inmates. In light of the possibility that the sodium thiopental will be improperly administered, and the inmate therefore will be improperly anesthetized, it is critical that execution personnel be able to ascertain if that occurs. But the second drug, pancuronium bromide, eliminates this possibility: pancuronium is a neuromuscular blocking agent that paralyzes the inmate’s muscles. This paralytic effect is so complete that even an anesthesiologist in a clinical setting must vigilantly monitor diagnostic indicators to assess the anesthetic effect. (Id. 10–11.) Because the drug does not affect the brain or nerves themselves, however, an unanesthetized patient remains completely conscious, and suffers slow suffocation and excruciating pain from the third drug (potassium chloride), all while appearing to be in a peaceful sleep. (See id. 11.) For this very reason, Tennesee in 2001 declared the use of pancuronium bromide or any other neuromuscular blocking agent on non- livestock animals inhumane and illegal. See Tenn. Code Ann. § 44-17-303(c); 44-17-303(j) (providing criminal sanctions for using any substance that acts as a neuromuscular blocking agent No. 07-5562 Workman v. Bredesen, et al. Page 33

when euthanizing non-livestock animals). Thus, Tennessee protects dogs and cats from the risk of excruciating pain in execution, but not death-row inmates. In this case, however, Tennessee contends that pancuronium bromide serves useful purposes: it speeds the death process, prevents involuntary muscular movement that may interfere with the functioning of the IV equipment, and contributes to the dignity of the death process. (See Workman’s Mem. In Support of Motion for TRO, Ex. 7 (Tennessee Report on the Administration of Death Sentences at 7).) In this context, these purposes are not so noble. First, although speeding the death process is generally a good idea, that is not the case where a slower death would more likely be painless. The problem arises when the inmate is conscious yet dying a torturous death, unable to scream out to alert anyone that there is a problem. That death may ultimately be quicker (because of the pancuronium bromide), but it will not satisfy any minimal constitutional standard. Second, it would also be expedient if any muscle spasms did not interfere with the IV equipment; but, again, whatever inconvenience that might cause pales in comparison to the torturous death the inmate might otherwise suffer. Finally, I assume whatever “dignity” pancuronium bromide adds to the death process comes from the perspective of the witnesses to the death—the execution certainly would look like a peaceful death. But where is the dignity when we know that under that peaceful surface could be a person (even if a convicted murderer) writhing in agony? No dignity arises from wilful blindness to possible torture. The excruciating (yet invisible) pain the final drug, potassium chloride, would cause absent proper anesthesia is not the only troubling aspect of the drug. For example, although the drug’s purpose is to cause electrical arrest of the human heart, Dr. Ramsey explains that the dosage provided for in the Revised Protocol is “wholly ineffective” to achieve this purpose. (Workman’s Mem. In Support of Motion for TRO, Ex. 18 at 2.) Thus, if an inmate is improperly anesthetized, asphyxiation because of the paralysis caused by the pancuronium bromide kills the inmate while he suffers the painful effects of the potassium chloride. These risks are unacceptable. b. Other Courts Recognize this Risk The risk of pain discussed above is not mere conjecture; substantially similar execution methods in other states have resulted in botched executions, leading those states to suspend the practice. A few months ago in Florida, for example, inmate Angel Diaz appeared not to receive an effective amount of sodium thiopental because the IV lines were improperly seated in his veins. (Workman’s Mem. In Support of Motion for TRO, Ex 21 (Florida Commission Report to Governor, at 8–9).) Observers stated that Diaz “looked like he was in a lot of pain,” was “gasping for air for 11 minutes,” was “grimacing” and “seem[ed] to speak.” (Id., Ex. 20 (Second Dose Needed to Kill Inmate, at 1).) Following this execution, Governor Bush ordered all executions stayed while a committee reviewed this execution and the lethal-injection protocols. (Id., Ex. 21, at 2.) Similar incidents suggesting a lack of proper anesthesia have occurred in California (Id., Ex. 22) (noting that inmate’s stomach and chest “heaved more than 30 times,” which suggests improper anesthetization)); North Carolina (Brown v. Beck, 2006 U.S. Dist. LEXIS 60084 (E.D.N.C April 7, 2006) (noting accounts of inmates convulsing, twitching, gagging, and choking); Ohio (Workman’s Mem. In Support of Motion for TRO, Ex. 27 (Trouble Finding Inmate’s Vein Slows Lethal Injection in Ohio (noting that after chemicals began flowing, inmate sat up several times to say “it’s not working” and asked if he could have “something by mouth to end this”; after execution team closed the curtain, witnesses reported cries of pain for five or ten minutes))); Arkansas (Id., Ex. 30 (Stoic Murderer Meets His Fate By Quiet Means (noting that inmate cried out and coughed sporadically after appearing to nod off into unconsciousness))). No. 07-5562 Workman v. Bredesen, et al. Page 34

In light of these concerns, many of these states have stayed executions pending further review of these procedures. See, e.g., Morales v. Hickman, 415 F. Supp. 2d 1037 (N.D. Cal. 2006); Nooner v. Norris, No. 06-00110 (E.D. Ark. June 26, 2006). Additionally, the United States Attorney General has agreed to a preliminary injunction for federal inmates challenging the federal lethal- injection protocols as unconstitutional. Roane v. Gonzales, No. 05-2337 (D.D.C. Feb. 16, 2007). As the court in Cooey v. Taft explained, “This Court would be remiss if it did not take note of the evidence [from other states] . . . [that] raises grave concerns about whether a condemned inmate would be sufficiently anesthetized under [this protocol] prior to and while being executed.” 430 F. Supp. 2d 702, 707 (S.D. Ohio 2006) (vacated on other grounds by Cooey v. Strickland, 479 F.3d 412 (6th Cir. March 2, 2007)). The majority raises the unremarkable point that no court has yet made a final ruling on the merits that these lethal-injection procedures violate the Eighth Amendment. This is of course not surprising considering that these lethal-injection challenges are in their infancy—the Supreme Court’s decision in Hill v. McDonough, 126 S. Ct. 2096 (2006), less than a year ago, breathed life into these claims. (See, e.g., David Stout, Justices Open Door to Lethal Injection Challenges, New York Times, June 12, 2006.) But the majority also misses the point that more and more courts have determined that there is a substantial likelihood that these procedures violate the Eighth Amendment. See, e.g., Taylor v. Crawford, 2006 U.S. Dist. LEXIS 51008 (D. Mo. 2006) (“Missouri’s current lethal injection procedure subjects condemned inmates to an unnecessary risk of unconstitutional pain and suffering. Without appropriate monitoring of the anesthesia, there is a strong argument that these executions might even be torturous.”). Ultimately, regardless of the current judicial scorecard on lethal-injection challenges, the compelling evidence remains that Workman is subjected to unnecessary risk of a torturous death and therefore has shown a substantial likelihood of success on the merits. The majority also states that the Eighth Amendment’s concern regarding evolving standards of decency “does not help Workman” because there is a consensus that lethal injection is the most humane method of execution. But everyone knows that this consensus is built entirely on the perception that the process actually is painless and dignified. The challenge here is not to proper, painless lethal injection; the challenge is to the grave risk that this particular lethal-injection procedure causes true suffering—exactly opposite of what the consensus believes the method achieves. Additionally, the majority notes that, despite what it sees as overwhelming evidence that lethal injection is indeed as humane as the consensus intends, Tennessee has—as if unprompted—taken the affirmative act of re-evaluating this apparently impeccable procedure and somehow “refined” it even more. What of course really happened here is that Tennessee recognized that numerous courts across the country were enjoining this procedure on Eighth Amendment grounds because of the real risk of a torturous death. And for that, Tennessee should be applauded. At the present time, however, there is no basis for concluding that Tennessee has sufficiently lessened this risk. Indeed, the evidence suggests the risk remains pronounced. 2. Irreparable Injury Nobody contests that Workman will suffer irreparable harm if his execution is not stayed. Worse than simply the harm of execution, however, Workman also faces substantial risk of excruciating pain in the process. 3. Harm to Others Tennessee no doubt has an interest in the finality of justice; under the law, if the execution is constitutional, it should be carried out. But the issue here is the additional harm caused by adding No. 07-5562 Workman v. Bredesen, et al. Page 35

five days to Workman’s 25 years on death row; that cannot outweigh the harm to Workman, who may be put to death through a procedure that inflicts so much pain it cannot be used on non- livestock animals. 4. The Public Interest Lethal injection has become the predominant method of execution in large part because the public perceived it to be the most humane form of execution. But the method—perhaps veiled by drug-induced paralysis—may be horrific. The public deserves assurances that this is not the case. As the Tennessee Governor has explained, “The administration of the death penalty in a constitutional and appropriate manner is a responsibility of the highest importance.” (Workman’s Mem. In Support of Motion for TRO, Ex. 3 (Governor’s Executive Order No. 43).) IV. Timeliness The majority further contends that the district court erred by failing to consider whether Workman’s filing of his § 1983 action was timely. As an initial matter, despite the State’s and the majority’s heavy reliance on the contention that Workman has unreasonably delayed in bringing his § 1983 suit, the State failed to raise this as a defense before the district court. (Tr. of Dist. Ct. Proceedings, Motion for TRO, May 4, 2007.). See Thurman v. Yellow Freight Sys., 97 F.3d 833, 835 (6th Cir. 1996) (holding that arguments not raised before the district court are waived). The majority does not contend that Workman was dilatory in challenging the Revised Protocol, nor could it. Once the Revised Protocol was made public on April 30, Workman promptly filed his administrative grievance with the Tennessee Department of Correction on May 2 and, having exhausted his administrative remedies, filed his § 1983 action in the district court on May 4. (See Workman’s Mem. In Support of Motion for TRO, Ex. 11.) Instead, the majority determines that Workman unreasonably delayed because he did not file his § 1983 suit promptly after the conclusion of his habeas proceedings, which, in the majority’s view, terminated in 2000. I do not take lightly the Supreme Court’s admonition that we must ensure that death-row inmates timely pursue available relief rather than engage in last-minute requests that could have been brought earlier. In many instances, applicable challenges can be lodged sufficiently in advance to avoid stays of already-scheduled executions. In some instances, however, such stays are necessary and proper. This is one of those instances. To begin, it would have been legally futile for Workman to bring his § 1983 challenge to Tennessee’s lethal-injection protocol at the conclusion of his habeas proceedings, assuming that the majority is right that those proceedings terminated in 2000, rather than with subsequent activity in this Court. Until the Supreme Court’s May 24, 2004 decision in Nelson v. Campbell, 541 U.S. 637, 645-47 (2004), the law of this Circuit precluded inmates from challenging lethal-injection protocols through § 1983 actions. See In re Sapp, 118 F.3d 460 (6th Cir. 1997); In re Williams, 359 F.3d 811 (6th Cir. 2004); see also Cooey v. Strickland, 479 F.3d 412 (6th Cir. 2007) (Gilman, J. dissenting) (agreeing with the district court’s holding that petitioner was not required to bring his § 1983 method-of-execution challenge prior to Nelson). Thus, had Workman filed his § 1983 complaint when the majority maintains he should have, it presumably would have been dismissed. Moreover, had Workman’s § 1983 case been pending on February 1, when Governor Bredesen rescinded the prior protocol, there is a very real possibility that it would have been dismissed at that time. The State moved to dismiss two § 1983 lethal-injection challenges pending in the Tennessee federal courts on the grounds that they were moot, in light of the Governor’s Executive Order. See Payne v. Little, No. 06-00825 (M.D. Tenn.); Harbison v. Little, No. 06-1206 (M.D. Tenn.). In Harbison, the State argued that No. 07-5562 Workman v. Bredesen, et al. Page 36

There is no lethal injection protocol currently in effect; thus, there is nothing to litigate. In light of this, the issues presented by the present action are moot, as there is no actual case or controversy, and this Court lacks jurisdiction under Article III of the United States Constitution. Harbison, Mem. In Support Of Defendants’ Motion To Dismiss (emphasis added). In Payne, the State made the identical argument that because the prior protocol had been rescinded, “[t]here is no lethal injection protocol currently in effect; thus there is nothing to litigate.” Payne, Mem. In Support Of Defendants’ Motion To Dismiss. Next, Workman’s execution was in no way imminent in 2004, when his § 1983 claims first became cognizable, because he was still litigating his conviction and death sentence and his execution was stayed for a two-year period between September 1, 2004 and October 17, 2006. Workman filed his motion for relief from the district court’s judgment denying his habeas petition on August 27, 2004. At the same time, he filed a motion for a stay. The district court granted the stay on September 1, 2004, and ordered it to remain in effect until the court had disposed of Workman’s motion for relief from judgment. On September 22, 2004, this Court declined to vacate the stay. Because the district court did not deny Workman’s motion for relief from judgment until October 17, 2006, the stay did not expire until that date. It makes no sense to say that Workman should have filed his § 1983 challenge to the lethal- injection protocol even before he knew whether he would succeed on his pending challenge in the district court to his conviction and sentence, or on his state coram nobis petition. Such a requirement forces inmates, such as Workman, to simultaneously litigate their claims that they are entitled to relief from their convictions and sentences, and that, even if not entitled to relief from their convictions and sentences, they are nonetheless entitled to relief from a state’s particular execution methodology. See Cooey, 479 F.3d at 429 (Gilman, J. dissenting). Further, such parallel litigation is inefficient and burdensome for both litigants and the courts. See id. Indeed, there is no point to requiring an inmate to file his § 1983 action to a lethal-injection protocol when success on a challenge to a conviction or sentence will moot the § 1983 case.1 Thus, where any § 1983 challenge Workman filed before the Supreme Court’s 2004 decision in Nelson would not have been cognizable under the law of this Circuit; where Workman’s execution was not imminent because it was stayed for more than two years by the district court; where Workman was actively pursuing relief that, if granted, would have mooted his § 1983 claims; where the Governor rescinded Tennessee’s lethal-injection protocol on February 1, 2007 and moved to dismiss pending lethal-injection challenges in the Tennessee federal courts on the grounds that they were moot; and where the actual lethal-injection protocol under which Workman will be executed was not established until April 30, 2007, I simply cannot conclude that Workman unreasonably delayed in bringing his § 1983 action. Two final points remain. First, the record shows that Workman diligently applied to the Tennessee Supreme Court to vacate his execution date on precisely the grounds about which he now complains, namely, that setting his execution date within just a few days of the establishment of the Revised Protocol would prevent him from challenging the constitutionality of that protocol. (See Workman’s Mem. In Support of Motion For TRO, Ex. 4.) The State opposed Workman’s motion.

1 To the extent that the State and the majority argue that the § 1983 statute-of-limitations analysis in this Court’s recent divided decision in Cooey applies here, I cannot agree. At a minimum, we should be circumspect about following Cooey where it is not yet a final judgment owing to the pendency of a petition for rehearing en banc. No. 07-5562 Workman v. Bredesen, et al. Page 37

(See id., Ex. 5.) In so doing, the State represented to the Tennessee Supreme Court that “the courts could take action later should circumstances warrant staying Workman’s execution date.” (See id., Ex. 6 (Tenn. Sup. Ct. Order).) Although not tantamount to a waiver of its rights to contest Workman’s challenge now, it might reasonably be said that insofar as Workman may be charged with “delay,” the State was complicit as well. Second, the majority and the State urge that the time has come to execute Workman because his conviction is now 25 years old and he has received multiple stays in the past. Of course, neither Congress nor the courts have seen fit to specify limitations on how long death-row inmates may be kept alive prior to their executions, or limitations on the number of stays that a defendant may be entitled to. Thus, I am simply not persuaded that, in and of themselves, Workman’s 25 years on death row and his past stays have any bearing on whether the district court properly issued a TRO here. This is not to minimize the interests of the State and the victims in seeing Workman’s sentence carried out. But we must remember that while justice should be prompt, “prompt injustice is not the answer.” Haitian Refugee Ctr. v. Smith, 676 F.2d 1023, 1040 n.43 (5th Cir. 1982) (emphasis added); see also Evans v. Muncy, 498 U.S. 927, 930 (1990) (Marshall, J., dissenting) (explaining that the state’s interest in finality does not permit a court to “look the other way when late-arriving evidence upsets its determination that a particular defendant can lawfully be executed”). V. Conclusion It is unfortunate that the majority chooses to foreclose the limited inquiry—an inquiry that does no more than preserve the status quo for a mere five days—that could very well confirm its conclusion that Philip Workman has nothing to fear from Tennessee’s new lethal-injection protocol. The majority’s reasons for doing so are unconvincing. Whatever harm the State might sustain by the issuance of the TRO—if indeed “harm” it can be called—pales next to the damage done to our Constitution by allowing a single defendant to perish under a method of execution that violates his rights. Of course, at this most preliminary of stages we cannot know whether Workman’s allegations ultimately will prove meritorious. Our task is only to ascertain whether the allegations he raises are sufficiently disturbing to warrant a brief and temporary halt to his execution. Considering the record in this case and the deference owed to the district court, I would deny the State’s motion to vacate the TRO. Accordingly, I respectfully dissent.