WILLIAM PRYOR, Circuit Judge:

This appeal presents the questions whether due process forbids giving a jury's findings of negligence and strict liability in a class action against cigarette manufacturers preclusive effect in a later individual suit by a class member and, if not, whether federal law preempts the jury's findings. Florida smokers and their survivors filed a class action against several tobacco companies, and after a yearlong trial designed to answer common questions concerning the companies' tortious conduct against all members of the class, a jury found that each company had breached its duty of care and sold defective cigarettes. The Florida Supreme Court upheld the jury verdicts of negligence and strict liability in Engle v. Liggett Group, Inc., 945 So.2d 1246 (Fla. 2006) (Engle III), and decertified the class to allow individual actions about the remaining issues of specific causation, damages, and comparative fault. The Engle decision made clear that the jury findings of negligence and strict liability had preclusive effect in the later individual actions, and the Florida Supreme Court reaffirmed that ruling in Philip Morris USA, Inc. v. Douglas, 110 So.3d 419 (Fla. 2013). R.J. Reynolds Tobacco Company and Philip Morris USA Inc. challenge a jury verdict against them in one of those individual actions in the district court. They argue that giving the Engle findings preclusive effect violates the Due Process Clauses, U.S. Const. Amends. V, XIV, and they urge us to overrule our decision to the contrary in Walker v. R.J. Reynolds Tobacco Co., 734 F.3d 1278 (11th Cir. 2013). They argue, in the alternative, that federal law preempts giving preclusive effect to the Engle findings of negligence and strict liability. Because we reaffirm our holding in Walker and conclude that federal law does not preempt the Engle jury findings, we affirm the judgments against R.J. Reynolds and Philip Morris.

I. BACKGROUND

In 1994, six individuals filed a putative class action in Florida court against the major domestic cigarette manufacturers, including R.J. Reynolds and Philip Morris, and two tobacco industry organizations. Id. at 1281. They alleged claims of strict liability, negligence, breach of express warranty, breach of implied warranty, fraud, conspiracy to commit fraud, and intentional infliction of emotional distress. Id. The strict liability count alleged that the companies manufactured "cigarettes containing nicotine," "manufactured their defective tobacco products by manipulating the levels of nicotine so as to addict the consuming public," "failed to design, manufacture, distribute and sell a safer alternative cigarette that would not addict smokers," and "failed to warn" members of the class of the dangers. The negligence count alleged that the companies "breached their duty of reasonable care" through several "acts and omissions," including the "failure to design and manufacture products that were not addictive," the "failure to ... adequately or sufficiently reduce or remove the level of nicotine in cigarettes," and the "failure to warn the smoking consumers of the addictive nature of nicotine." A Florida district court of appeal approved the certification of the following class: all Florida citizens and residents, "and their survivors, who have suffered, presently suffer or have died from diseases and medical conditions caused by the addiction to cigarettes that contain nicotine." R.J. Reynolds Tobacco Co. v. Engle, 672 So.2d 39, 40-42 (Fla. Dist. Ct. App. 1996) (Engle I).

The trial court in Engle divided the proceedings in three phases. Walker, 734 F.3d at 1281. In Phase I, a jury "decide[d] issues common to the entire class, including general causation, the Engle defendants' common liability to the class members..., and the class's entitlement to punitive damages." Douglas, 110 So.3d at 422. Phase I was a year-long trial on "common issues relating exclusively to defendants' conduct and the general health effects of smoking." Liggett Grp. Inc. v. Engle, 853 So.2d 434, 441 (Fla. Dist. Ct. App. 2003) (Engle II). Phase I required "hundreds of witnesses, thousands of documents and exhibits, and tens of thousands of pages of testimony." Douglas, 110 So.3d at 431. In Phase II, the jury determined the liability of the tobacco companies to three class representatives, awarded them compensatory damages, and fixed the amount of class-wide punitive damages. Walker, 734 F.3d at 1281. The trial court planned to have new juries decide specific causation and damages for the remaining class members in Phase III. Id.

In his opening statement in Phase I, the plaintiffs' attorney stated, "The evidence will show, ladies and gentlemen, that there is no dispute or controversy in the medical and scientific communities but that cigarette smoking causes lung cancer, heart disease, chronic obstructive pulmonary disease, emphysema and many other diseases." He stated that "the evidence will establish overwhelmingly" that "[n]icotine is addictive." And he explained that the tobacco companies "have the technology to make a safer cigarette" but not one that is profitable. He also stated that "the evidence will show that the tobacco companies have so successfully misled the American people that many highly intelligent people, in 1998, are confused."

The smokers presented a substantial body of evidence that all of the cigarettes manufactured by the named defendants contained carcinogens that cause disease, including cancer and heart disease, and that nicotine addicts smokers. Douglas, 110 So.3d at 423. They presented evidence that the tobacco companies "failed to address the health effects and addictive nature of cigarettes, manipulated nicotine levels to make cigarettes more addictive, and concealed information about the dangers of smoking." Id. For example, Dr. Julius Richmond, a former Surgeon General of the United States and professor at the Harvard Medical School, testified that cigarettes contain carcinogens and that cigarettes cause pulmonary disease, emphysema, lung cancer, heart disease, and bladder disease. Dr. Ronald Davis, a former director of the Office on Smoking and Health and former medical director for the Michigan Department of Public Health, testified similarly that cigarette smoking is addictive and that those who smoke have a heightened risk of stroke, emphysema, cancer, and heart disease. Dr. David Burns, a professor of medicine at the University of California, San Diego, School of Medicine, with a specialty in pulmonary and critical care medicine, testified that nicotine is addictive and that cigarette smoking causes cancers, lung disease, and heart disease. He was an associate scientific editor of a 1981 Surgeon General's Report, and he explained that "the purpose of the report was to make it very clear to the public that there is no safe cigarette and there is no safe level of consumption." He testified, "[W]ith the exception of the tobacco industry, no other scientific group in the last 30 years has reviewed this evidence and reached a conclusion other than that cigarette smoking causes disease." Dr. John Holbrook, professor of medicine at the University of Utah School of Medicine, who is board certified in the field of internal medicine, testified that, in his experience, the tobacco industry "attempted to confound and obfuscate science" in its funding of medical research. Dr. W. Jarrard Goodwin, a professor at the University of Miami School of Medicine, with a specialty in otolaryngology, testified that smoking causes cancer of the mouth, larynx, and pharynx. Dr. Edward Staples, director of the artificial heart program at the University of Florida, testified that cigarette smoking causes emphysema, lung cancer, coronary artery disease, and atherosclerosis. Dr. Neal Benowitz, a doctor at San Francisco General Hospital and professor of medicine, psychiatry, and biopharmaceutical sciences at the University of California in San Francisco, testified that 90 percent of individuals begin smoking before the age of 20 and, within two or three years, those young people will become addicted to nicotine. He stated that tobacco companies could reduce the level of nicotine in cigarettes to nonaddictive quantities. Some of the evidence of design defects applied only to some brands of cigarettes. For example, the smokers presented evidence that people who smoke light cigarettes tend to smoke more and inhale more deeply. But the common thrust of the smokers' evidence was that all of the companies' cigarettes cause disease and addict smokers.

The tobacco companies put on evidence to defend themselves against the several theories of liability. For example, the companies repeatedly challenged the evidence that cigarette smoking causes disease. Dr. George Hensley, a former professor at the University of Miami School of Medicine with a specialty in pathology, testified that smoking does not cause pancreatic cancer. Dr. Hugh Gilmore, a cardiology professor at the University of Miami School of Medicine, testified that smoking is not a risk factor for the development of aortic aneurysms or congestive heart failure. And Dr. Alden Cockburn, a urologist and a clinical professor at the University of South Florida, testified that smoking is a risk factor for bladder cancer but was not definitively proven to be a cause of bladder cancer.

In closing argument, the smokers' attorney explained that "[t]he common issue trial has addressed the conduct of the tobacco industry." He recounted some of the expert testimony. He argued, without focusing on any specific brand or manufacturer of cigarettes, that scientists agree that nicotine is addicting, and he argued that there is no scientific debate as to whether cigarette smoking causes certain diseases, including cancer and heart disease. He said, "None of them qualified their answer one iota. Does cigarette smoking cause these diseases? Yes, yes, yes. Clear, crisp and definitive." He also referred the jury to a collection of documents that discussed how the companies manipulated nicotine levels. He mentioned different methods of manipulating nicotine levels but not different brands.

In closing argument, the tobacco companies' attorneys responded to the smokers' many arguments. The companies contended that cigarettes are not proven to be addictive. They maintained that smokers can quit and that nicotine is a "far cry from heroin or cocaine." And the companies argued that they have tried to make cigarettes safer. They argued that they have not "spiked" cigarettes with nicotine but have reduced the level of nicotine in some cigarettes.

The trial court instructed the jury in Phase I about the claim of strict liability and negligence without regard to specific brands of cigarettes. For the claim of strict liability, the trial court explained that "the issues are whether one or more of the defendants designed, manufactured and marketed cigarettes which were defective and unreasonably dangerous to smokers." For the claim of negligence, the trial court instructed the jury as follows:

The verdict form included a series of yes-or-no questions. The tobacco companies requested a more detailed verdict form, in which the jury would be asked to identify "specific defects and tortious actions," but the trial court rejected that proposal. Id. The jury returned its verdict after eight days of deliberation. The first question on the verdict form asked whether smoking cigarettes causes a list of enumerated diseases and medical conditions. The jury answered "yes" for 20 specific diseases, including various forms of cancer. The second question asked whether "cigarettes that contain nicotine [are] addictive or dependence producing." The jury answered "yes." The verdict form then contained nine questions about the conduct of each tobacco company. One of the nine questions asked the jury to decide whether each tobacco company was strictly liable. It asked if the tobacco company "place[d] cigarettes on the market that were defective and unreasonably dangerous." Another question asked if each tobacco company was negligent. It asked if the tobacco company "failed to exercise the degree of care which a reasonable cigarette manufacturer would exercise under like circumstances." The jury answered "yes" to each of these nine questions for each tobacco company. The last question on the verdict form asked the jury whether the actions of the tobacco companies entitled the class to punitive damages, and the jury answered "yes" for each tobacco company.

The trial court denied the tobacco companies' motion for directed verdict. Id. Regarding strict liability, the court ruled that the evidence supported a finding that all of the tobacco companies' cigarettes were defective even if some of the cigarettes had brand-specific dangers:

Engle v. R.J. Reynolds Tobacco, 2000 WL 33534572, at *2 (Fla. Cir. Ct. 2000). Regarding negligence, the court ruled that the evidence supported a finding that the tobacco companies were negligent in producing and selling all of their cigarettes:

Id. at *4.

In Phase II, the same jury determined that the tobacco companies were liable to the three class representatives and awarded them compensatory damages totaling $12.7 million. Walker, 734 F.3d at 1282. The jury awarded punitive damages of $145 billion to the class. Id. The tobacco companies filed an interlocutory appeal of the judgments in Phases I and II. Id.

The Florida Supreme Court approved in part and vacated in part the jury verdicts. Engle III, 945 So.2d at 1254. The Florida Supreme Court concluded that the trial court did not abuse its discretion in certifying the class for purposes of Phase I and II. Id. at 1267. But the court decertified the class for Phase III "because individualized issues such as legal causation, comparative fault, and damages predominate." Id. at 1268. The Florida Supreme Court "retain[ed]" the findings of liability by the jury from Phase I "other than those on the fraud and intentional infliction of emotion[al] distress claims, which involved highly individualized determinations, and the finding on entitlement to punitive damages questions, which was premature." Id. at 1269. The court explained, "Class members can choose to initiate individual damages actions," and those retained findings, which include the findings that the companies acted negligently and that they sold defective products, "will have res judicata effect in those trials." Id. The court affirmed the damages award in favor of two of the class representatives and vacated the judgment in favor of the third class representative because the statute of limitations barred his claims. Id. at 1276. The court vacated the award of punitive damages. Id. at 1262-65.

After members of the Engle class filed thousands of individual actions in state and federal courts, these courts had to determine the extent to which the smokers could rely on the approved findings from Phase I to establish certain elements of their claims. Walker, 734 F.3d at 1283. In Brown v. R.J. Reynolds Tobacco Co., 611 F.3d 1324 (11th Cir. 2010), we stated that, under Florida law, courts should give preclusive effect to the findings only to the extent that the smoker can "show with a `reasonable degree of certainty' that the specific factual issue was determined in [his] favor." Id. at 1335 (quoting Seaboard Coast Line R. Co. v. Indus. Contracting Co., 260 So.2d 860, 865 (Fla. Dist. Ct. App. 1972)). We remanded to the district court to make that determination after considering the "entire trial record." Id. But several of the Florida district courts of appeal disagreed with our decision that a member of the Engle class had to establish from the trial record that an issue was actually decided. These district courts of appeal all held that the Phase I findings established the duty and breach elements of the smokers' claims, though they disagreed about how the smokers would prove causation in individual cases. See Philip Morris USA, Inc. v. Douglas, 83 So.3d 1002, 1010 (Fla. Dist. Ct. App. 2012); R.J. Reynolds Tobacco Co. v. Brown, 70 So.3d 707, 715-16 (Fla. Dist. Ct. App. 2011); R.J. Reynolds Tobacco Co. v. Martin, 53 So.3d 1060, 1066-70 (Fla. Dist. Ct. App. 2010).

In Douglas, the Florida Supreme Court ruled that the approved findings from Phase I established common elements of the claims of Engle class members. 110 So.3d at 428-30. The court explained that, although the evidence submitted during Phase I included both general and brand-specific defects, "the class action jury was not asked to find brand-specific defects in the Engle defendants' cigarettes." Id. at 423. The jury was asked to determine "all common liability issues," and it heard evidence that the tobacco companies' cigarettes were "defective because they are addictive and cause disease." Id. The court explained that the approved findings concerned conduct that "is common to all class members and will not change from case to case" and that "the approved Phase I findings are specific enough" to establish some elements of the smokers' claims. Id. at 428. That is, the jury findings "conclusively establish" that the tobacco companies manufactured defective products and that the companies failed to exercise the degree of care of a reasonable person. Id. at 430. And the jury findings establish general causation. Id. at 428. Going forward, "to prevail on either strict liability or negligence Engle claims, individual plaintiffs must establish (i) membership in the Engle class; (ii) individual causation, i.e., that addiction to smoking the Engle defendants' cigarettes containing nicotine was a legal cause of the injuries alleged; and (iii) damages." Id. at 430.

The Florida Supreme Court then held that giving preclusive effect to the approved findings from Phase I did not violate the right to due process of the tobacco companies. Id. The companies had argued that "accepting the Phase I findings as res judicata violates their due process rights because it is not clear from the Phase I verdict which theories of liability the Engle jury actually decided to reach those findings." Id. The Douglas court concluded that the tobacco companies had notice and an opportunity to be heard and that the Engle proceedings did not arbitrarily deprive them of property. Id. at 431. It explained that "the Phase I verdict against the Engle defendants resolved all elements of the claims that had anything to do with the Engle defendants' cigarettes or their conduct." Id. at 432.

The Douglas court stated, "[T]he defendants' due process argument is an attack on our decision in Engle to give the Phase I findings res judicata — as opposed to issue preclusion — effect in class members' individual damages actions." Id. The Douglas court explained that, when it gave "res judicata effect" to the Phase I approved findings, Engle III, 945 So.2d at 1269, it meant claim preclusion, not issue preclusion. Douglas, 110 So.3d at 432. The Douglas court stated that claim preclusion prevents the same parties from relitigating the same cause of action. Id. Issue preclusion prevents the parties from relitigating "the same issues that were litigated and actually decided in a second suit involving a different cause of action." Id. at 433. The Douglas court ruled that the individual Engle actions involved the same causes of action. Id. The Douglas court stated, "[T]o decide here that we really meant issue preclusion even though we said res judicata in Engle would effectively make the Phase I findings regarding the Engle defendants' conduct useless in individual actions." Id. And the Douglas court concluded that the tobacco companies "do not have the right to have issue preclusion, as opposed to res judicata, apply to the Phase I findings." Id. at 435.

In Walker, we held that giving res judicata effect to the findings of the jury in Engle did not violate the rights of the tobacco companies to due process. Walker, 734 F.3d at 1280-81. R.J. Reynolds had appealed the jury verdicts in favor of two smokers after the district courts instructed the juries that R.J. Reynolds sold defective cigarettes and was negligent. Id. at 1286. We explained that we were obligated to give "full faith and credit to the decision in Engle, as interpreted in Douglas," unless it "would arbitrarily deprive R.J. Reynolds of its property without due process of law." Id. at 1287. We stated that no court "has ever held that due process requires application of the federal common law of issue preclusion," and we did not decide whether it does. Id. at 1289. We concluded that, even if due process requires that an issue be actually decided, the Florida Supreme Court ruled in Douglas that the approved findings from Phase I concerned conduct that is common to all class members and established negligence and defect elements of the class members' claims. Id. We concluded that the "actually decided" requirement was satisfied and that it is "no concern of ours" what the Florida Supreme Court calls the "relevant doctrine." Id.

In this appeal, R.J. Reynolds and Philip Morris challenge a jury verdict in favor of Earl Graham, as personal representative of the estate of his deceased wife, Faye Graham, a member of the Engle class. Mr. Graham filed an individual Engle action in the district court against R.J. Reynolds, Philip Morris, and other defendants later dismissed. He alleged that his wife developed lung cancer and died because of her addiction to cigarettes manufactured by R.J. Reynolds and Philip Morris. He asserted claims of strict liability, breach of warranty, negligence, fraudulent concealment, and conspiracy to fraudulently conceal.

Under the Engle framework articulated in Douglas, the jury was not asked to find that the cigarettes Faye Graham smoked were defective or that the tobacco companies were negligent. Graham v. R.J. Reynolds Tobacco Co., 782 F.3d 1261, 1273 (11th Cir. 2015), reh'g en banc granted, op. vacated, 811 F.3d 434 (11th Cir. 2016). The district court treated those findings as having already been established. Id. For the claims of negligence and strict liability, the jury was asked to determine only whether Faye Graham was a member of the Engle class and whether smoking cigarettes manufactured by R.J. Reynolds or Philip Morris "was a legal cause" of Faye Graham's injuries. Id. The district court instructed the jury that, to find legal causation, Graham's addiction to cigarettes must have "directly and in natural and continuous sequence produced or contributed substantially to producing" her injuries.

The jury found for Graham on the claims of strict liability and negligence. Id. The jury awarded Graham $2.75 million in damages and determined that Faye Graham was 70 percent at fault, R.J. Reynolds was 20 percent at fault, and Philip Morris was 10 percent at fault. Id. at 1273-74. The district court entered judgment against R.J. Reynolds for $550,000 and against Philip Morris for $275,000. Id. at 1274. The district court denied the tobacco companies' motion for judgment as a matter of law. Id. Theresa Graham later replaced Earl Graham as personal representative of the estate.

A panel of this Circuit reversed the judgment of the district court. Id. at 1285. The panel held that the Engle findings of strict liability and negligence are preempted by federal law. Id. We later granted the petition for rehearing en banc filed by Graham and vacated the panel opinion. Graham, 811 F.3d at 434-35. In addition to briefing the preemption issue, we allowed the parties to brief whether giving effect to the jury's findings in Engle would "violate the tobacco companies' rights under the Due Process Clause of the Fourteenth Amendment to the United States Constitution notwithstanding the panel's holding in Walker." The Florida Supreme Court has since ruled that federal law does not preempt "state tort" actions against the tobacco companies and that, even if federal law preempted a ban on the sale of cigarettes, the Engle Phase I findings do "not amount to ... a ban" that might conflict with federal law. R.J. Reynolds Tobacco Co. v. Marotta, No. SC16-218, 214 So.3d 590, 601, 2017 WL 1282111, at *9 (Fla. Apr. 6, 2017).

II. STANDARD OF REVIEW

We review de novo the denial of a motion for judgment as a matter of law. Mendoza v. Borden, Inc., 195 F.3d 1238, 1244 (11th Cir. 1999) (en banc). We also review de novo questions of constitutional law, Nichols v. Hopper, 173 F.3d 820, 822 (11th Cir. 1999), and whether federal law preempts a state law claim, Atwater v. Nat'l Football League Players Ass'n, 626 F.3d 1170, 1179 (11th Cir. 2010).

III. DISCUSSION

We divide our discussion in two parts. First, we explain why giving full faith and credit to the Engle jury findings of negligence and strict liability does not deprive R.J. Reynolds and Philip Morris of property without due process of law. Second, we conclude that the Engle jury findings of negligence and strict liability are not preempted by federal law.

A. Giving Preclusive Effect to the Negligence and Strict Liability Findings Does Not Violate Due Process.

The Full Faith and Credit Act, 28 U.S.C. § 1738, requires federal courts to "give preclusive effect to a state court judgment to the same extent as would courts of the state in which the judgment was entered," Kahn v. Smith Barney Shearson Inc., 115 F.3d 930, 933 (11th Cir. 1997) (quoting Battle v. Liberty Nat'l Life Ins. Co., 877 F.2d 877, 882 (11th Cir. 1989)), subject to the requirements of the Due Process Clause, see Kremer v. Chem. Const. Corp., 456 U.S. 461, 481, 102 S.Ct. 1883, 72 L.Ed.2d 262 (1982). R.J. Reynolds and Philip Morris argue that the Due Process Clause mandates that an issue be actually decided in one case before it is given preclusive effect in another. They argue that relying on the approved jury findings in individual actions by Engle members is an application of issue preclusion and that the Florida courts did not actually decide issues of strict liability and negligence for all class members. They argue that by abandoning the "actually decided" requirement, the Florida courts abrogated a fundamental protection against arbitrary deprivations of property in violation of the Due Process Clause. See Honda Motor Co. v. Oberg, 512 U.S. 415, 430, 114 S.Ct. 2331, 129 L.Ed.2d 336 (1994).

We need not determine whether the Due Process Clause requires that an issue be actually decided in an earlier case before the judgment from that case is given preclusive effect on that issue. We will assume, without deciding, that the "actually decided" requirement is a fundamental requirement of due process under Fayerweather v. Ritch, 195 U.S. 276, 25 S.Ct. 58, 49 S.Ct. 193 (1904). Even with that assumption, no violation of due process occurred when the district court gave the Engle findings preclusive effect. Based on our review of the Engle proceedings, we are satisfied that the Engle jury actually decided common elements of the negligence and strict liability of R.J. Reynolds and Philip Morris.

The Florida Supreme Court made clear in Douglas that the Engle jury decided common elements of the negligence and strict liability of the tobacco companies for all class members. And for that reason, the Florida Supreme Court explained that the findings were binding in individual Engle actions. It stated, "Because these findings go to the defendants' underlying conduct, which is common to all class members and will not change from case to case, we held that these approved `Phase I common core findings ... will have res judicata effect' in class members' `individual damages actions.'" Douglas, 110 So.3d at 428 (alteration in original) (quoting Engle III, 945 So.2d at 1269).

The Florida Supreme Court rejected the same argument that R.J. Reynolds and Philip Morris make here about what the Engle jury decided. R.J. Reynolds and Philip Morris asserted that some of the evidence presented at the Engle trial applied to specific brands of cigarettes. They argued that, although the Engle jury found that the tobacco companies "place[d] cigarettes on the market that were defective and unreasonably dangerous," the jury did not necessarily find that all cigarettes the defendants placed on the market were defective and unreasonably dangerous. The Florida Supreme Court rejected this argument and stated that "this Court in Engle necessarily decided that the approved Phase I findings" are "specific enough to establish a causal link between their conduct and damages to individual plaintiffs who prove injuries caused by addiction to smoking the Engle defendants' cigarettes." Id. That is, the Phase I findings establish the causal link between the tobacco companies' conduct and the class members' injuries because the companies acted wrongfully toward all of the class members. Whether that conduct was the legal cause of the individual class members' injuries, and whether the individual class members were entitled to damages, was left for later individual trials.

After reviewing the Engle trial record, we are satisfied that the Florida Supreme Court determined that the Engle jury found the common elements of negligence and strict liability against Philip Morris and R.J. Reynolds. Both companies admit that the smokers presented common "proof that the Engle defendants' cigarettes were defective because they are addictive and cause disease" in addition to brand-specific evidence. Id. at 423. In two days of closing arguments, the smokers' attorneys recounted the ample body of evidence that smoking cigarettes causes disease without focusing on the differences in the designs of various brands. The trial court instructed the jury to "determine `all common liability issues' for the class concerning `the conduct of the tobacco industry.'" Id. Moreover, the jury's answers on the verdict form, when read together with the entire record, were consistent with the general theories that the tobacco companies' cigarettes are defective and the sale of their cigarettes is negligent because all of those cigarettes cause disease and are addictive.

The first two questions on the verdict form are most naturally read to apply to all cigarettes manufactured by the tobacco companies. Question 1 asked whether "smoking cigarettes cause one or more of the following diseases or medical conditions." The jury answered "yes" to 20 of 23 diseases. This question does not admit of any limitation, nor did the accompanying jury instruction, and its natural interpretation is that it was asking about all cigarettes manufactured by the tobacco companies, not just some. Similarly, question 2 asked whether "cigarettes that contain nicotine [are] addictive or dependence producing," and the jury answered "yes." The evidence at trial was that nicotine, and not some other ingredient, made cigarettes addictive. In closing arguments, the tobacco companies' counsel told the jury that the question should be understood to inquire whether "all cigarettes that contain nicotine [are] addictive or dependence-producing," not whether there is "one cigarette or a brand of cigarettes or two brands of cigarettes" that are addictive.

The strict liability and negligence questions presented to the jury used the same unmodified noun — "cigarettes" — that was used to refer to all cigarettes manufactured by the tobacco companies in questions 1 and 2. The strict liability interrogatory asked whether "one or more of the defendant tobacco companies place[d] cigarettes on the market that were defective and unreasonably dangerous," and the negligence interrogatory inquired whether the smokers had "proven that one or more of the defendant tobacco companies failed to exercise the degree of care which a reasonable cigarette manufacturer would exercise under like circumstances." The jury answered "yes" to both questions for R.J. Reynolds and Philip Morris. When asked about strict liability, the jury found that R.J. Reynolds and Philip Morris had sold defective cigarettes "both before and after July 1, 1974," and, with respect to the negligence claim, that they had acted negligently by selling, manufacturing, and distributing cigarettes "both before and after July 1, 1969." That the jury found that these tobacco companies' tortious conduct swept across both time periods is consistent with a general theory of liability that applied to all their cigarettes.

After the jury returned a verdict in favor of the class on all counts, the trial court ruled that there was sufficient evidence to support those verdicts, including negligence and strict liability, and cited evidence that applied to all of the cigarettes made by the tobacco companies. For example, it stated, "The evidence more than sufficiently proved that nicotine is an addictive substance which when combined with other deleterious properties, made the cigarette unreasonably dangerous." Engle, 2000 WL 33534572, at *2. The only way to make sense of these proceedings is that the Florida courts determined that the Engle jury actually decided issues common to the class, and the district court did not abrogate a protection against arbitrary deprivations of property in affording the Phase I jury's findings preclusive effect in Graham's case.

R.J. Reynolds and Philip Morris argue that if the Florida Supreme Court had determined that the Engle jury actually decided common elements of negligence and strict liability for all class members, it would not have used the term "claim preclusion" in Douglas to refer to the preclusive effect of the jury findings and thereby evade the "actually decided" requirement, but we disagree. The Florida Supreme Court explained that issue preclusion applies in actions involving different causes of action and claim preclusion applies in actions involving the same causes of action. Douglas, 110 So.3d at 432-33. And in explaining the differences between claim preclusion and issue preclusion, the Florida Supreme Court reiterated that the Engle jury made findings about the tobacco companies' conduct that applied to all class members. It said, "No matter the wording of the findings on the Phase I verdict form, the jury considered and determined specific matters related to the [Engle] defendants' conduct. Because the findings are common to all class members, [individual plaintiffs are] entitled to rely on them...." Id. at 433 (alterations in original) (quoting Martin, 53 So.3d at 1067).

The terminology employed by the Florida Supreme Court was unorthodox, but "[i]n determining what is due process of law, regard must be had to substance, not to form." Fayerweather, 195 U.S. at 297, 25 S.Ct. 58. The Supreme Court of the United States has acknowledged that "[t]he preclusive effects of former adjudication are discussed in varying and, at times, seemingly conflicting terminology, attributable to the evolution of preclusion concepts over the years." Migra v. Warren City Sch. Dist. Bd. of Educ., 465 U.S. 75, 77 n.1, 104 S.Ct. 892, 79 L.Ed.2d 56 (1984). As long as the state proceedings "satisf[ied] the minimum procedural requirements" of due process, Kremer, 456 U.S. at 481, 102 S.Ct. 1883, what the Florida Supreme Court "calls the relevant doctrine... is no concern of ours," Walker, 734 F.3d at 1289.

Apart from their argument that the jury did not actually decide common issues of negligence and strict liability, R.J. Reynolds and Philip Morris do not deny that they were afforded due process. That is, they do not contend that they were denied notice or an opportunity to be heard, the central features of due process. See Fuentes v. Shevin, 407 U.S. 67, 80, 92 S.Ct. 1983, 32 L.Ed.2d 556 (1972). The Florida courts provided them notice that the jury findings would establish the "conduct elements of the class's claims." Douglas, 110 So.3d at 429. And the year-long trial provided them "a full and fair opportunity to litigate the issues of common liability in Phase I." Walker, 734 F.3d at 1288. Both tobacco companies seized that opportunity, presenting "testimony that cigarettes were not addictive and were not proven to cause disease and that they had designed the safest cigarette possible." Douglas, 110 So.3d at 423. And they continue to contest liability in individual actions by class members, in which new juries determine issues of individual causation, apportionment of fault, and damages. Id. at 430; Engle III, 945 So.2d at 1254.

The Due Process Clause does not require a state to follow the federal common law of res judicata and collateral estoppel. "State courts are generally free to develop their own rules for protecting against the relitigation of common issues or the piecemeal resolution of disputes." Richards v. Jefferson Cty., 517 U.S. 793, 797, 116 S.Ct. 1761, 135 L.Ed.2d 76 (1996). For example, a state might allow offensive, non-mutual collateral estoppel. E.g., In re Owens, 125 Ill.2d 390, 126 Ill.Dec. 563, 532 N.E.2d 248, 252 (1988). And courts, both state and federal, frequently manage class actions by splitting them into separate phases. See generally William B. Rubenstein, Newberg on Class Actions §§ 10.6, 11.3 (5th ed.). Engle is not the first time that "a defendant's common liability [was] established through a class action and given binding effect in subsequent individual damages actions." Douglas, 110 So.3d at 429 (collecting cases); see also Brown v. Electrolux Home Prods., Inc., 817 F.3d 1225, 1239 (11th Cir. 2016) (discussing several "tools to decide individual damages" in a class action, including "(1) bifurcating liability and damage trials with the same or different juries; (2) appointing a magistrate judge or special master to preside over individual damages proceedings; [and] (3) decertifying the class after the liability trial and providing notice to class members concerning how they may proceed to prove damages" (quoting In re Visa Check/MasterMoney Antitrust Litig., 280 F.3d 124, 141 (2d Cir. 2001))). The Due Process Clause requires only that the application of principles of res judicata by a state affords the parties notice and an opportunity to be heard so as to avoid an arbitrary deprivation of property. Fuentes, 407 U.S. at 80, 92 S.Ct. 1983.

We recognize that the Engle Court defined a novel notion of res judicata, but we cannot say that the substance of that doctrine or its application in these trials was so unfair as to violate the constitutional guarantee of due process. "The very nature of due process negates any concept of inflexible procedures universally applicable to every imaginable situation," Cafeteria & Rest. Workers Union, Local 473, AFL-CIO v. McElroy, 367 U.S. 886, 895, 81 S.Ct. 1743, 6 L.Ed.2d 1230 (1961), and our review of the record establishes that the tobacco companies had notice that the Engle trial involved common evidence and theories of negligence and strict liability that applied to all cigarettes manufactured by all tobacco companies and sold to all members of the class during the relevant periods. The tobacco companies were given an opportunity to be heard on the common theories in a year-long trial followed by an appeal to the Florida Supreme Court and later individual trials and appeals on the remaining issues of proximate causation, comparative fault, and damages. See Engle III, 945 So.2d at 1254-56.

Contrary to the dissent's view, see Dissenting Op. of Tjoflat, J., at 1213-14, no tobacco company can be held liable to any smoker without proof at trial that the smoker belongs to the Engle class, that she smoked cigarettes manufactured by the company during the relevant class period, and that smoking was the proximate cause of her injury. Every tobacco company must also be afforded the opportunity to contest the smokers' pleadings and evidence and to plead and prove the smokers' comparative fault. Indeed, in this appeal, after the district court instructed it, the jury reduced Graham's damages award for his deceased spouse's comparative fault. And in other Engle progeny litigation, tobacco companies have won defense verdicts. E.g., Suarez v. R.J. Reynolds Tobacco Co., No. 09-79584-CA-01, 2015 WL 12776786 (11th Fla. Cir. Ct., Nov. 25, 2015) (final judgment). "[S]tate proceedings need do no more than satisfy the minimum procedural requirements" of due process to receive full faith and credit. Kremer, 456 U.S. at 481, 102 S.Ct. 1883. The record in this appeal establishes that R.J. Reynolds and Philip Morris were afforded the protections mandated by the Due Process Clause.

"Under the Full Faith and Credit Act, federal courts generally should respect state court judgments, even where erroneous." Lops v. Lops, 140 F.3d 927, 938 (11th Cir. 1998); see also Hickerson v. City of New York, 146 F.3d 99, 107 (2d Cir. 1998) ("[T]o second-guess that court's determination of this issue would violate the full faith and credit statute."). We decide only whether applying Florida law in this case violates due process. We do not endorse or condemn the use of a class action in Phase I of the Engle litigation. Nor do we endorse or condemn the explication of res judicata by the Supreme Court of Florida. We say only that applying Florida law in this trial did not violate the tobacco companies' rights to due process of law.

R.J. Reynolds and Philip Morris argue that we are not compelled to give full faith and credit to Douglas because Graham was not a party in Douglas and Florida law does not allow non-mutual issue preclusion. Because state courts would not be bound by the Douglas decision in this circumstance, they argue, we are also not bound. But this argument is a straw man.

We do not give full faith and credit to the decision in Douglas; we instead give full faith and credit to the jury findings in Engle. The Florida Supreme Court in Engle interpreted those findings to determine what the jury actually decided, and the Florida Supreme Court in Douglas decided a matter of state law when it explained the preclusive effect of the Engle jury's Phase I findings. We are bound by the decisions of state supreme courts on matters of state law when we exercise diversity jurisdiction, subject to the constraints of due process. See Walker, 734 F.3d at 1284. We conclude that giving preclusive effect to the findings of negligence and strict liability by the Engle jury in individual actions by Engle class members against R.J. Reynolds and Philip Morris does not deprive these tobacco companies of property without due process of law.

B. Federal Law Does Not Preempt the Jury Findings of Negligence and Strict Liability.

"The Supremacy Clause of Art. VI of the Constitution provides Congress with the power to pre-empt state law." La. Pub. Serv. Comm'n v. FCC, 476 U.S. 355, 368, 106 S.Ct. 1890, 90 L.Ed.2d 369 (1986). "State action may be foreclosed by express language in a congressional enactment, by implication from the depth and breadth of a congressional scheme that occupies the legislative field, or by implication because of a conflict with a congressional enactment." Lorillard Tobacco Co. v. Reilly, 533 U.S. 525, 541, 121 S.Ct. 2404, 150 L.Ed.2d 532 (2001) (citations omitted). Conflicts arise in two ways: "when compliance with both federal and state regulations is impossible or when the state law `stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress,'" Hillman v. Maretta, ___ U.S. ___, 133 S.Ct. 1943, 1950, 186 L.Ed.2d 43 (2013) (citation omitted) (quoting Hines v. Davidowitz, 312 U.S. 52, 67, 61 S.Ct. 399, 85 S.Ct. 581 (1941)). "`[T]he purpose of Congress is the ultimate touchstone' in every pre-emption case." Medtronic, Inc. v. Lohr, 518 U.S. 470, 485, 116 S.Ct. 2240, 135 L.Ed.2d 700 (1996) (quoting Cipollone v. Liggett Grp., Inc., 505 U.S. 504, 516, 112 S.Ct. 2608, 120 L.Ed.2d 407 (1992)). "Congress' intent, of course, primarily is discerned from the language of the pre-emption statute and the `statutory framework' surrounding it." Id. at 486, 116 S.Ct. 2240 (quoting Gade v. Nat'l Solid Wastes Mgmt. Ass'n, 505 U.S. 88, 111, 112 S.Ct. 2374, 120 L.Ed.2d 73 (1992) (Kennedy, J., concurring in part and concurring in the judgment)).

This appeal presents an issue of conflict preemption. A party asserting conflict preemption faces a high bar:

Wyeth v. Levine, 555 U.S. 555, 565, 129 S.Ct. 1187, 173 L.Ed.2d 51 (2009) (second and third alteration in original) (quoting Lohr, 518 U.S. at 485, 116 S.Ct. 2240).

R.J. Reynolds and Philip Morris argue that the obstacle form of conflict preemption defeats the findings of negligence and strict liability in Engle. They argue that this Circuit avoided finding a violation of due process in Walker by construing the Engle findings as embracing a theory that all cigarettes manufactured by the tobacco companies are defective and the sale of all of those cigarettes is negligent because all of those cigarettes are dangerous — that is, that all of those cigarettes are addictive and cause disease. Federal law, they contend, preempts state law claims premised on the theory that all of the cigarettes manufactured by the tobacco companies are inherently dangerous.

We disagree. We conclude that federal tobacco laws do not preempt state tort claims based on the dangerousness of all the cigarettes manufactured by the tobacco companies. In other words, federal law does not preempt the Engle jury findings.

Congress has enacted six tobacco-specific laws that are relevant to this appeal. In 1965, Congress passed the Federal Cigarette Labeling and Advertising Act, Pub. L. No. 89-92, 79 Stat. 282, which made it unlawful to sell cigarettes without the following warning label: "Caution: Cigarette Smoking May Be Hazardous to Your Health." Id. § 4, 79 Stat. at 283. And the Act prohibited requiring any additional "statement relating to smoking and health" on cigarette packages or in cigarette advertising. Id. § 5, 79 Stat. at 283. Congress then passed the Public Health Cigarette Smoking Act of 1969, Pub. L. No. 91-222, 84 Stat. 87, which changed the language of the warning label to "Warning: The Surgeon General Has Determined That Smoking Is Dangerous to Your Health." Id. § 2, 84 Stat. at 88. The Act made it "unlawful to advertise cigarettes on any medium of electronic communication subject to the jurisdiction of the Federal Communications Commission." Id., 84 Stat. at 89. And it amended the preemption provision in the 1965 Act by adding the following statement: "No requirement or prohibition based on smoking and health shall be imposed under State law with respect to the advertising or promotion of any cigarettes the packages of which are labeled in conformity with the provisions of this Act." Id., 84 Stat. at 88 (codified as amended at 15 U.S.C. § 1334(b)). Congress again amended the Labeling Act by passing the Comprehensive Smoking Education Act, Pub. L. No. 98-474, 98 Stat. 2200 (1984). The Act replaced the warning with a series of warnings that must appear on cigarette packages and advertisements on a rotating basis. Id. § 4, 98 Stat. at 2201-03. The Act also required the Secretary of Health and Human Services to "establish and carry out a program to inform the public of any dangers to human health presented by cigarette smoking." Id. § 3, 98 Stat. at 2200. The Alcohol and Drug Abuse Amendments of 1983, Pub. L. No. 98-24, 97 Stat. 175, required the Secretary of Health and Human Services to issue a report to Congress every three years on, among other things, "the addictive property of tobacco." Id. § 2, 97 Stat. at 178. The Comprehensive Smokeless Tobacco Health Education Act of 1986, Pub. L. No. 99-252, 100 Stat. 30, regulates smokeless tobacco products. The Act requires that a warning appear on smokeless tobacco products, id. § 3, 100 Stat. at 30-32, prohibits the advertising of smokeless tobacco products "on any medium of electronic communications subject to the jurisdiction of the Federal Communications Commission," id. § 3(f), 100 Stat. at 32, and requires the Secretary of Health and Human Services to create a program to inform the public about the health effects of using smokeless tobacco products, id. § 2, 100 Stat. at 30. Last, the ADAMHA Reorganization Act, Pub. L. No. 102-321, 106 Stat. 323 (1992), conditions certain block grants on states making it unlawful "for any manufacturer, retailer, or distributor of tobacco products to sell or distribute any such product to any individual under the age of 18." Id. § 202, 106 Stat. at 394 (codified at 42 U.S.C. § 300x-26(a)(1)). We do not consider the Family Smoking Prevention and Tobacco Control Act, Pub. L. No. 111-31, 123 Stat. 1776 (2009), because the Act does not affect actions, like the individual Engle actions, that were pending in federal or state court during its passage, id. § 4, 123 Stat. at 1782.

Affording preclusive effect to the Engle jury findings does not frustrate the objectives of these federal laws on tobacco. The only significant requirement imposed on cigarette manufacturers by the six federal laws in question is the warning label requirement for cigarette packages and advertising. Three of the six statutes — the Federal Cigarette Labeling and Advertising Act, the Public Health Cigarette Smoking Act of 1969, and the Comprehensive Smoking Education Act — concern this warning label. Fittingly, the Labeling Act expressly preempts state laws that would impose labeling requirements. See 15 U.S.C. § 1334; Altria Grp., Inc. v. Good, 555 U.S. 70, 79, 129 S.Ct. 538, 172 L.Ed.2d 398 (2008). The other three statutes impose no significant requirements on cigarette manufacturers: the Comprehensive Smokeless Tobacco Health Education Act of 1986 concerns smokeless products, not cigarettes; the Alcohol and Drug Abuse Amendments imposed a requirement on the Secretary of Health and Human Services to submit reports about cigarettes; and the ADAMHA Reorganization Act conditions block grants to states.

Contrary to R.J. Reynolds and Philip Morris's argument, the statement of purpose in the Labeling Act, 15 U.S.C. § 1331, does not preserve cigarette sales. The second listed purpose of establishing a program to "deal with cigarette labeling and advertising" states, "[C]ommerce and the national economy may be (A) protected to the maximum extent consistent with this declared policy and (B) not impeded by diverse, nonuniform, and confusing labeling and advertising regulations." Id. Congress sought to protect "commerce and the national economy" specifically from the effect of "diverse, nonuniform and confusing cigarette labeling and advertising" rules, id., not from more stringent regulation generally. See Altria Grp., 555 U.S. at 78-79, 129 S.Ct. 538 (explaining that the "Act's pre-emption provisions promote its second purpose" by preventing States from "enforcing rules that are based on an assumption that the federal warnings are inadequate"); Reilly, 533 U.S. at 542-43, 121 S.Ct. 2404 (paraphrasing the second purpose as "to protect the national economy from interference due to diverse, nonuniform, and confusing cigarette labeling and advertising regulations"); Marotta, 214 So.3d at 599, 2017 WL 1282111, at *7 ("Thus, Congress clearly intended to `protect the national economy from the burden imposed by diverse, nonuniform, and confusing cigarette labeling and advertising regulations,' but did not clearly intend to extend broad immunity from common law liability to cigarette manufacturers." (citation omitted)).

Nothing in these six statutes reflects a federal objective to permit the sale or manufacture of cigarettes. As a result, we cannot say that Congress created a regulatory scheme that does not tolerate tort liability based on the dangerousness of all cigarettes manufactured by the tobacco companies but tolerates tort actions based on theories with a more limited scope. Cf. Altria Grp., 555 U.S. at 90, 129 S.Ct. 538 (holding that federal law did not preempt common-law fraud claim against cigarette manufacturer based on advertising of light cigarettes); Boerner v. Brown & Williamson Tobacco Co., 394 F.3d 594, 600 (8th Cir. 2005) (holding that the Labeling Act did not preempt design defect claim against cigarette manufacturer); Spain v. Brown & Williamson Tobacco Corp., 363 F.3d 1183, 1197 (11th Cir. 2004) (holding that the Labeling Act did not preempt negligent and wanton design and manufacture claims against cigarette manufacturer). Federal law is silent both by its terms and by its operation.

Determinations of strict liability and negligence based on the Engle findings create no conflict with a federal objective. R.J. Reynolds and Philip Morris do not contend that the Engle jury based its findings of liability on a determination that the warnings on cigarette packages and advertisements were inadequate such that the jury's findings imposed labeling requirements preempted by federal law. Rules governing the design of cigarettes or even banning the sale of cigarettes do not frustrate accomplishing a rule that requires a certain label when and if cigarettes are sold. See Hunter v. Philip Morris USA, 582 F.3d 1039, 1048 (9th Cir. 2009) (explaining that product-liability claim against cigarette manufacturer "does not present an obstacle to the congressional policy concerning the regulation of tobacco" because the federal laws "concern labeling, research and education and do not provide strong evidence of a federal policy against more stringent state regulation"); Marotta, 214 So.3d at 600, 2017 WL 1282111, at *9 ("Strict liability and negligence claims, such as those brought ... under Engle, do not interfere with the regulation of advertising and promotion of cigarettes and, therefore, do not clearly conflict with congressional objectives.").

That the express-preemption provision in the Labeling Act does not cover the negligence and strict liability findings in Engle supports an inference that there is no implied preemption of those findings. See Wyeth, 555 U.S. at 574-75, 129 S.Ct. 1187; Riegel v. Medtronic, Inc., 552 U.S. 312, 327, 128 S.Ct. 999, 169 L.Ed.2d 892 (2008). Granted, "[i]f a federal law contains an express pre-emption clause, it does not immediately end the inquiry because the question of the substance and scope of Congress' displacement of state law still remains." Altria Grp., 555 U.S. at 76, 129 S.Ct. 538; see also Geier v. Am. Honda Motor Co., 529 U.S. 861, 874, 120 S.Ct. 1913, 146 L.Ed.2d 914 (2000). But, with the Federal Cigarette Labeling and Advertising Act and the Public Health Cigarette Smoking Act of 1969, in Cipollone the Supreme Court interpreted the express-preemption provision as exclusively defining the preemptive scope of the Acts:

Cipollone, 505 U.S. at 517, 112 S.Ct. 2608 (citations omitted) (quoting Malone v. White Motor Corp., 435 U.S. 497, 505, 98 S.Ct. 1185, 55 L.Ed.2d 443 (1978); Cal. Fed. Sav. & Loan Ass'n v. Guerra, 479 U.S. 272, 282, 107 S.Ct. 683, 93 L.Ed.2d 613 (1987) (plurality opinion)).

The Supreme Court has explained that "in Cipollone, we engaged in a conflict pre-emption analysis of the Federal Cigarette Labeling and Advertising Act, and found `no general, inherent conflict between federal preemption of state warning requirements and the continued vitality of state common-law damages actions.'" Freightliner Corp. v. Myrick, 514 U.S. 280, 288-89, 115 S.Ct. 1483, 131 L.Ed.2d 385 (1995) (citation omitted) (quoting Cipollone, 505 U.S. at 518, 112 S.Ct. 2608). Although the Supreme Court considered only the 1965 and 1969 statutes in Cipollone, "[s]ince the Labeling Act's passage, Congress's basic goals have remained largely unchanged." Graham, 782 F.3d at 1277. We find nothing in the four statutes passed later that alters the preemptive scope of federal law on tobacco in a way that is relevant to this appeal.

R.J. Reynolds and Philip Morris argue that, by passing legislation that does not ban cigarettes, Congress has established a policy of allowing the sale of tobacco products, but this argument is contrary to settled law that inaction by Congress cannot serve as justification for finding federal preemption of state law. See Wyeth, 555 U.S. at 602-03, 129 S.Ct. 1187 (Thomas, J., concurring in the judgment) (collecting cases); Sprietsma v. Mercury Marine, 537 U.S. 51, 65, 123 S.Ct. 518, 154 L.Ed.2d 466 (2002) ("History teaches us that a Coast Guard decision not to regulate a particular aspect of boating safety is fully consistent with an intent to preserve state regulatory authority...."); Schneidewind v. ANR Pipeline Co., 485 U.S. 293, 306, 108 S.Ct. 1145, 99 L.Ed.2d 316 (1988) ("This Court generally is reluctant to draw inferences from Congress' failure to act."). "[O]therwise, deliberate federal inaction could always imply pre-emption, which cannot be. There is no federal pre-emption in vacuo, without a constitutional text or a federal statute to assert it." P.R. Dep't of Consumer Affairs v. Isla Petroleum Corp., 485 U.S. 495, 503, 108 S.Ct. 1350, 99 L.Ed.2d 582 (1988).

R.J. Reynolds and Philip Morris also rely on the discussion of federal law regulating cigarettes in FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 120 S.Ct. 1291, 146 L.Ed.2d 121 (2000), but that decision does not support their argument for preemption. In Brown & Williamson, the Supreme Court considered whether the Food and Drug Administration had jurisdiction over tobacco products. Id. at 125-26, 120 S.Ct. 1291. The Court held that it did not. Id. at 126, 120 S.Ct. 1291. The Supreme Court reasoned that, if the Administration had jurisdiction, the Food, Drug, and Cosmetic Act would require the administration to remove cigarettes from the market. Id. at 135, 120 S.Ct. 1291. The Supreme Court considered the six federal statutes that regulate cigarette labeling and concluded that Congress would not have enacted these laws if it intended the Administration to ban cigarettes. See id. at 137-38, 120 S.Ct. 1291. "[T]he collective premise of these statutes is that cigarettes and smokeless tobacco will continue to be sold in the United States." Id. at 139, 120 S.Ct. 1291. The Supreme Court stated that Congress has "foreclosed the removal of tobacco products from the market" in this context, id. at 137, 120 S.Ct. 1291 — surmising that Congress would not have bothered to regulate a product that it intended to have removed from the market nationwide by a federal agency.

Although federal agencies have only the authority granted to them by Congress, states are sovereign. Brown & Williamson does not address state sovereignty, and it does not consider the preemptive reach of federal legislation on tobacco. Marotta, 214 So.3d at 598, 2017 WL 1282111 at *6 ("[W]hile Brown & Williamson held that the FDA did not have the authority to regulate tobacco products, it said nothing about the states' power to do the same."). Cipollone does.

State governments retain their historic police powers to protect public health. See U.S. Const. Amend. X. "It is one of the happy incidents of the federal system that a single courageous state may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country." New State Ice Co. v. Liebmann, 285 U.S. 262, 386-87, 52 S.Ct. 371, 76 S.Ct. 747 (1932) (Brandeis, J., dissenting). Over a hundred years ago, Tennessee, like some other states, passed a law making it a crime to sell cigarettes. 6 Clark Bell, Medico-Legal Studies 50-65 (1902). Although that experiment in prohibition, like so many others, failed, Tennessee did not violate the federal Constitution. In upholding the law as not infringing the power of Congress under the Commerce Clause, the Supreme Court described the cigarette ban as the type of legislation that states may enact "for the preservation of the public health or safety" under their police powers. Austin v. Tennessee, 179 U.S. 343, 349, 21 S.Ct. 132, 45 S.Ct. 224 (1900). Today, state and local governments continue to enact public health measures to respond to the dangers associated with smoking, see, e.g., Paul A. Diller, Why Do Cities Innovate in Public Health? Implications of Scale and Structure, 91 Wash. U. L. Rev 1219, 1234-35 (2014) (discussing state and local bans of flavored cigarettes passed before the Tobacco Control Act banned cigarette flavorings); Patrick Kabat, Note, "Till Naught but Ash is Left to See": Statewide Smoking Bans, Ballot Initiatives, and the Public Sphere, 9 Yale J. Health Pol'y L. & Ethics 128, 138-45 (2009) (surveying state prohibitions of smoking in public places), and to combat other public health risks, see, e.g., Cal. Health & Safety Code § 114377 (banning certain trans fats); N.Y. State Rest. Ass'n v. N.Y.C. Bd. of Health, 556 F.3d 114 (2d Cir. 2009) (upholding New York City law requiring caloric disclosure on chain restaurant menus against preemption challenge); Trans Fat and Menu Labeling Legislation, Nat'l Conference of State Legislatures (Jan. 2013), http://www.ncsl. org/research/health/trans-fat-and-menu-labeling-legislation.aspx (all Internet materials as visited July 9, 2016, and available in Clerk of Court's case file) (listing six states that had enacted menu labeling legislation as of 2010).

Florida may employ its police power to regulate cigarette sales and to impose tort liability on cigarette manufacturers. We may not supersede the "historic police powers of the States" unless it is the "clear and manifest purpose of Congress." Wyeth, 555 U.S. at 565, 129 S.Ct. 1187 (quoting Lohr, 518 U.S. at 485, 116 S.Ct. 2240). And "[t]hat assumption applies with particular force when Congress has legislated in a field traditionally occupied by the States," Altria Grp., 555 U.S. at 77, 129 S.Ct. 538, like public health, Lohr, 518 U.S. at 475, 116 S.Ct. 2240.

R.J. Reynolds and Philip Morris would have us presume that Congress established a right to sell cigarettes based on a handful of federal labeling requirements. We decline to do so. We discern no "clear and manifest purpose" to displace tort liability based on the dangerousness of all cigarettes manufactured by the tobacco companies.

IV. CONCLUSION

We AFFIRM the judgments against R.J. Reynolds and Philip Morris.

JULIE CARNES, Circuit Judge, concurring in part and dissenting in part:

I concur in the majority opinion's decision that federal law does not preempt the jury findings in the underlying Engle litigation. As to defendants' Due Process Clause challenge, the latter presents a close question on which reasonable minds can differ. I do not disagree that the majority opinion articulates reasonable arguments in explaining why it rejects defendants' challenge. On balance, however, I agree with Judges Tjoflat and Wilson that on the particular and unusual facts of the underlying Engle litigation, its jury findings are too non-specific to warrant them being given preclusive effect in subsequent trials. Concluding that defendants' due process rights were therefore violated, I respectfully dissent as to the Majority's contrary holding.

TJOFLAT, Circuit Judge, dissenting:

In 1998, the Circuit Court of the Eleventh Judicial Circuit in and for Dade County, Florida held a trial — Phase I of the Engle class action — to determine whether the largest domestic tobacco companies (the "Engle defendants") engaged in conduct proscribed by Florida tort law.1 The Phase I jury found that each of the Engle defendants engaged in nine different kinds of proscribed conduct.2

Eight years later, the Florida Supreme Court "retain[ed] the jury's Phase I findings other than those on the fraud and intentional infliction of emotion distress claims." Engle v. Liggett Group, Inc. (Engle III), 945 So.2d 1246, 1269 (Fla. 2006). It then instructed progeny courts tasked with adjudicating causation and damages in individual class-member tort actions to give "res judicata effect" to the retained findings. Id. at 1254.

But progeny courts had trouble understanding Engle III's res judicata instruction. For starters, issuing such a mandate was strange because courts that render a judgment ordinarily do not attempt to predetermine the res judicata effect of that judgment.3 Courts tasked with determining whether to enforce a rendering court's judgment make those determinations themselves.4 On top of that, the Phase I findings only established that the Engle defendants engaged in proscribed conduct; the findings revealed nothing about what the defendants actually did. Thus, the findings were useless in helping plaintiffs prove, as Florida law requires,5 that their injuries were caused by a defendant's proscribed conduct. Judge May of Florida's Fourth District Court of Appeal lamented that "parties to the tobacco litigation [were left to] ... play legal poker, placing their bets on questions left unresolved by Engle." R.J. Reynolds Tobacco Co. v. Brown (Jimmie Lee Brown II), 70 So.3d 707, 720 (Fla. 4th Dist. Ct. App. 2011) (May, J., concurring).

Seven years after it had issued its res judicata mandate, the Florida Supreme Court finally stepped in to explain it. The Court conceded that the Phase I findings were "useless in individual actions." Philip Morris USA, Inc. v. Douglas (Douglas III), 110 So.3d 419, 433 (Fla. 2013) (emphasis added). Nevertheless, by retaining certain Phase I findings and instructing progeny courts to give those findings res judicata effect in Engle III, the Court claimed it had intended to allow class members to simply "assume[ ]" "injury as a result of the Engle defendants' conduct." Id. at 430 (emphasis added). Thus, regardless of the tort a class member alleged, she only needed to prove that she was injured as a result of "`smoking cigarettes' manufactured by [a defendant]" to recover. Id. at 426. In effect, then, the Douglas III Court proscribed the very act of selling cigarettes, albeit under color of traditional tort law. So long as a defendant's sale of cigarettes caused a plaintiff's injury — that is, so long as a plaintiff was injured by smoking cigarettes — the plaintiff had no need to identify, for example, the defendant's negligent conduct or unreasonably dangerous product defect.

This case was litigated pursuant to the state law set forth in Douglas III. Earl Graham, as personal representative of the estate of Faye Graham, alleged in his complaint all of the torts for which Engle III had retained findings. Yet, he was never required to identify any proscribed conduct other than the sale of cigarettes. With respect to both negligence and strict liability, the District Court instructed the jury to determine only "whether smoking cigarettes manufactured by [the] Defendant was a legal cause of Faye Graham's death."

The Majority purport to give effect to the "state law" created in Douglas III. Ante at 1185-86. They recognize that it is "unorthodox," "novel," and amounts to an irrebuttable presumption of liability. Id. at 1183-84, 1184-85, 1185. Yet, they believe that due process is flexible enough to accommodate such a law. Id. at 1185. It is not.6

This is not to say, as the Majority imply, that I would "require a state to follow the federal common law of res judicata and collateral estoppel." Id. at 1184. I agree that states are free to fashion "novel" and even "unorthodox" laws. Id. at 1183, 1184-85. I do not agree, however, that federal courts must apply such laws when doing so deprives litigants of an opportunity to be heard on essential elements of their case.7 Id. at 1185-86. I also disagree with the Majority's conclusion that Florida's proscribing the sale of cigarettes is not preempted by federal law.

To navigate the surprising evolution from Phase I's nonprobative findings of fact to Douglas III's sweeping new tort law, I start from the beginning of the Engle litigation and proceed painstakingly to the end. As the Table of Contents indicates, I begin with Phase I of Engle and proceed through Walker v. R.J. Reynolds Tobacco Co. (Walker II), 734 F.3d 1278 (11th Cir. 2013), and to the opinion the Court issues today. Along the way, I comment on the decisions in light of relevant legal principles. My commentaries are set aside by conspicuous section breaks or headings, and my preemption discussion is set within its own part as it is more legally complex than the basic principles of procedural fairness that animate the rest of the opinion.

As I detail below, Engle-progeny opinions examining the same basic legal issues vary drastically in both their analysis and recitation of the facts. The Majority, for example, portray Engle III differently from the way all other courts, including the Florida Supreme Court, see that case.8 Unfortunately, the one theme that remains constant throughout — with a few exceptions — is that Engle-progeny courts have rested their thumbs on the scales to the detriment of the unpopular Engle defendants.

I dissent for eight reasons. First, I reject the Majority's false narrative of Engle III. Second, in injecting their false narrative into the case, the Majority improperly act as advocates and relieve the plaintiff of his burden of proving preclusion. Third, the Majority fail to provide the defendants with an opportunity to be heard on the accuracy and applicability of their narrative. Fourth, even if that narrative were not false, Engle III, as portrayed by the Majority, would not be entitled to full faith and credit because its key holdings were rendered without affording the Engle defendants notice or opportunity to be heard. Fifth, and most importantly, we cannot deprive R.J. Reynolds ("RJR") and Philip Morris of their property because they have never been afforded an opportunity to be heard on whether their unreasonably dangerous product defect(s) or negligent conduct caused Ms. Graham's death. Sixth, we cannot give effect to a state law that amounts to an unreasonable and arbitrary presumption of liability.9 Seventh, we cannot give effect to a state law that operates to deprive the defendants of their Seventh Amendment right to a jury trial on contested and material elements of the claims against them. Eighth, the way in which the Engle-progeny litigation has been carried out has resulted in a functional ban on cigarettes, which is preempted by federal regulation premised on consumer choice.

TABLE OF CONTENTS

I. Procedural History of Engle ... 1195

A. Certifying the Engle Class ... 1196

B. Engle Trial to Proceed in Three Phases ... 1197

1. Phase I ... 1198

2. Phase II ... 1202

3. Posttrial Motions ... 1204

C. Appeal to the Third District Court of Appeal in Engle II ... 1207

D. Petition for Review to the Florida Supreme Court in Engle III ... 1208

II. What "Res Judicata" Traditionally Means ... 1213

A. Res Judicata 101: The Elements of Issue and Claim Preclusion ... 1214

B. Res Judicata 102: Procedures to Invoke Issue and Claim Preclusion ... 1218

III. Engle III Instructed Courts to Disregard Traditional Res Judicata Law so as to Hold the Defendants Liable without Regard to the Phase I Findings ... 1221

A. The U.S. District Court for the Middle District of Florida in Brown I Rejected the Florida Supreme Court's Interference with Its Duties as a Recognizing Court ... 1223

B. In Brown II, We Upheld the District Court's Decision as a Recognizing Court to Apply Florida's Traditional Issue-Preclusion Doctrine to the Phase I Findings... 1234

C. The Florida District Courts of Appeal Rejected Brown II on the Basis of Engle III's Instruction ... 1236

1. The Martin I Circuit Court Concluded That Engle III's Instruction Required It to Hold the Defendants Liable If the Plaintiff Simply Proved Class Membership Irrespective of the Phase I Findings... 1237

2. The First District Court of Appeal in Martin II Agreed That Engle III's Instruction Required It to Hold the Defendants Liable to all Class Members Irrespective of the Phase I Findings ... 1239

3. The Fourth District Court of Appeal in Jimmie Lee Brown II Held That Engle IlI's Instruction Meant Issue Preclusion but That the Plaintiff Did Not Need to Identify a Specific Defect or Negligent Conduct ... 1245

D. In Light of Martin II and Jimmie Lee Brown II, the Middle District of Florida in Waggoner Ruled That the Preclusive Application of the Phase I Findings to Hold the Defendants Liable Would Not Violate Due Process ... 1248

E. The Second District Court of Appeal in Douglas II Accepted Martin II's Reasoning, But Certified the Due Process Question to the Florida Supreme Court ... 1252

IV. The Florida Supreme Court in Douglas III Held That the Engle III Court Had (1) Implicitly Determined That the Phase I Findings Were Full-Blown Liability Determinations and (2) Implicitly Entered Judgment Against All Defendants on Behalf of All Class Plaintiffs ... 1255

V. The Walker Panel Effectively Rewrote and then Gave Full Faith and Credit to Douglas III Before Issuing a New Opinion That Gave Full Faith and Credit to Engle III, Yet Left the Original Opinion's Inapposite Reasoning Intact ... 1273

VI. The Majority Repeat and Add to the Walker Panel's Errors ... 1285

VII. The Functional Ban on Cigarettes is Preempted by Federal Law ... 1292

A. Obstacle Preemption ... 1293

B. Federal Regulation of Tobacco Consumers' Ability to Choose ... 1295

C. Florida Has Imposed a Duty Not to Sell Cigarettes Contrary to Federal Law ... 1297

D. The Majority Misinterpret the Statutory Framework of Tobacco Regulation ... 1299

Conclusion ... 1300

I.

PROCEDURAL HISTORY OF ENGLE

The Engle litigation epic began in 1994 when six plaintiffs filed a putative class action in the Circuit Court for Miami-Dade County, Florida against the Engle defendants seeking over $100 billion in both compensatory and punitive damages for injuries allegedly caused by smoking cigarettes. Walker II, 734 F.3d at 1278. The plaintiffs asserted an array of claims, including "strict liability, negligence, breach of express warranty, breach of implied warranty, fraud, conspiracy to commit fraud, and intentional infliction of emotional distress." Liggett Grp. Inc. v. Engle (Engle II), 853 So.2d 434, 441 (Fla. 3d Dist. Ct. App. 2003).

A. Certifying the Engle Class

On May 5, 1994, the plaintiffs moved the Circuit Court pursuant to Florida Rule of Civil Procedure 1.220(b)(3)10 to certify a class consisting of all smokers in the United States and their survivors. They estimated that the class would include "in excess of one million addicted smokers." R.J. Reynolds Tobacco Co. v. Engle, 672 So.2d 39, 41 (Fla. 3d Dist. Ct. App. 1996). The defendants opposed the motion, arguing that it failed to establish the "predominance" and "superiority" requirements imposed by Rule 1.220(b)(3). Id. at 39. They further argued that a nationwide class would be unmanageable and would unduly burden Florida's courts and taxpayers. Id. at 41-42. The Circuit Court disagreed. It granted the plaintiffs' motion and certified a nationwide class.

The defendants appealed the decision to the District Court of Appeal, Third District.11 Id. at 39. The Third District found that the plaintiffs' motion satisfied the Rule 1.220(b)(3) "predominance" requirement but agreed with the defendants that a nationwide class was too large in that it "would unduly burden Florida courts and taxpayers," and would "require the sustained attention of all ... circuit judges in Dade County, if not the entire state." Id. at 40, 41. After the nationwide class had been rejected, the plaintiffs responded with their fallback position — a statewide class, which, they later represented, would consist of roughly 40,000 members.

Appeased, the Court affirmed the certification order on January 31, 1996, but limited the class to "[a]ll Florida citizens and residents," "and their survivors, who have suffered, presently suffer or have died from diseases and medical conditions caused by their addiction to cigarettes that contain nicotine." Id. at 40-42. In their motion for rehearing, the defendants rejected a 40,000-claimants estimate, insisting that a statewide class would consist of an unmanageable host of hundreds of thousands of class members. Nevertheless, the Court denied their motion on May 10, 1996, and the Florida Supreme Court denied review on October 2, 1996. R.J. Reynolds Tobacco Co. v. Engle, 682 So.2d 1100 (Fla. 1996). Three months later, plaintiffs' counsel wrote thousands of Florida physicians informing them of the class action and stating that the class included "well over one-half million" people.

By the end of 1997, as the case proceeded through its pretrial stages, the class had indeed grown to hundreds of thousands of claimants.12 In light of the class size and plaintiffs' counsel's concession that addiction to nicotine was an individual issue, the Engle defendants moved to decertify the class. The Court heard the motion on January 15, 1998.13 It denied the motion with this comment:

The defendants appealed the Court's ruling to the Third District. That Court dismissed the appeal for lack of jurisdiction, but stated that the defendants had a right to obtain review of "the propriety of the order by plenary appeal from any adverse final judgment." Engle II, 853 So.2d at 443.

B. Engle Trial to Proceed in Three Phases

In February 1998, the Circuit Court announced that it had developed a tentative three-phase trial plan to manage the litigation.14 Engle III, 945 So.2d 1246, 1256 (Fla. 2006). In Phase I, the Court would conduct a jury trial of the issues common to the entire statewide class. The plan defined the issues as those "which form integral elements of the claims" the named plaintiffs were asserting for themselves and the members of their class.15 The jury would evaluate evidence exclusively related to the defendants' conduct and would determine whether such conduct rendered the defendants answerable in punitive damages.

If the jury found that the defendants had engaged in the tortious conduct alleged, the litigation would proceed to Phase II-A to determine whether that conduct caused the class representatives' injuries. In Phase II-B, the same jury would also decide whether the entire class was entitled to punitive damages, and, if so, make a "lump-sum" award. Engle III, 945 So.2d at 1257. Finally, in Phase III, new juries would try the individual class members' claims — that the Engle defendants' tortious conduct caused their harm. Id. at 1268. The punitive damages, if any, awarded in Phase II-B would be divided among the class members who prevailed.

1. Phase I

The Phase I trial commenced on July 6, 1998. In accordance with the plan, the Phase I jury considered evidence pertaining to the defendants' conduct between 1953 and 1994 and to whether cigarettes manufactured during that time were addictive and caused diseases. Over the course of the yearlong trial, the plaintiffs presented evidence that was sweeping in its scope, spanning decades of tobacco-industry history. Ante at 1175-76. Witnesses testified that cigarettes were addictive and could cause a variety of diseases, including lung cancer. Douglas III, 110 So.3d at 423 (Fla. 2013). Witnesses also described differences among cigarette brands, filtered and nonfiltered, in terms of their tar and nicotine levels and the way in which they were designed, tested, manufactured, advertised, and sold. Id. at 423-24.

With such wide-ranging evidence and disparity among cigarettes, the defendants registered early on their concerns that the jury would have a hard time sorting through the evidence and connecting it to particular defendants and particular assertions of wrongdoing. They repeatedly argued, for example, "that [the] wide spectrum of views ... represented by counsel... [make it] hard [to] figur[e] out where we're going as a common question." The defendants later summarized their concerns:

Undeterred, the Court responded that it would make sense of the scattershot theories and evidence by means of jury instructions at the end of Phase I.

In March 1999, the plaintiffs rested, and the defendants moved the Court for decertification of the class and a directed verdict on all counts. After eight months of trial, the defendants pressed the Court to address the manageability problems that had been looming since the beginning. Although the plaintiffs had, to that point, successfully urged the Court to postpone such issues until "later," the defendants insisted that "later is here. Later is now."

Given the jumble of evidence and theories that had been put forward, the defendants argued, the jury would be unable to match theories with evidence as required unless it was instructed with precision:

Such reasoning undergirded the defendants' motions for directed verdict as well. In those motions, the defendants argued that the plaintiffs had spread themselves too thin by sporadically referencing, while never fully substantiating, numerous theories of liability. The defendants worried that these shotgun-style allegations would unfairly disadvantage them if their motions were denied:

The plaintiffs did not confront the merits of such arguments directly, countering instead with two process-oriented arguments. First, they argued that the defendants failed to satisfy the directed verdict standard because "the burden of the defendants is an almost impossible burden. In most instances because the defendants have to convince the Court that there is not minimal but zero, zero evidence and zero inferences from the evidence that would support our claims." Second, the plaintiffs argued that the Court should defer its ruling because the law demands "that in those rare instances where the Court really doesn't feel there's enough to go to a jury, the Court should wait," let the jury render a verdict, and then rule, so the appellate court can reinstate the jury verdict if it disagrees with the trial judge.

Persuaded by the plaintiffs, the Court reserved ruling on the motion,16 and the defendants went on to present their case.17 On June 9, 1999, the parties rested, and a charge conference with counsel followed. The plaintiffs conceded that there were "many hundreds and hundreds" of things at issue for each claim. To account for the many theories presented, and, concomitantly, to provide the jury with the widest possible range of bases upon which to premise tortious-conduct findings, the plaintiffs proposed that the jury be instructed on eight different theories of negligence and five theories of strict liability. The defendants did not take issue with instructing the jury on an array of tortious-conduct theories. They did object, however, to the plaintiffs' proposed verdict-form questions, which were generic rather than disaggregated and specific. They warned that a verdict form that failed to specify the particular theories on which the jury based its findings could not be "meaningful[ly] imported into Phase II and Phase III":

Because, the defendants argued, a generic verdict form would make it "completely impossible to import intelligently and rationally the findings from the verdict form in Phase I to any particular plaintiff in Phase II and III," relying on such a verdict form to preclude defendants' defenses in later phases would result in a "due process violation under the U.S. Constitution as well as the Florida Constitution."

The defendants accordingly requested a verdict form that would elicit specific findings that class members could later allege, in a meaningful way and in accordance with due process, in their Phase III complaints. See Walker II, 734 F.3d at 1282 (The defendants "requested that the trial court submit to the jury a ... detailed verdict form that would ... ask[ ] the jury [among other things] to identify the brands of cigarettes that were defective."). Plaintiffs repeatedly opposed such requests, arguing that specificity burnished a slippery slope to complexity and delay: "[O]nce you start [being more specific], then you've got to include a lot more.... And that becomes a 20, 25-page verdict form for the jury to complete, yes, no, and be here for a long time." The Court sided with the plaintiffs.

Hence, the first two questions on the finalized verdict form made no distinction between cigarette brands and did not even refer to the defendants' conduct.19 Instead, the questions asked the jury to determine whether cigarettes could cause certain diseases and addiction. The remaining verdict-form questions charged the jury to determine whether the defendants had engaged in tortious conduct, but did not require the jury to reveal the theory or theories on which it premised its tortious-conduct findings.20 Thus, as the defendants had feared, the verdict form did not prompt the jury to indicate whether it had accepted, for example, just one or all eight instructed theories of negligence.21 Nor did it prompt the jury to reveal which of the five instructed theories of strict liability22 it accepted or which particular brands of cigarettes or cigarette features it identified as defective and unreasonably dangerous. With respect to strict liability and negligence — the two claims at issue in this appeal — the form simply asked the jury to respond "yes" or "no" to whether "one or more of the Defendant Tobacco Companies" (1) "place[d] cigarettes on the market that were defective and unreasonably dangerous"23 and (2) "failed to exercise the degree of care which a reasonable cigarette manufacturer would exercise under like circumstances."24

The defendants objected to both questions, arguing that "[t]he [defect] question does not require specificity as to the product (brand or brand style), the defect, or the time of occurrence, which renders it useless for application to individual plaintiffs in other Phases of this case," and "the [negligence] question does not require specificity as to the product (brand and brand style), the alleged negligent act, and the date of the act, which renders it useless for application to individual plaintiffs in other Phases of this case." They objected to verdict-form questions related to other tortious-conduct claims as well, insisting that the answers to such questions would be useless in Phase III because a different jury would be unable to discern what conduct the Phase I jury deemed tortious, making it impossible to prove that such conduct caused harm.

The Court overruled the defendants' objections, and the jury, in the verdicts they returned on July 7, 1999, answered "yes" to every question.25 Walker II, 734 F.3d at 1282. The defendants moved the Court to set aside the verdicts in accordance with their motions for directed verdict and alternatively for a new trial.26 They also moved the Court to decertify the class. On July 29th, the Court summarily denied these motions.

2. Phase II

The trial of Phase II-A — the cases of three class representatives, Mary Farnan, Frank Amodeo, and Angie Della Vecchia,27 against six tobacco companies28 — began on November 1, 1999. All alleged that they were addicted to cigarettes, smoked a variety of the companies' brands, both filtered and nonfiltered,29 and in time contracted cancer. All sought damages against each defendant on theories of strict liability and negligence.30 Because the Phase I trial did not involve the class representatives' claims (or those of any class members), the Phase I jury was not instructed to determine whether any of the brands these plaintiffs actually smoked were defective, unreasonably dangerous, or negligently produced.31 In Phase I, the jury had determined "issues ... concerning the conduct of the tobacco industry." In Phase II-A, the same jury32 was tasked with deciding inter alia whether the tortious conduct it identified in Phase I caused the class representatives' injuries.

The jury had the Phase I trial record before it, and the three plaintiffs augmented that record by alleging the various brands of cigarettes they smoked, their inability to stop smoking, and that cigarette smoking caused the cancer they contracted.33 After they rested their cases, the defendants moved the Court for directed verdicts on the ground that the plaintiffs failed to prove all elements of their claims, including whether the cigarettes the plaintiffs smoked were defective, unreasonably dangerous, or negligently produced. Evidence that cigarettes could cause disease, the defendants argued, did not establish that their tortious conduct caused the plaintiffs' diseases.

The Court deferred its ruling on the motion until after the jury rendered its verdicts on the plaintiffs' claims. In the Court's view, the jury's answers to the Phase I verdict-form questions, coupled with the plaintiffs' testimony that they could not stop smoking and their experts' testimony that their smoking caused their cancer, were all the plaintiffs needed to make out a case for the jury under the theories of strict liability and negligence they were advancing.

The Court's instructions to the jury reflected this view.34 The Court began by explaining that the issues the jury decided in Phase I were not being litigated anew. What it had to decide now was whether the defendants' "conduct" on which it based its Phase I verdict was the "legal cause of injury to Mary Farnan, Frank Amodeo and Angie Della Vecchia."35 Turning to the verdict form it would be submitting to the jury, the Court informed the jury of the issues it had to decide by answering "yes" or "no" to a series of questions, each prefaced with a finding the jury made in Phase I.

The first question was prefaced with this statement: "In your [Phase I verdict], you found that smoking cigarettes causes ... lung cancer and laryngeal (throat) cancer." The question that followed asked, "[W]as smoking cigarettes a legal cause" of the plaintiff's cancer? If the jury answered "yes," it would proceed to the question pertaining to the claims of strict liability. The preface read, "You found in your [Phase I verdict] that each of the Defendant Tobacco Companies placed cigarettes on the market that were defective and unreasonably dangerous, both before and after July 1 of 1974 (except for Brooke, whose liability is limited to after July, 1974)." That preface was followed by a question: "Were defective and unreasonably dangerous cigarettes placed on the market by one or more of the Defendant tobacco companies a legal cause of [the plaintiff's cancer]"?

In addition to answering this question regarding strict liability, the jury had to answer the question pertaining to the claims of negligence. The preface to the question was, "[I]n your [Phase I verdict], you found that all of the Defendant Tobacco Companies failed to exercise the degree of care which a reasonable cigarette manufacturer would exercise under like circumstances, both before and after July 1 of 1969 (except for Brooke whose liability is limited to after July 1, 1969)." The corresponding question was, "As to each of the Defendants ... please state whether that Defendant's negligence was a legal cause of [plaintiff's cancer]."

The Court sent the case to the jury on April 5, 2000. The jury returned its verdicts on April 7, 2000, responding "yes" to each of the questions and therefore, pursuant to the Court's instructions, proceeded to determine the amount of the plaintiffs' compensatory damages, which were offset by comparative fault. The total award was $12.7 million. Engle II, 853 So.2d at 441.

The trial of Phase II-B began on May 22, 2000. In Phase I, the jury determined that the defendants' conduct warranted the imposition of punitive damages,36 so the Phase II-B trial focused on the monetary sum that should be imposed. On July 14, 2000, the jury fixed that sum at $145 billion to be awarded incrementally to class members who prevailed in subsequent Phase III lawsuits. Id. at 1257.

3. Posttrial Motions

At the conclusion of Phase II-B, the defendants moved the Court for the entry of judgment (as to Phases II-A and II-B) in accordance with their motion for directed verdict.37 They also moved the Court to decertify the plaintiffs' class.38 The plaintiffs filed no opposition to the defendants' motions. The Court took the motions under advisement and, without entertaining oral argument, denied them on November 6, 2000, in an Amended Final Judgment and Amended Omnibus Order ("Omnibus Order").39

In denying the defendants' motion for the entry of judgment in accordance with their motion for directed verdict, the Omnibus Order addressed the plaintiffs' claims separately40 and concluded that each claim had substantial evidentiary support.41

The evidence introduced during the trial of Phase II-A was sufficient to prove that the plaintiffs had become addicted to the defendants' cigarettes and that smoking those cigarettes caused the plaintiffs' disease, cancer. That was all the plaintiffs had to show to prevail on their claims of strict liability, the Omnibus Order indicated, because the evidence introduced during the trial of Phase I established that both before and after July 1, 1974, the defendants had "placed cigarettes on the market that were defective and unreasonably dangerous."42 The plaintiffs also prevailed on their claims of negligence because, the Omnibus Order indicated, the defendants "failed to exercise the degree of care which a reasonable cigarette manufacturer would exercise under like circumstances."43 The plaintiffs were relieved of the burden of proving that specific defects in the defendants' cigarettes or specific tortious conduct caused their injuries. Instead, plaintiffs were required to prove only that smoking the defendants' cigarettes caused their injuries.

The Court had previously forecast that it would ease the plaintiffs' burden of proof in this way in a colloquy with Philip Morris' counsel during closing arguments in Phase II-A. The Court said,

The Court upheld the jury's punitive-damages award because "[i]n Phase I of the trial, the jury, having heard the testimony concerning the behavior and conduct of the defendants, decided that punitive damages were indeed appropriate in this case."44 The Court found that the $145 billion award was not unreasonable because "the amount of the jury verdict is within the parameters of the evidence at trial — within the limits of the highs and lows, albeit on the high side, but when the enormity of the facts and issues of this case are considered, the award cannot be said to be unreasonable."

The Court made one further reference to Phase I. "[I]t should be noted that the jury in ... Phase I ... found each of the defendants Guilty as to all counts with the exception of count 7 for Equitable relief which the court dismissed previously under the plaintiffs request for Medical Monitoring."

C. Appeal to the Third District Court of Appeal in Engle II

The defendants appealed the Omnibus Order to the Third District Court of Appeal. They argued that plaintiffs' counsel's race-based incendiary remarks throughout trial merited the judgment's complete reversal. They argued alternatively that the punitive damages should be set aside as foreclosed by Florida precedent and that the class should be decertified because the Phase I findings were useless. The findings of tobacco-company misconduct were "generalized"; hence, the defendants contended, the "Phase III juries [would be] unable to determine whether the conduct found to be wrongful in Phase I was the legal cause of any Phase III claimant's injury."45 Thus, the defendants warned, they would be faced with "an infinite re-examination of issues by different juries and the consequent risk of inconsistent verdicts, in violation of [their] constitutional right to have one — and only one — jury decide the same or interrelated issues."

On May 21, 2003, the Third District, persuaded by the defendants' arguments, held that "the entire judgment must be reversed and the class decertified." Engle II, 853 So.2d at 470. The Court began its opinion by noting that "[a]lthough the emotional appeal of the class representatives' claims is compelling, our job as appellate judges is not to be swayed by emotion where to do so results in violating established legal principles." Id. at 442. The Court found that the plaintiffs had "incit[ed] juror prejudice against an unpopular industry," concocted ostensibly "common" issues only by "creat[ing] a composite plaintiff who smoked every single brand of cigarettes, saw every single advertisement, read every single piece of paper that the tobacco industries ever created or distributed, and knew about every single allegedly fraudulent act." Id. at 467 n.48. Doing so enabled the class "to try fifty years of alleged misconduct that they never would have been able to introduce in an individual trial, which was untethered to any individual plaintiff." Id. Making matters worse moving forward, "there were no specific findings as to any act by any defendant at any period of time." Id. The Court acknowledged what the defendants had been arguing — the Phase I findings were useless.

The Court concluded that "Florida's class action rules, substantive tort law, and state and federal guarantees of due process and a fair trial, [all] require[d] class decertification." Id. at 450. In reaching this conclusion, the Court noted that "virtually all courts that have addressed the issue have concluded that certification of smokers' cases is unworkable and improper." Id. at 444 (collecting cases). This is in large part because "issues of liability, affirmative defenses, and damages, outweigh[] any `common issues' in th[e] case." Id. at 445. The impropriety of class certification was especially clear in this particular case, the Court explained, because "the jury did not determine whether defendants were liable to anyone." Id. at 450 (emphasis in original). And, "[a]s evidenced by the proceedings in Phase 2, each claimant will have to prove that his or her illness not only was caused by smoking, but was also proximately caused by defendants' alleged misconduct." Id. at 446 (emphasis added). Because "each class member had unique and different experiences that will require the litigation of substantially separate issues, class representation is not `superior' to individual suits." Id. at 446-47.

In addition to decertifying the class, the Court vacated the punitive-damages award on a host of independent grounds. First, the award violated "well-established Florida precedent" by

Id. at 450. Second, the size of the punitive-damages award was excessive under state and federal law, noting that "the $145 billion verdict is roughly 18 times the defendants' proven net worth." Id. at 457. Third, as explained in Young v. Miami Beach Improvement Co., 46 So.2d 26 (Fla. 1950), the punitive award was precluded by settlement agreements between the tobacco companies and the states, "which expressly included claims for punitive damages." Engle II, 853 So.2d at 467-70.

Lastly, the Court held that "Plaintiffs' counsel's improper race-based appeals for nullification caused irreparable prejudice and require reversal." Id. at 458. "The trial was book-ended with prejudicial attorney misconduct which incited the jury to disregard the law because the defendants are tobacco companies." Id. The Court explained that "Plaintiffs' counsel began making racially-charged arguments on the first day of trial," and perpetuated through closing. Id. Specifically,

Id. at 459-60. After citing many further examples of prejudicial conduct, the Court explained that "the improper comments of plaintiffs' counsel further deprived the defendants of due process and a fair trial, thus additionally requiring reversal." Id. at 466.

The Court ultimately summarized its holding thus: "The fate of an entire industry and of close to a million Florida residents, cannot rest upon such a fundamentally unfair proceeding." Id. at 470.

D. Petition for Review to the Florida Supreme Court in Engle III

The plaintiffs petitioned the Florida Supreme Court for review under Article V, Section 3(b)(3) of the Florida Constitution, which grants the Court jurisdiction to "review any decision of a district court of appeal that ... expressly and directly conflicts with a decision ... of the supreme court on the same question of law." The defendants opposed the Court's exercise of such jurisdiction by arguing that Engle II did not in fact come in "express and direct conflict" with any such precedent. The Supreme Court rejected the defendants' arguments and accepted jurisdiction based on a conflict between Engle II and its decision in Young v. Miami Beach Improvement Co. as to whether a settlement agreement between the state and the defendants would bind private citizens in their punitive-damages claims. Engle III, 945 So.2d at 1254, 1260. Following the Supreme Court's acceptance of jurisdiction, the parties jointly briefed the issues the Third District resolved in reaching its decision in Engle II.46

On July 6, 2006, a divided Supreme Court issued its decision.47 Resolving the issues the parties had briefed, the Court quashed the Third District's judgment as to each of its holdings except for its rejection of the punitive-damages award.48 Engle III, 945 So.2d at 1254-56. However, in its rejection of the holdings, the Court actually echoed many of the Third District's criticisms regarding the unmanageably expansive class action litigation. Id. at 1267-71. Most notably, the Court "agree[d] with the Third District that problems with the three-phase trial plan" required the class to be decertified. Id. at 1267-68.

To the Florida Supreme Court, however, decertification would not serve as an acceptable outcome for the class members who had been standing idly by while their attorneys tried Phases I and II of their case. To thus accommodate such class members, the Supreme Court sua sponte fashioned a "pragmatic solution" in which it preserved some of the Phase I findings for use in the class members' cases to establish tobacco-company liability. Id. at 1269.

The Court implemented its pragmatic solution in two steps. First, it certified, pursuant to Florida Rule of Civil Procedure 1.220(d)(4)(A),49 a class limited to liability issues; that is, limited to eight of the ten Phase I findings,50 the findings on "Questions 1 (general causation), 2 (addiction of cigarettes), 3 (strict liability), 4(a) (fraud by concealment), 5(a) (civil-conspiracy-concealment), 6 (breach of implied warranty), 7 (breach of express warranty), and 8 (negligence)."51 Engle III, 945 So.2d at 1255, 1268. It labeled these findings "common core findings." Id. at 1269. The Court did not certify the issue of legal causation — whether the Engle defendants' tortious conduct caused the class members' harm — noting that it was "highly individualized and [did] not lend [itself] to class action treatment." Id. at 1254. The Court acknowledged that "no Florida cases address whether it is appropriate under rule 1.220(d)(4)(A) to certify class treatment for only limited liability issues." Id. at 1268. It nevertheless certified the issues class because "several decisions by federal appellate courts applying a similar provision in the Federal Rules of Civil Procedure provide persuasive authority for this approach." Id.

Under step two of its pragmatic solution, the Court declared that these "common core findings ... will have res judicata effect" in the subsequent "damages actions" the class members would bring. Id. at 1269. The Phase I findings, which, as the Third District observed, were decided with reference to a "composite plaintiff who smoked every single brand of cigarettes, saw every single advertisement, read every single piece of paper that the tobacco industry ever created or distributed, and knew about every single allegedly fraudulent act," Engle II, 853 So.2d at 467 n.48, would now have the legal effect of a partial final judgment resolving issues for individual class members.52 The Court's pragmatic solution was therefore intended to enable class members — in suing an Engle defendant on claims of strict liability, negligence, breach of express warranty, breach of implied warranty, fraud, and conspiracy to commit fraud — to plead the Phase I findings to conclusively establish elements of their claims and thereby foreclose the defendant from denying such elements.53 Although the Phase I jury found none of the defendants foreclosed of their defenses,54 id. at 450, the Florida Supreme Court did. All that remained for progeny plaintiffs to prove, and for progeny juries to consider, was "individualized issues such as legal causation, comparative fault, and damages." Engle III, 945 So.2d at 1268.

After ruling on these two matters without providing the parties notice or opportunity to be heard on them,55 the Supreme Court remanded the case to the Third District "with directions that the class should be decertified without prejudice to the class members filing individual claims within one year of the issuance of our mandate in this case with res judicata effect given to certain Phase I findings," and "for further proceedings consistent with [its] opinion." Engle III, 945 So.2d at 1254, 1277.

On August 7, 2006, the tobacco companies moved the Supreme Court for rehearing. Their motion contended that the Court's certification of an issues class under Rule 1.220(d)(4)(A) and its pronouncement that the Phase I jury findings would "have res judicata effect" in the cases brought by class members denied them due process in that the Court provided them with no notice that it was contemplating such action and no opportunity to be heard. The denial of due process aside, the tobacco companies contended that the Court erred in certifying the issues class. The "basic principle of class-action law throughout the country ... [is] that certification — under any subdivision of the rules — must be addressed and determined before there is a trial on the merits." The companies' final contention was that the Phase I jury findings relating to the claims of strict liability and negligence, among others, could not be given "res judicata effect" because the findings were too generalized to provide a basis for individual causation consistent with due process.

The Florida Supreme Court withdrew its July 6, 2006, opinion, Engle v. Liggett Grp., 945 So.2d 1246 (Fla. 2006), and on December 21, 2006, published Engle III as a substitute. Engle III made minor modifications to the withdrawn opinion, but none are pertinent here. That same day, the Court summarily denied the tobacco companies' motion for rehearing in an order it chose not to publish. The order instructed the companies not to file another motion for rehearing.56 The Engle defendants petitioned the U.S. Supreme Court for certiorari relief, but their petition was denied. R.J. Reynolds Tobacco Co. v. Engle, 552 U.S. 941, 128 S.Ct. 96, 169 L.Ed.2d 244 (2007).

II.

WHAT "RES JUDICATA" TRADITIONALLY MEANS

In this appeal, RJR and Philip Morris challenge a judgment in favor of Earl Graham, as personal representative of the estate of his deceased wife, Faye Graham, on claims of strict liability and negligence. Ante at 1180. Under traditional Florida tort law, a plaintiff alleging strict liability in the products-liability context must prove inter alia (a) that the product in question was defective57 and (b) that the "defect caused the injury or harm alleged." Aubin v. Union Carbide Corp., 177 So.3d 489, 513 (Fla. 2015). Similarly, under traditional Florida tort law, a plaintiff alleging negligence must prove inter alia (a) that the defendant breached a duty of care owed to her and (b) that the defendant's breach caused her harm. Williams v. Davis, 974 So.2d 1052, 1056 (Fla. 2007). In the wake of Engle III's res judicata dicta,58 these traditional requirements have gone by the wayside in Engle-progeny cases.

In this case, for example, the District Court held the defendants liable even though Mr. Graham never proved that his late wife's injury was caused by the defendants' product defect(s) or negligent conduct. Instead, the Court allowed Mr. Graham to take advantage of state-law conclusive presumptions — which did not exist when the parties litigated Phase I and apply only in Engle-progeny cases — under which "injury as a result of the Engle defendants' conduct is assumed." Douglas III, 110 So.3d at 429.

The conclusive presumptions on which Engle-progeny plaintiffs rely effectively transform the Phase I findings from "useless," Douglas III, 110 So.3d at 433, to dispositive. For example, the Phase I finding that each defendant "place[d] cigarettes on the market that were defective and unreasonably dangerous" now establishes as a matter of law that (a) every cigarette smoked by every class plaintiff was defective and unreasonably dangerous59 and (b) such (unidentified) unreasonably dangerous defect(s) caused every class member's injury, including Ms. Graham's.60 Similarly, the Phase I finding that each defendant "failed to exercise the degree of care which a reasonable manufacturer would exercise under like circumstances" now establishes under state law that (a) the Engle defendants breached their duty of care to every class plaintiff61 and (b) their (unidentified) breach(es) caused every class member's injury, including Ms. Graham's.

That Engle III's dicta62 regarding the res judicata effect of the Phase I findings could so drastically alter the Phase I findings and Florida's preclusion doctrines and tort law is startling. Even more alarming is that progeny courts, including the Majority today, have consistently failed to address the resulting constitutional violations.63 In this dissent, I lay bare these violations, which have been carried forward and incrementally exacerbated for twenty years.

So far, I have traced the relevant procedural history preceding this case through Engle III. Below, I continue the narrative by detailing layer upon layer of judicial error committed by numerous state and federal courts, culminating finally with the Majority's errors today. To illuminate that narrative, I pause to explain some fundamental principles of common and constitutional law that progeny courts have either failed to understand or chosen to ignore. Specifically, I provide an overview of preclusion law and explain the U.S. Constitution's role in its effective operation. I then explain how progeny courts have interpreted Engle III's "res judicata" dicta as a mandate to disregard traditional preclusion law, tort law, and the Constitution; an invitation that many progeny courts have accepted.

A. Res Judicata 101: The Elements of Issue and Claim Preclusion

The term "res judicata" refers to all the ways in which the judgment of one court will have a binding effect in a subsequent case. Res judicata, Black's Law Dictionary 1425 (9th ed. 2009). This definition is the most common, but "lumps under a single name two quite different effects of judgments."64 Id. The first — "issue preclusion" or "collateral estoppel" — is the effect of foreclosing relitigation of matters that have been litigated and decided. Id. The second — "claim preclusion," "merger," or "bar" — is the effect of foreclosing any litigation of matters that have never been litigated because they should have been advanced in an earlier suit.65 Id.

Both issue preclusion and claim preclusion operate across a two-lawsuit continuum.66 First, parties litigate a dispute to a final judgment on the merits. Second, in a later, separate suit between the parties, one party brings to court evidence of an earlier judgment and contends that issue or claim preclusion should apply to prevent her opponent from litigating a previously decided issue67 or cause of action.68 In this two-lawsuit scheme, the first court is the "rendering" court and the second is the "recognizing" court. In this subpart, I elaborate on the elements of each doctrine.

Issue preclusion, as developed in the common law, "bars relitigation of an issue of fact or law that has been decided in a prior suit." Baloco v. Drummond Co., 767 F.3d 1229, 1251 (11th Cir. 2014). Drawing from its common-law roots, the doctrine only applies when

Id. Although some states articulate these elements differently, the core requirements are largely the same across all jurisdictions.

In Florida, the elements are set forth in a five-prong test. For issue preclusion to apply there must be (1) identical parties,69 (2) identical issue(s), (3) full litigation of the particular matter, (4) determination of the particular matter, and (5) a "final decision" in the prior proceeding by a court of competent jurisdiction. Dadeland Depot, Inc. v. St. Paul Fire and Marine Ins. Co., 945 So.2d 1216, 1235 (Fla. 2006) (quoting Dep't of Health & Rehab. Servs. v. B.J.M., 656 So.2d 906, 910 (Fla. 1995)).

Elements (2), (3), and (4) of the Florida doctrine culminate in an "actually decided" requirement, which is fundamental to issue preclusion. The requirement originated with early English authorities, which explained that preclusion requires a determination "directly upon point"; recognizing courts could not preclude parties from litigating issues on the basis that such issues might have been or probably were decided. The Duchess of Kingston's Case, 20 Howell's State Trials 538 (House of Lords 1776). Rather, courts could estop litigation only when the "estoppell" was "certaine to every intent, and not ... taken by argument or inference." 2 Coke, The First Part of the Institutes of the Laws of England; Or, A Commentary on Littleton ¶ 352a (1817).

This early English common-law requirement is now deeply ingrained in the American judicial system. Federal and state issue-preclusion doctrines have included the requirement for well over a century. See, e.g., Cromwell v. Sac County, 94 U.S. 351, 353, 4 Otto 351, 24 S.Ct. 195 (1876) ("[T]he inquiry must always be as to the point or question actually litigated and determined in the original action, not what might have been thus litigated and determined." (emphasis added)); Burlen v. Shannon, 99 Mass. 200, 203 (1868) (noting that "according to all the well considered authorities, ancient and modern," the inference that an issue was decided by prior litigation had to "be inevitable, or it [could not] be drawn"). And, to this day, federal and state courts uniformly adhere to it.70 Florida is no exception. See Brown v. R.J. Reynolds Tobacco Co. (Brown II), 611 F.3d 1324, 1334 ("Florida courts have enforced the `actually adjudicated' requirement with rigor." (citation omitted)).

The universality of the actually decided requirement is no accident; the requirement helps facilitate due process. When a rendering court decides an issue and a recognizing court later accords that issue preclusive effect, two consequences result: First, the precluded party is gagged from litigating that issue. Fayerweather v. Ritch, 195 U.S. 276, 307, 25 S.Ct. 58, 68, 49 S.Ct. 193 (1904). Second, the parties are bound to the rendering court's decision with respect to that issue. Id. at 299, 25 S.Ct. at 64. A litigant is therefore susceptible to being denied her due process right of having an opportunity to be heard on each issue of her case, duPont v. Southern, 771 F.2d 874, 880 (5th Cir. 1985), unless the recognizing court, before giving preclusive effect to an issue determination, first identifies with specificity what the rendering court allegedly decided and determines it was, indeed, actually decided.

Though similar to issue preclusion in some respects, claim preclusion is a distinct doctrine carrying its own elements. Unlike issue preclusion, which can be asserted offensively or defensively, claim preclusion is an affirmative defense.71 Fed. R. Civ. P. 8(c)(1); Fla. R. Civ. P. 1.110. To invoke claim preclusion, a defendant must prove in a recognizing court that the plaintiff's cause of action was adjudicated on the merits in a previous case involving the same parties.72 Fla. Dep't of Transp. v. Juliano, 801 So.2d 101, 105 (Fla. 2001). Thus, under both Florida and federal law, claim preclusion carries four elements: (1) "a final judgment on the merits"; (2) a "decision ... rendered by a court of competent jurisdiction"; (3) "the same cause of action ... involved in both cases"; and (4) "the parties, or those in privity with them, are identical in both suits."73 Baloco v. Drummond Co., 767 F.3d 1229, 1246 (11th Cir. 2014).

Like issue preclusion's actually decided requirement, elements (1) and (3) of claim preclusion are ubiquitous and deeply ingrained because they help protect parties' due process rights.74 Element (1), the final-judgment requirement,75 has long been a "cardinal rule" in Florida and all other traditional common-law jurisdictions. Douglas III, 110 So.3d at 438 (Canady, J., dissenting) (quoting Juliano, 801 So.2d at 105) (citing Kimbrell v. Paige, 448 So.2d 1009, 1012 (Fla. 1984)). The requirement is important because a defendant who successfully invokes claim preclusion bars a plaintiff from litigating a previously adjudicated cause of action, both as to "issues that were raised ... [and] issues that could have been raised but were not raised in the first case."76 Juliano, 801 So.2d at 105. Barring a cause of action that was never fully litigated to a final judgment unjustly "blockades [an] unexplored path[] that may lead to the truth." Brown v. Felsen, 442 U.S. 127, 132, 99 S.Ct. 2205, 2210, 60 L.Ed.2d 767 (1979).

Element (3) of claim preclusion, the same-cause-of-action requirement, has similar constitutional significance. Litigants enjoy a "due process right to fully and fairly litigate each issue in their case." duPont, 771 F.2d at 874; see also Bell v. Burson, 402 U.S. 535, 542, 91 S.Ct. 1586, 1591, 29 L.Ed.2d 90 (1971) ("It is a proposition which hardly seems to need explication that a hearing which excludes consideration of an element essential to the decision ... does not meet [the requirements of the Due Process Clause]."). Claim preclusion — which bars litigation both as to issues that were and were not litigated in a prior case, Juliano, 801 So.2d at 105 — stands in tension with this due process right. The doctrine is reconciled with due process by means of the same-cause-of-action requirement, which functions to "bar[] only those claims that could have been raised in the prior litigation." Griswold v. County of Hillsborough, 598 F.3d 1289, 1293 (11th Cir. 2010) (emphasis added); see also Dennard v. State, No. SC15-300, 2016 WL 1252516, at *2 (Fla. Mar. 30, 2016) (explaining that res judicata only extends to "claims that could have been raised in the prior action" (emphasis in original) (quotation marks and citation omitted)).

B. Res Judicata 102: Procedures to Invoke Issue and Claim Preclusion

When applied properly, issue and claim preclusion facilitate the worthy aim of efficiency: "By `preclud[ing] parties from contesting matters that they have had a full and fair opportunity to litigate,' these two doctrines protect against `the expense and vexation attending multiple lawsuits, conserv[e] judicial resources, and foste[r] reliance on judicial action by minimizing the possibility of inconsistent decisions.'" Taylor v. Sturgell, 553 U.S. 880, 892, 128 S.Ct. 2161, 2171, 171 L.Ed.2d 155 (2008) (alterations in original) (quoting Montana v. United States, 440 U.S. 147, 153-154, 99 S.Ct. 970, 59 L.Ed.2d 210 (1979)). The doctrines, however, carry the risk of depriving litigants of their property without ever affording them an opportunity to be heard on a central element of their case. Hence, recognizing courts should apply the doctrines "only after careful inquiry." Felsen, 442 U.S. at 132, 99 S.Ct. at 2210. "[I]n properly seeking to deny a litigant two days in court, [recognizing] courts must be careful not to deprive him of one." Criales v. Am. Airlines, Inc., 105 F.3d 93, 97 (2d Cir. 1997).

Recognizing courts therefore strictly abide by certain common-law procedures designed to help protect the integrity of their proceedings and litigants' due process rights. Such procedures are so ubiquitous and rudimentary that litigants and courts have had little, if any, reason to test their boundaries. See Honda Motor Co., Ltd. v. Oberg, 512 U.S. 415, 430, 114 S.Ct. 2331, 2340, 129 L.Ed.2d 336 (1994) ("Because the basic procedural protections of the common law have been regarded as so fundamental, very few cases have arisen in which a party has complained of their denial."). The rare court that does deviate from, or abrogate, such procedures risks violating litigants' due process rights. See Douglas III, 110 So.3d at 430-31 ("[E]liminating the basic common law protections against an arbitrary deprivation of property violates due process." (citing Oberg, 512 U.S. at 432, 114 S.Ct. at 2341)). I detail some of these procedures in a hypothetical.

A lawsuit is tried to a jury in a rendering court on claims and defenses framed by the plaintiff's complaint and the defendant's answer. After receiving the jury's verdict, the court enters a final judgment for the plaintiff. In doing so, the rendering court does not declare or predict whether, and if so to what extent, a recognizing court will give preclusive effect to its judgment, that is, to any of the claims or defenses or to any of the issues that were litigated. To do so would result in mere dicta, because those determinations are within the recognizing court's sole purview.77

Later, the plaintiff sues the defendant78 in a Title VII action in a different court, a recognizing court. Her complaint alleges several discrete acts of conduct severe or pervasive enough to create a hostile work environment. The defendant denies each allegation. The plaintiff, invoking issue preclusion, then moves the court to strike the defendant's denial of two of the acts on the ground that they were adjudicated in her favor in the previous lawsuit. The defendant opposes the motion, so the court requires the plaintiff — the party with the burden of proof79 — to present the portions of the previous lawsuit's record that establish the adjudication of the issues. The plaintiff responds by introducing from that record the complaint and answer, the jury instructions, the jury's verdict, and the final judgment.

Upon receiving the plaintiff's evidence, the court decides whether to grant her motion to strike. First, the court determines whether the plaintiff has established the elements of issue preclusion under the rendering state's laws.80 Because every state has a presumption against preclusion, recognizing courts must not apply preclusion if any doubt exists that the elements of preclusion have been satisfied. Issue preclusion's actually decided requirement, for example, is stringent: If a rendering court's jury instructions leave "it open to the jury to find for the defendant upon either of ... two [or more] propositions, and the verdict does not specify upon which the jury acted, there can be no certainty that they found upon one rather than the other," and preclusion is inappropriate.81 De Sollar v. Hanscome, 158 U.S. 216, 222, 15 S.Ct. 816, 818, 39 S.Ct. 956 (1895). In other words, if the jury in the previous case could have returned a verdict for the plaintiff without deciding whether the two acts at issue actually occurred, the recognizing court could not grant the motion to strike.82

If, on the other hand, the recognizing court concludes that the plaintiff has met her burden, and preclusion is appropriate under the rendering state's laws, the court will grant the plaintiff's motion unless the defendant objects further. If the defendant objects on due process grounds, the recognizing court must ensure that applying the rendering state's preclusion law will not violate the defendant's due process rights.83 See Hansberry v. Lee, 311 U.S. 32, 40, 61 S.Ct. 115, 117, 85 S.Ct. 22 (1940) ("[When a due process objection is raised] it becomes the duty of [the recognizing court] to examine the course of procedures in both litigations to ascertain whether the litigant whose rights have thus been adjudicated has been afforded ... due process."); Douglas III, 110 So.3d at 430-31 (expressing the same principle); Adams v. State Farm Bureau Life Ins. Co., 493 F.3d 1276, 1285 (11th Cir. 2007) ("[W]e have stated that res judicata can only be applied to an action if it is first shown that doing so would be consistent with due process." (citing Twigg v. Sears & Roebuck & Co., 153 F.3d 1222, 1226 (11th Cir.1998))). In conducting its due process inquiry, the recognizing court must determine (a) whether the determination in the rendering court was made with adequate notice and opportunity to be heard, (b) whether state preclusion law contains adequate safeguards to ensure that courts do not arbitrarily deprive litigants of property,84 and (c) whether such safeguards were, in fact, applied.

To conduct its inquiry appropriately, the recognizing court must "look past the linguistic label[s] employed by the [rendering court]" and conduct a meaningful review.85 Davila v. Delta Air Lines, Inc., 326 F.3d 1183, 1189 (11th Cir. 2003); see also Gooch v. Life Inv'rs Ins. Co. of Am., 672 F.3d 402, 420-21 (6th Cir. 2012) ("[In conducting this inquiry] it is incumbent upon us to apply the same scrutiny to state-court judgments that the Supreme Court would apply."); Criales v. Am. Airlines, Inc., 105 F.3d 93, 97 (2d Cir. 1997) ("[W]e would not permit the choice of labels to distort substance, especially where the consequence would be so drastic as to deprive a party of the opportunity to be heard."). If its due process inquiry so warrants,86 the recognizing court then grants the plaintiff's motion to strike.

Given the essential inquiries for which a recognizing court is responsible, a rendering court cannot "predetermine the res judicata effect of [its] judgment."87 Matsushita Elec. Indus. Co. v. Epstein, 516 U.S. 367, 396, 116 S.Ct. 873, 888, 134 L.Ed.2d 6 (1996) (Ginsburg, J., concurring in part and dissenting in part) (citation omitted). This is so even if the rendering court, like the Supreme Court in Engle III, is convinced that its proceedings were constitutionally sound.

III.

ENGLE III INSTRUCTED PROGENY COURTS TO DISREGARD TRADITIONAL RES JUDICATA LAW SO AS TO HOLD THE DEFENDANTS LIABLE WITHOUT REGARD TO THE PHASE I FINDINGS

When Engle III accepted jurisdiction under Article V, Section 3(b)(3) of the Florida Constitution, it assumed the role of an appellate rendering court, reviewing the proceedings in Engle I and Engle II for certain issues that were "properly briefed[,] argued and [] dispositive of the case." Murray v. Regier, 872 So.2d 217, 225 n.5 (Fla. 2002) (citing Savona v. Prudential Ins. Co. of America, 648 So.2d 705, 707 (Fla.1995)). The Florida Supreme Court fulfilled this role by considering the briefed issues and quashing much of the Third District's judgment.

When Engle III retroactively certified an issues class limited to eight of the ten of Phase I findings, and declared that those "findings ... will have res judicata effect" in future "damages actions" to be brought by individual members of the decertified class, it usurped the role of a recognizing court.88 Engle III, 945 So.2d at 1269. Recognizing progeny courts could have disregarded Engle III's res judicata instruction as mere dicta,89 and some did.90 Many others, however, in deference to the state's highest court, interpreted the instruction as a binding mandate. See, e.g., R.J. Reynolds Tobacco Co. v. Martin (Martin II), 53 So.3d 1060, 1066-67 (Fla. 1st Dist. Ct. App. 2010) (interpreting the Florida Supreme Court's res judicata instruction as a mandate that "district courts of appeal do not have the prerogative to overrule"); Jimmie Lee Brown II, 70 So.3d at 715 ("We are constrained by the Florida Supreme Court's decision in Engle III."). As shown below, recognizing courts that interpreted the instruction as a mandate treated preclusion as a foregone conclusion, thereby abandoning their recognizing-court duties and putting their integrity at risk while sparing progeny plaintiffs their burden of proving the elements of preclusion.

In accordance with mutuality requirements under Florida preclusion law,91 we are not bound by previous recognizing-court determinations.92 Nevertheless, I review such cases to demonstrate how progeny courts have incrementally grown ever-more absurd in their reasoning, ever-more disingenuous in their portrayal of facts, and ever-more cavalier in their abrogation of due process. The Majority's opinion is best understood in the context of the steady downslide that preceded it.

A. The U.S. District Court for the Middle District of Florida in Brown I Rejected the Florida Supreme Court's Interference with Its Duties as a Recognizing Court

Within the one-year limitations period Engle III provided, 9,000 class members — smokers and personal representatives of deceased smokers — filed suit against the Engle defendants in state and federal court, the "Engle-progeny cases."93 Approximately 4,000 members brought suit in the Circuit Court of Duval County, Florida.94 The tobacco companies, invoking the Class Action Fairness Act of 2005, Pub. L. No. 109-2,119 Stat. 4 (codified in scattered sections of 28 U.S.C.), successfully removed the cases to the U.S. District Court for the Middle District of Florida.95 Graham v. R.J. Reynolds Tobacco Co. (Graham I), No. 3:09-cv-13603-MMH-JBT (M.D. Fla. May 28, 2013) was one of them.

After the cases removed to the Middle District of Florida were assembled,96 the tobacco companies moved the District Court in one of the cases, Brown v. R.J. Reynolds Tobacco Co. (Brown I), 576 F.Supp.2d 1328 (M.D. Fla. 2008),97 to decide the preclusive effect, if any, of the Phase I findings based on Engle III's declaration that "the Phase I common core findings [it] approved will have res judicata effect" in the progeny cases.98 Engle III, 945 So.2d at 1269. The District Court granted the motion.

The preclusion issue was framed by Brown I's amended complaint99 and the defendants' answers.100 I quote parts of these pleadings because they set the stage for, and were integral to, the District Court's decision.

The amended complaint was materially identical to the complaints filed in the other Engle-progeny cases in that all asserted the same Engle III-approved tort claims and sought compensatory and punitive damages. None of the complaints specified the brand(s) of the defendants' cigarettes the plaintiff smoked, how the defendants' tortious conduct caused the plaintiff's injuries, or even what the tortious conduct was in the first place. The facts on which a specific tort claim rested consisted of a citation to the Engle III decision and the Phase I findings. Amended Complaint at 1, 5, 12-14, Brown I, 576 F.Supp.2d 1328 (No. 3:07-cv-00761).

I begin with the pertinent allegations of the complaint and then move to the defendants' answers.

AMENDED COMPLAINT

Plaintiffs, as Personal Representatives of the Estates of Decedents, hereby sue the Defendants as follows:

Id. at 2-7.

The complaints in Brown I and the other progeny cases were pleaded pursuant to Rule 8(a) of the Federal Rules of Civil Procedure101 and the Supreme Court's instructions in Ashcroft v. Iqbal, 556 U.S. 662, 129 S.Ct. 1937, 173 L.Ed.2d 868 (2009). Iqbal requires that

Id. at 678-79, 129 S.Ct. at 1949-50.

The amended complaint did not satisfy Rule 8(a) and Iqbal's pleading standards because it merely recited conclusory statements from the Engle III opinion. Moreover, it failed altogether to identify the tortious conduct that caused the plaintiffs' injuries.

In drafting their amended complaints in Brown I and other progeny cases, progeny plaintiffs simply lifted language from the Engle III opinion as a means of pleading res judicata offensively, using the doctrine as a substitute for alleging the facts necessary to establish the elements of their causes of action. E.g., Amended Complaint at 4, Brown I, 576 F.Supp.2d 1328 (No. 3:07-cv-00761). This method of pleading is foreign to what Rule 8 prescribes102 and helps explain the difficulty federal district courts have encountered in adjudicating progeny cases.

Such difficulty could have been mitigated if the following procedures had been followed: To satisfy the requirements of Rule 8 and Iqbal, plaintiffs' counsel should have drafted a condensed version of the Phase I complaint, one tailored to the individual plaintiff's claims. With respect to negligence, for example, the plaintiffs' complaints should have identified each defendant's negligent conduct, noting when it took place and explaining how it caused the plaintiff's injury. Each defendant, in turn, would answer the complaint and might choose to deny inter alia (1) that it engaged in the alleged conduct, (2) that such conduct was negligent, and (3) that such conduct caused the plaintiff's injuries. Next, the plaintiff, invoking res judicata, would likely respond with a motion to strike denials (1) and (2) as being foreclosed by the Phase I findings and/or by Engle III. If a defendant then opposed that motion, the plaintiff, having the burden of proof on the applicability of preclusion, would introduce into evidence the part of the Engle proceedings that established the foreclosure. The court would then examine such evidence and rule.103

That things did not operate in this way suggests that Engle-progeny cases have more to do with a change in substantive law than with an invocation of traditional claim or issue preclusion. If, for example, Engle III established as a substantive rule of tort law (1) that all cigarettes are defective, unreasonably dangerous, and negligently made and (2) that one who smokes them can recover damages for smoking-related injury — because it is conclusively presumed that the defect or negligence caused the injury — then the way progeny cases have been pleaded makes more sense. If that were the tort law,104 the class plaintiffs would satisfy the Iqbal standard simply by alleging that they purchased a defendant's cigarettes, became addicted, and suffered injury as a result. That is precisely what they have been allowed to do; they neither allege in their complaints nor proffer evidence that the defendants wrongful conduct caused their injuries. Even under this scenario, however, if a defendant moved to dismiss a claim under Rule 12(b)(6), the court would have to determine whether Engle III actually implemented such a change in substantive tort law — that it relieved the class plaintiffs of the burden of proving injury causation. That determination, in turn, would hinge on a review of the Engle proceedings.

In contrast to the plaintiffs' deficient complaints, the Engle defendants' answers have been pleaded in accordance with the Federal Rules of Civil Procedure. Those answers admit or deny the plaintiffs allegations and assert affirmative defenses. Below, I provide RJR's answer from Brown I as a template of a typical Engle-defendant answer. The answer begins with a Preliminary Statement, which is followed by a response to each of the amended complaint's numbered paragraphs and thirty-four affirmative defenses.105

PRELIMINARY STATEMENT

Answer, Defenses, and Jury Demand of Defendant R.J. Reynolds Tobacco Co. at 2-7, Brown I, 576 F.Supp.2d 1328 (No. 3:07-cv-00761) (emphasis added).

In deciding the preclusion issue, the District Court, sitting as a recognizing court, and the parties drew on Florida's res judicata doctrines, claim and issue preclusion. In briefing the preclusion issue, plaintiffs' counsel argued that the Engle III Court issued four implied holdings. The first three holdings relate to the tort claims pleaded in the class action complaint. The fourth relates to the duty of recognizing trial courts in progeny cases.

Plaintiffs first argued that Engle III, by invoking "res judicata" — which the plaintiffs interpreted as claim preclusion — implicitly held that the Phase I findings established the elements of the plaintiffs' tort claims.106 Plaintiffs Response to Tobacco's Rule 16(c) Motion at 4, Brown I, 576 F.Supp.2d 1328 (No. 3:07-cv-00761). Relatedly, the plaintiffs argued, the Engle III Court also implicitly held that the Phase I findings foreclosed all of the tobacco companies' defenses to the plaintiffs' tort claims.107 See id. ("The [Phase I] jury is ... conclusively presumed to have considered all issues related to the claims of defect, negligence, conspiracy to defraud, and the other counts."). Defendants could, of course, still contend that a plaintiff was not a class member — because, for example, she was not addicted — that her disease was not caused by smoking, that she was comparatively negligent, and that her damages should be less than those demanded. Id. at 19. In sum, the plaintiffs argued, Engle III held that progeny trials would be "`damages' trials" rather than causation trials; plaintiffs had no need to prove that their injury was caused by a defendant's tortious conduct. Id. Instead, as plaintiffs argued Engle III also held, the only issue of causation the plaintiffs needed to establish was "that smoking cigarettes caused a plaintiff's particular injury."108 Id. (emphasis in original).

Finally, the plaintiffs also argued that by commanding recognizing trial courts to give the Phase I findings res judicata effect, the Engle III Court was informing those courts that it had predetermined the preclusive effect of the Phase I findings. See id. at 13 ("We know — from Engle itself — that Florida law permits a verdict of this type to be given res judicata effect." (emphasis in original)). Thus, under Florida law, the plaintiffs contended, recognizing trial courts no longer had any business evaluating whether the elements of preclusion — including the final-judgment and actually decided requirements109 — had been satisfied or whether due process had been afforded to Engle defendants in the rendering court. See id. at 12-13 ("[I]t is the law of Florida, deemed so by Florida's highest court, that whatever Phase I can be called or labeled, it is sufficient to be the basis of claims preclusion." (emphasis in original)). If due process had been denied, so be it.

Before it addressed the preclusive effect that should be afforded to the Phase I findings, the District Court had to decide a preliminary question concerning its jurisdiction. The plaintiffs argued that the Rooker-Feldman doctrine110 deprived the Court of jurisdiction to "independently review[] the state court rulings." Brown I, 576 F.Supp.2d at 1334. That is, the Court could not entertain the defendants' argument that because it was "impossible to know what allegations formed the basis of each [Phase I] finding, affording preclusive effect to the general Phase I findings would be an arbitrary application of the common law rules of preclusion" and a denial of due process of law. Id. at 1344. Due to the lack of jurisdiction, the plaintiffs continued, the Court had to apply preclusion without evaluating the due process implications of doing so. Id. at 1334. The District Court rejected the plaintiffs' arguments and held Rooker-Feldman inapplicable.111

Next, the Court considered what preclusive effect it should give to the Phase I findings. The defendants argued that it was "apparent that the Florida Supreme Court intended that the findings function as issue preclusion (or collateral estoppel) in subsequent proceedings," id. at 1338, but that using the findings to establish their liability in the instant case "would be an arbitrary application of the common law rules of preclusion" and thus would deny them due process. Id. at 1344-45. The plaintiffs' response was that "the Engle findings should act as claim preclusion (or res judicata) since the Supreme Court of Florida explicitly used the legal term `res judicata' in its decision," and the defendants had received all the process they were due in the Engle proceedings. Id. at 1338.

The Court considered the plaintiffs' argument "problematic." Id. at 1339. First, the Florida Supreme Court, as the

Id. at 1339 (citations omitted). "Second, as acknowledged by the Florida Supreme Court, the Phase I jury verdict did not establish liability as to any Defendant."113 Id. at 1340. Thus, "the Phase I findings did not serve to merge the claims asserted by Plaintiffs into an enforceable judgment" against the tobacco companies.114 Id.

Although claim preclusion was not viable due to the absence of a final judgment, the District Court assessed whether plaintiffs could successfully assert issue preclusion. Id. In conducting its assessment, the Court performed its recognizing-court duties of evaluating the due process implications of preclusion and determining whether the party asserting preclusion had carried the burden of proving its elements.115 Id. at 1340-47. Issue preclusion, the Court determined, could not be invoked because neither its Florida-law nor constitutional requirements had been satisfied.

The Court found that the plaintiffs had not carried their burden of proving issue preclusion's actually decided element, a requirement that carries constitutional significance.116 According to the Court, the "jury form, and any verdict delivered from the form" were "flaw[ed]" and "nonspecific[]" such that "this Court `would have to embark on sheer speculation' to determine what issues were actually decided during the Phase I trial and how to apply them to the individual claims before this Court." Id. at 1342 (quoting Hoag v. New Jersey, 356 U.S. 464, 472, 78 S.Ct. 829, 829, 2 L.Ed.2d 913 (1958)). The Court simply could not determine "what acts or omission committed by what Defendant breached what duty to which Plaintiff causing what injury." Id. Accordingly, to preclude defendants from litigating such issues would violate the Supreme Court's instruction "that courts not apply the doctrine of issue preclusion to prior determinations unless the court `is certain that the precise fact was determined by the former judgment.'" Id. at 1345 (quoting De Sollar v. Hanscome, 158 U.S. 216, 221, 15 S.Ct. 816, 39 S.Ct. 956 (1895)).

Moreover, "since it is impossible to determine the precise issues decided by the Phase I jury ... the traditional elements of issue preclusion — e.g., identicality, criticality, and necessity to the prior determination — cannot be satisfied." Id. at 1346. Accordingly, the Court concluded that it was "foreclosed from applying the Phase I findings as establishing any part of Plaintiffs' claims." Id. (citations omitted).

Because the plaintiffs were unsuccessful in invoking claim and issue preclusion, the Engle defendants had the right to deny that their tortious conduct caused the plaintiffs' injuries. In its order rejecting the plaintiffs' arguments that Rooker-Feldman precluded it from deciding the due process issue the companies had presented, the District Court certified that its rulings qualified for interlocutory appeal under 28 U.S.C. § 1292(b). Id. at 1348. We agreed and granted the parties' application to appeal. Notice of Interlocutory Appeal at 1, Brown I, 576 F.Supp.2d 1328 (No. 3:07-cv-00761). Meanwhile, further proceedings in the Engle-progeny cases in the Middle District were stayed pending our decision. E.g., Order at 1-2, Waggoner v. R.J. Reynolds Tobacco Co., 835 F.Supp.2d 1244 (2011) (No. 3:09-cv-10367).

* * *

Although Engle III sought to predetermine preclusion such that recognizing courts would not consider whether the elements of preclusion had been satisfied or whether applying preclusion would deny due process, the Brown I Court firmly rejected this attempted usurpation of its recognizing-court responsibilities — "the rendering court," the Court affirmed, "may not decide the preclusive effect of its own judgments."117 Brown I, 576 F.Supp.2d at 1339. In performing its recognizing-court duties, the Court found that under both Florida law and the U.S. Constitution, plaintiffs could invoke neither claim or issue preclusion.

B. In Brown II, We Upheld the District Court's Decision as a Recognizing Court to Apply Florida's Traditional Issue-Preclusion Doctrine to the Phase I Findings

On appeal, we affirmed the District Court's rejection of the plaintiffs' Rooker-Feldman argument for the reasons that Court gave and in light of the Supreme Court's recent decision in Exxon Mobil Corp. v. Saudi Basic Indus. Corp., 544 U.S. 280, 125 S.Ct. 1517, 161 L.Ed.2d 454 (2005).118 Brown II, 611 F.3d at 1330.

We also affirmed the District Court's rejection of the plaintiffs' argument that claim preclusion, rather than issue preclusion, was what the Engle III Court had in mind when it used the term "res judicata." Although "the plaintiffs argued before the district court and suggested in their brief to this Court that the Florida Supreme Court was referring to claim preclusion in Engle III," the plaintiffs, at oral argument "clarified that their position [was] that the Phase I approved findings are entitled to issue preclusive effect." Id. at 1333 n.7. "[I]f the plaintiffs had continued to argue for claim preclusion, we would have rejected that position" because claim preclusion's final-judgment requirement had not been satisfied. Id. at 1332, 1333 n.7.

After the plaintiffs stipulated that claim preclusion was not viable, we evaluated the viability of issue preclusion. "Issue preclusion," we observed, "operates more narrowly to prevent re-litigation of issues that have already been decided between the parties in an earlier lawsuit." Id. at 1332 (citations omitted). Like the District Court before us, we recognized that Florida's doctrine of issue preclusion carried an actually decided requirement.119 Id. (citing Rohan v. Trakker Maps, Inc., 633 So.2d 1176, 1177 (Fla. Dist. Ct. App. 1994)). Unlike the District Court, however, we did not take any position as to whether that requirement was required under the U.S. Constitution, "because," we assumed,120 "under Florida law the findings could not be used" "to establish facts that were not decided by the [Phase I] jury." Id. at 1334.

The parties' dispute, therefore, came down to what the Phase I jury decided. The defendants argued that the jury decided only what it indicated on the Phase I verdict form — "those [facts] framed by the specific factual issue set out in the questions posed to them on the verdict form." Id. The plaintiffs, on the other hand, advocated a more expansive reading of the Phase I findings that relied on "flesh[ing] out" "the jury's answers" "using the record as a whole" and "going outside the record." Id. at 1335. This process of "fleshing out," the plaintiffs contended, would lead the Court to conclude that when the jury answered "yes" to, for example, the verdict-form question about defendants "plac[ing] cigarettes on the market that were defective and unreasonably dangerous," it meant that "all cigarettes the defendants sold were defective and unreasonably dangerous." Id. (emphasis added).

Though we welcomed the plaintiffs to scour the trial record — without looking beyond it — for proof of what the jury determined, we were skeptical that such proof existed. Id. "[T]he plaintiffs have pointed to nothing in the record, and there is certainly nothing in the jury findings themselves," we observed, "to support [the plaintiffs'] factual assertion" that the Phase I jury found that defendants' tortious conduct tainted all cigarettes.121 Id.

With the dispute over claim and issue preclusion resolved, and the due process issue avoided, we remanded the case to the District Court to provide plaintiffs an opportunity to prove,122 inter alia, that "the jury's [unreasonably-dangerous-defect finding] ... establishes that all of the cigarettes that the defendants sold" "were defective and unreasonably dangerous."123 Id. at 1336.

* * *

Brown II, as a recognizing-court decision, became the Eleventh Circuit's controlling precedent regarding the preclusive effect of Engle III in the litigation of progeny cases in the district courts. The Phase I findings resolved factual issues, not causes of action. Id. at 1333 ("[F]actual issues and not causes of action were decided in Phase I."). Absent evidence that the Phase I jury decided more facts than those it disclosed in its findings — which we were skeptical existed, but welcomed plaintiffs to locate — plaintiffs could not rely upon the Phase I findings to identify particular cigarette defect(s) and tortious conduct, let alone prove that such defect(s) and conduct caused their harm. See id. at 1335 ("[T]here is certainly nothing in the jury findings ... to support [the plaintiffs'] factual assertion" "that all cigarettes the defendants sold were defective and unreasonably dangerous."). By necessary implication, we held that a District Court judgment based solely on the Phase I findings would deprive the defendant of its property without due process of law.124

C. The Florida District Courts of Appeal Rejected Brown II on the Basis of Engle III's Instruction

Brown II was decided on July 22, 2010. After the mandate issued, the stays were lifted in twelve "lead" Middle District of Florida cases, including Graham.125 The Court in Waggoner v. R.J. Reynolds Tobacco Co., 835 F.Supp.2d 1244 (M.D. Fla. 2011), was selected as the recognizing court for "determin[ing] which Engle facts should be given preclusive effect under Florida law as the Eleventh Circuit had outlined it" in Brown II.126 Id. at 1253. The Waggoner Court was unable to undertake the assignment, however, until June 2011.

On June 2, 2011, the lawyers representing the parties in the twelve lead cases appeared before the District Court for a Rule 16(c)127 management conference. Id. at 1256. In the interim, between the issuance of the mandate in Brown II and the commencement of the Rule 16(c) conference, the First District Court of Appeal, in Martin II, 53 So.3d 1060,128 had reached a preclusion holding contrary to Brown II's. Before the Rule 16(c) conference adjourned, the Fourth District Court of Appeal, in Jimmie Lee Brown II agreed with Brown II's analysis, but reached the same result as the First District's in Martin II, albeit with serious reservations as to whether its decision would deny the tobacco companies of their property without due process of law.

1. The Martin I Circuit Court Concluded That Engle III's Instruction Required It to Hold the Defendants Liable if the Plaintiff Simply Proved Class Membership Irrespective of the Phase I Findings

The Martin case was brought on October 24, 2007, in the Circuit Court for Escambia County, Florida. Beverly Martin sued RJR, Philip Morris USA, Inc., and Lorillard Tobacco to recover for the death of her husband, Benny Martin. Martin II, 53 So.3d at 1064 n.2. In her first amended complaint,129 she asserted four of the Engle III-approved causes of action: strict liability, negligence, fraud by concealment, and conspiracy to commit fraud by concealment. Id. at 1065. She also sought punitive damages. Id. Her tort claims and the prayer for punitive damages were based solely on the "Common Liability Findings" asserted in the Brown I complaint. Amended Complaint at 4, Brown I, 576 F.Supp.2d 1328 (No. 3:07-cv-00761). Those findings, the complaint alleged, were sufficient to "conclusively establish" her tort claims. First Amended Complaint at ¶¶ 29-61, Martin v. R.J. Reynolds Tobacco Co. (Martin I), No. 2007-CA-2520 (Fla. Cir. Ct. 2009).

The defendants' answers to the amended complaint raised the same due process objection as their answers did in Brown I. Philip Morris USA Inc.'s Answer, Martin I (No. 2007-CA-2520), 2008 WL 6722672 at *12. On August 25, 2008, their attorneys and those representing Engle defendants in the other progeny cases pending in the Escambia County Circuit Court jointly moved the Circuit Court, pursuant to Florida Rule of Civil Procedure 1.200,130 to determine, as a recognizing court, what the Florida Supreme Court meant when it declared that the Phase I findings would be given "res judicata effect" in progeny cases. Defendants' Rule 1.200 Motion at 1-3, In re: Engle Progeny Cases Tobacco Litigation (Fla. Cir. Ct. Feb. 24, 2009) (No. 2008-CA-80000). The parties asserted positions similar to those they asserted in Brown I. Id.; Order Denying Defendants' Rule 1.200 Motion at 1-3, In re: Engle Progeny Cases Tobacco Litigation (No. 2008-CA-80000).

The Court ruled on the motion in an order entered on February 24, 2009. Order Denying Defendants' Rule 1.200 Motion, In re: Engle Progeny Cases Tobacco Litigation (No. 2008-CA-80000). It could not say whether the Florida Supreme Court had intended to invoke "res judicata, collateral estoppel, estoppel by judgment, stare decisis, or some other mechanism." Id. at 3. Whatever the mechanism was, it was "unorthodox." Id. at 2. The intended effect of the mechanism, however, was clear: the Florida Supreme Court had intended to facilitate rather than "void the class action litigation." Id. at 3. The Phase I findings "must [be] use[d]" even if they appeared useless. Id.

The Court's Rule 1.200 ruling governed the trial in Martin I. All Ms. Martin had to prove to hold RJR liable was Mr. Martin's class membership — his addiction to an RJR cigarette and a smoking-related injury.131 Martin II, 53 So.3d at 1066. Thus, the jury was not required to determine whether the cigarettes Mr. Martin smoked were defective and unreasonably dangerous; whether RJR committed one or more negligent acts that caused him to smoke; whether it concealed information about the health effects or addictive nature of smoking that would have caused him to stop smoking had he been aware of it; or whether any Engle defendant acted to conceal such information pursuant to a conspiracy of which RJR was a member.132 Id. at 1064-66; Jury Instructions, Martin I (No. 2007-CA-2520), 2009 WL 2599305. The jury found RJR liable on all four claims and assessed Ms. Martin's damages at $5 million. Martin II, 53 So.3d at 1066. That amount was reduced to $3.3 million based on the jury's apportionment of fault. Id. Ms. Martin was also awarded $25 million in punitive damages. Id. RJR appealed the judgment.

2. The First District Court of Appeal in Martin II Agreed That Engle III's Instruction Required It to Hold the Defendants Liable to all Class Members Irrespective of the Phase I Findings

The appeal was "the first ... `Engle progeny' case to reach a district court of appeal following the Florida Supreme Court's decision" in Engle III. Id. at 1062, 128 S.Ct. 96. The "crux" of RJR's appeal, as the First District saw it, was "the extent to which an Engle class member can rely upon the findings from the class action when she individually pursues one or more Engle defendants for damages." Id. In other words, to what extent could Ms. Martin use the Engle findings to establish the elements of her claims? Id. Reiterating its argument from previous cases, RJR pointed out that the Phase I findings

Id. Therefore, RJR contended, Ms. Martin should have been required to prove, and RJR should have been allowed to contest, that the brand of cigarettes Mr. Martin smoked was defective, unreasonably dangerous, and negligently produced.133 Id. Further, Ms. Martin should have been required to identify the particular conduct the jury deemed tortious and the particular product feature(s) the jury deemed defective and prove that such conduct and feature(s) caused Mr. Martin's injury. Id. Because the Circuit Court had simply presumed that RJR's tortious conduct caused her husband's injuries, it had violated RJR's due process right to litigate essential elements of its case.134 Id.

The First District rejected RJR's characterization of the Phase I findings. In doing so, it did not look to the jury instructions or the special interrogatories. Instead, it looked to the Engle Omnibus Order for interpretational assistance. In that order, Judge Kaye, who had tried Phases I and II, determined that "the plaintiff[s] ha[d] presented evidence that could support [the Phase I findings]." Friedrich v. Fetterman & Assocs., 137 So.3d 362, 365 (Fla. 2013) (emphasis added). In other words, "There was more than sufficient evidence at trial to ... support the jury verdict." But the First District cited the Omnibus Order, not for what the Phase I jury could have determined, but for what it did determine. That a properly instructed jury could have determined that the "findings encompassed all brands" was, to the Martin II Court, proof that the jury made such a determination. Id. at 1068. That a properly instructed jury could have "determined the defendants ... breached their duty [to all class members] by [negligently] selling [defective] cigarettes" was proof that the jury determined that as well. Id. The First District implemented this strange sufficiency-of-the-evidence standard throughout.135 See, e.g., id. at 1069 ("[T]he record contains abundant evidence from which the jury could infer Mr. Martin's reliance on pervasive misleading advertising campaigns for the Lucky Strike brand in particular and for cigarettes in general." (emphasis added)).

Why did the First District, as a recognizing court, take upon itself the plaintiff's burden of proving what the Phase I jury decided? Why did it lighten that burden from one of necessary inference136 to sufficiency of the evidence? Why, in doing so, did it ignore the jury instructions and verdict form in favor of an inapposite ruling on a motion for directed verdict? The First District's interpretation of Engle III provides insight: RJR's characterization of the Phase I findings had to be rejected because it would "nullify" "the supreme court's [Engle III] decision" and "district courts of appeal do not have the prerogative to overrule Florida Supreme Court precedent." Id. at 1066-67 (citation omitted). Engle III had predetermined the res judicata question, and that was good enough for the First District, "[n]o matter the wording of the findings on the Phase I verdict form." Id. at 1067.

Although the First District perceived that the Florida Supreme Court wanted it to preclude defendants' defenses, it, like the Circuit Court in its Rule 1.200 order, could not tell which preclusion doctrine the Supreme Court had intended to invoke. The First District purported to "find it unnecessary to distinguish between" "issue preclusion versus claim preclusion." Id. Nevertheless, the Court expressly disavowed Brown II's assertion that plaintiffs, in accordance with issue preclusion's actually decided requirement, had to "trot out the class action trial transcript to prove applicability of the Phase I findings."137 Id.

Recall that the plaintiffs in Brown II had stipulated that the Florida Supreme Court had invoked issue, rather than claim, preclusion in Engle III. Brown II, 611 F.3d at 1333 n.7. Recall also that plaintiffs in that case had requested an opportunity to "flesh out" the Phase I verdict form "using the record as a whole." Id. at 1335. Here, the First District rejected issue preclusion and its actually decided requirement, because "[s]uch a requirement undercuts the supreme court's ruling in [Engle III]."138 Martin II, 53 So.3d at 1067.

In sum, the driving force behind the First District's unusual analysis was its interpretation of Engle III. It upheld the Circuit Court because that Court "correctly construed Engle [III] and instructed the jury accordingly on the preclusive effect of the Phase I findings." Id. at 1069.

RJR petitioned the Florida Supreme Court for review, but the Court declined in an opinion stating,

R.J. Reynolds Tobacco Co. v. Martin, 67 So.3d 1050 (2011) (Table). The U.S. Supreme Court denied RJR's petition for a writ of certiorari.

* * *

In entertaining RJR's appeal, the First District faced a compelling constitutional argument. As RJR contended, the plaintiff had invoked "the doctrine of res judicata... to prevent any jury determination of the critical facts on which [the plaintiff's] claims turn." Reply Brief of Appellant at 1, Martin II, 53 So.3d 1060 (No. 1D09-4934). RJR had been precluded from contesting, and Ms. Martin had been spared the burden of proving, that RJR's tortious conduct caused her late husband's injury. The Phase I findings on which such expansive preclusion had been premised plainly "[did] not establish that there was a defect in the Lucky Strike cigarettes smoked by Mr. Martin, let alone one that caused his death." Id. at 2. Nor did the findings "establish any negligent conduct, concealed information, or conspiratorial conduct that caused Mr. Martin's death." Id. The First District knew this. It knew it because the Phase I jury was not tasked with determining whether an Engle defendant's conduct caused a class member's injury. That determination, according to the original trial plan and Engle III, would be made by the progeny juries. But the First District also knew that Engle III, by declaring "res judicata," had signaled or implicitly held that the Phase I findings would assist class members in holding Engle defendants liable. See Martin II, 53 So.3d at 1069 ("[W]e interpret the supreme court's ruling in Engle to mean individual class plaintiffs, when pursuing RJR and the other class defendants for damages, can rely on the Phase I jury's factual findings."). The First District's dilemma, then, was to either acknowledge the worthlessness of the Phase I findings and "essentially nullify" Engle III in the process, Id. at 1066, or ignore the findings' worthlessness and uphold Engle III. Concluding that "district courts of appeal do not have the prerogative to overrule Florida Supreme Court precedent," the First District chose the latter option. Id. (citing Hoffman v. Jones, 280 So.2d 431, 434 (Fla. 1973)).

In so choosing, the First District attempted to mitigate the worthless-findings problem. The Phase I verdict form and jury instructions were transparently worthless, so the First District looked to another source, the Engle Omnibus Order, for interpretational assistance. In that order, Judge Kaye concluded that "the plaintiff[s] ha[d] presented evidence that could support [the Phase I findings]." Friedrich v. Fetterman & Assocs., 137 So.3d 362, 365 (Fla. 2013) (emphasis added). In other words, "[t]here was more than sufficient evidence at trial to ... support the jury verdict." But the First District cited the Omnibus Order, not for what the Phase I jury could have determined, but for what it did determine. That a properly instructed jury could have determined that the "findings encompassed all brands" was, to the Martin II Court, proof that the jury did determine that. Id. at 1068. That a properly instructed jury could have "determined the defendants ... breached their duty [to all class members] by selling cigarettes" was proof that the jury determined that as well. Id.

Ironically, in carrying out the Engle III Court's implicit instruction to hold defendants liable to all class members, the Martin II Court ignored the very explicit instruction from which the implied instruction was derived. The Martin II Court did not give "res judicata effect to certain Phase I findings" as Engle III directed. Engle III, 945 So.2d at 1254. Instead, in blatant disregard of the defendants' jury-trial rights,139 it gave res judicata effect to the evidence presented at the Phase I trial.

This sleight of hand carried class plaintiffs only part of the way to establishing RJR's liability. Even if all cigarettes were defective and unreasonably dangerous, and even if all defendants breached a duty to all class members, plaintiffs still could not prove that an unreasonably dangerous defect or a tortious act caused their harm unless they could identify the unreasonably dangerous defect or tortious act.140

The First District bridged the remaining gap by changing the law rather than the facts. Specifically, it implemented, without saying it was doing so, a conclusive presumption under which class members were allowed to presume rather than prove that defendants' tortious conduct caused their injury. In doing so, the Court disregarded Engle III's holding that "individualized issues such as legal causation" would be litigated in progeny trials. Engle III, 945 So.2d at 1268. Perhaps the Court was interpreting a coded message embedded in Engle III that departed from its explicit language. Or maybe the Court was breaking a lesser commandment in order to keep a greater one. In any case, under Martin II's reasoning, legal causation would be presumed.

The Court explained its reasoning as follows: It acknowledged that "[t]he Phase I jury ... [did] not [determine] `whether any class members ... were injured by Tobacco's conduct.'" Id. at 1067 (quoting Engle III, 945 So.2d at 1256). It held, however, that progeny plaintiffs need not establish that a defendant's tortious conduct caused their harm; rather, they only needed to show that addiction to cigarettes caused their harm. Id. at 1069.

The conclusive presumption the Martin II Court implemented had serious flaws. First, it violated RJR's due process rights.141 The conclusive presumption also violated RJR's right to have a jury determine whether it engaged in the wrongful conduct that caused Mr. Martin's death.142

In its attempt to give effect to Engle III's coded instructions, the First District departed far from Engle III's explicit language. Nothing in Engle III foreshadowed Martin II's reasoning. Nothing in Engle III suggests that all cigarettes had been found defective and unreasonably dangerous or that all defendants had been found to breach their duty to all class members. And nothing in the opinion even hints at the conclusive presumption the Court created. In fact, the conclusive presumption runs contrary to Engle III's assertion that legal causation would be litigated rather than presumed in progeny trials. In creating the presumption, the Court substantially altered midstream the elements of the tort claims asserted in the class action complaint and litigated in Phase I. Though the Martin II Court may have been doing its best to give effect to an instruction it interpreted as embedded in the Engle III opinion, the First District's reasoning is so removed from Engle III as to constitute a new substantive ruling. As will be shown below, the Waggoner Court treated it as such.

3. The Fourth District Court of Appeal in Jimmie Lee Brown II Held That Engle III's Instruction Meant Issue Preclusion but That the Plaintiff Did Not Need to Identify a Specific Defect or Negligent Conduct

Jimmie Lee Brown v. R.J. Reynolds Tobacco Co. (Jimmie Lee Brown I), No. 4D09-2664 (Fla. Cir. Ct. 2010), 2009 WL 2493781 was tried to a jury on issues framed by the parties' pleadings. The plaintiff, Jimmie Lee Brown, filed suit on behalf of Roger Brown, who was deceased, on March 1, 2007 in the Circuit Court for Broward County, Florida. Complaint at 1, Jimmie Lee Brown I (No. 4D09-2664). In it, he alleged, echoing the complaints in Brown I and Martin I, that the Common Liability Findings conclusively established the elements of the plaintiff's tort claims. Id. at 5-9. The defendants' answers, in turn, were similar to those filed in Brown I and Martin I. Answer, Jimmie Lee Brown I (No. 4D09-2664).

The trial proceeded in two phases. Jimmie Lee Brown II, 70 So.3d at 711. In the first phase, the Court asked the jury to determine whether the decedent "was a member of the Engle class, i.e. whether he was addicted to RJR cigarettes containing nicotine; and, if so, was his addiction a legal cause of his death." Id. The jury found that Roger Brown was an Engle class member.143 Id. In the second phase, a trial involving claims of strict liability and negligence,144 the Court gave the jury binding instructions similar to those in Martin I. Jury Instructions, Jimmie Lee Brown I (No. 4D09-1664), 2009 WL 2599305 at 1-12. The jury assessed Ms. Brown's damages at $1.2 million, which was reduced to $600,000 due to the decedent's fault. Jimmie Lee Brown II, 70 So.3d at 714.

In appealing the judgment to the Fourth District, RJR repeated the arguments it had made to the District Court in Brown I and to the First District in Martin II: Engle III's res judicata declaration did not relieve the plaintiff of "the burden to prove that RJR committed particular negligent acts in a violation of a duty of care owed to Mr. Brown." Id. Nor did it relieve the burden "to prove that the cigarettes Mr. Brown smoked contained a specific defect that injured Mr. Brown." Id. Because, RJR argued, "res judicata ... necessarily mean[t] ... issue preclusion ... post-Engle plaintiffs must demonstrate that the issues on which they seek preclusion were `actually litigated' in [Phase I]." Id.

The Fourth District agreed. Citing our decision in Brown II, the Court held that contrary to the First District's interpretation,145 the Supreme Court's reference to the res judicata effect of the Phase I findings "necessarily meant issue preclusion, not claim preclusion." Id. at 715. Recall that the First District in Martin II allowed Ms. Martin to hold RJR liable simply by proving class membership — his addiction to an RJR cigarette and a smoking-related injury. Martin II, 53 So.3d at 1066. The Fourth District repudiated that approach, holding instead that the "class membership" jury instruction could not be used "for the dual purpose of satisfying the element of legal causation with respect to addiction and legal causation on the underlying strict liability and negligence claims."146 Jimmie Lee Brown II, 70 So.3d at 714. "[P]laintiffs must prove more than mere class membership and damages." Id. at 715. They must, the Fourth District insisted, prove "legal causation and damages."147 Id. In doing so, they cannot use the Phase I findings "to establish facts that were not actually decided by the jury." Id. (quoting Brown II, 611 F.3d at 1333).

For all its lip service to the defendants' arguments and Brown II's reasoning, however, the Fourth District "[did] not go as far as Brown [II] to require trial courts to evaluate whether ... elements of ... plaintiffs' claims are established by the Engle findings." Id. Specifically, though the Court believed in the necessity of issue preclusion's actually decided requirement, it nevertheless held that plaintiffs were not "required to point to a specific defect" or "specific tortious conduct." Id. at 717, 718. Although it believed in the necessity of proving legal causation, it held that plaintiffs need not do so — they did not need to identify defect(s) or tortious conduct, let alone prove that such defect(s) and conduct caused their harm.

Why the sharp disconnect between analysis and holding? Because the Fourth District was "constrained by the Florida Supreme Court's decision in Engle III." Id. at 715. That decision makes it clear that "conduct ... was determined" and was not to be litigated in progeny cases. Id. at 717. To require plaintiffs to prove "that Tobacco committed particular negligent acts when asserting a negligence claim ... would render the Florida Supreme Court's opinion in Engle III meaningless." Id. at 718 (emphasis added). The Florida Supreme Court had issued a gag order, and the Fourth District had no choice but to obey.

Even the Fourth District's constitutional concerns could not justify a departure from Engle III's mandate. It was "concerned the preclusive effect of the Engle findings violates Tobacco's due process rights." Id. at 716. It was concerned, specifically, that allowing plaintiffs to invoke issue preclusion without its actually decided requirement constituted an "extreme application[] of the doctrine ... inconsistent with a federal right that is `fundamental in character.'" Id. (quoting Postal Tel. Cable Co. v. City of Newport, 247 U.S. 464, 476, 38 S.Ct. 566, 62 S.Ct. 1215 (1918)). Nevertheless, the Court affirmed the Circuit Court's judgment in which plaintiffs were allowed to do just that and thereby hold defendants liable without proving that defendants' wrongful conduct caused harm. Id. at 718. Thus, although the Fourth District disavowed Martin II, which it interpreted as "effectively" implementing "claim preclusion instead of issue preclusion," its approach tracked closely to the First District's — the Fourth District similarly implemented an unconstitutional conclusive presumption that class members' harm was caused by tortious conduct.148 Id. at 716.

Although the Court followed the Engle III mandate, it emphatically voiced its disapproval. In a special concurrence endorsed by the Court, id. at 716, Chief Judge May noted the "confusion in the trial courts" stemming from a "struggle with the extent to which [the Phase I] findings resolve ultimate issues in the trial of individual claims," Id. at 718 (May, J., concurring). He quoted from our Brown II decision to highlight our concern that the "jury findings themselves" provide no indication that "all cigarettes the defendants sold were defective and unreasonably dangerous." Id. at 720 (quoting 611 F.3d at 1335). Likewise, he quoted Justice Wells's dissent from Engle III, lamenting the many questions Engle III left unanswered, including, "How are the findings to be used in cases in which the findings are used?" Id. at 719 (quoting Engle III, 945 So.2d at 1284 (Wells, J., concurring in part and dissenting in part)). Such questions, Judge May, explained left trial courts and litigants no choice but to play "a form of legal poker." Id. at 720. One aspect of the game was clear: "the Engle factual findings are binding." Id. But "a lurking constitutional issue hovers over the poker game: To what extent does the preclusive effect of the Engle findings violate the manufacturer's due process rights?" Id. With this constitutional question, along with many other questions, lurking, "parties to the tobacco litigation [were left to] continue to play legal poker, placing their bets on questions left unresolved by Engle." Id.

In the wake of Chief Judge May's special concurrence, the Florida Supreme Court initially accepted RJR's petition to review the Fourth District's decision. R.J. Reynolds Tobacco Co. v. Jimmie Lee Brown, 133 So.3d 931, 931 (Fla. 2014) (per curiam). But then it changed its mind: "Upon further consideration," the Court explained, "we have determined that we should exercise our discretion and discharge jurisdiction. Accordingly, we hereby dismiss this review proceeding. No motion for rehearing will be entertained by the Court." Id. The Supreme Court denied review because while RJR's petition for review was pending, it agreed to answer a certified question from the Second District Court of Appeal in Philip Morris v. Douglas (Douglas II), 83 So.3d 1002, 1011 (Fla. 2d Dist. Ct. App. 2012), which asked the Florida Supreme Court to determine whether its method of affording res judicata to the Phase I findings denied the Engle defendants' due process rights.

* * *

Jimmie Lee Brown II provides the sharpest illustration of the dilemma facing Florida District Courts of Appeal. Engle III issued a mandate: use the Phase I findings — use them. The final-judgment, actually decided, and due process inquiries that recognizing courts ask before affording preclusive effect to prior adjudications149 were not necessary in progeny cases because the Engle III Court had already predetermined the res judicata effect of the Phase I findings. The District Courts of Appeal understood this much. What they did not understand, however, was how to execute. What preclusion doctrine were they supposed to use when plaintiffs could not satisfy all the elements of either claim or issue preclusion? What were they supposed to say in response to the defendants' legitimate due process concerns?

The First and Fourth Districts took different approaches. The First District dutifully accepted the Engle III mandate and jumped through hoops — distorting facts, disregarding its recognizing-court duties, and remaking Florida tort law — in an attempt to make the mandate work. The Fourth District was less accommodating. It rejected Engle III's invitation to disregard its recognizing-court duties, identifying preclusion-law elements the plaintiffs had not satisfied and the due process deprivations foisted upon defendants. But the Court could do no more than make note of such concerns. It was "constrained" by the state's highest court to implement the Engle III mandate. Jimmie Lee Brown II, 70 So.3d at 715. Thus, despite its resistance, the Fourth District begrudgingly upheld two unconstitutional conclusive presumptions and violated Engle defendants' jury-trial rights just like its more cooperative sister court did in Martin II.150

D. In Light of Martin II and Jimmie Lee Brown II, the Middle District of Florida in Waggoner Ruled That the Preclusive Application of the Phase I Findings to Hold the Defendants Liable Would Not Violate Due Process

The Waggoner Court faced the same "legal-poker-game" questions that previous courts had faced. How should a recognizing court respond to an invitation to abandon its recognizing-court duties? To what extent should it preclude defendants from litigating their case on the basis of Engle III and the Phase I findings? In addressing these questions, the Court was presented with a complaint that the plaintiffs had amended to take advantage of Martin II's favorable holding.151 In their complaint, the plaintiffs asserted the six tort claims Engle III had "approved," basing the claims solely on the Phase I findings.152

As they had from the beginning of the Rule 16(c) management conference, the Engle defendants continued to contend that "the Due Process Clause of the U.S. Constitution bars Plaintiff from using the Engle findings to establish the wrongful-conduct elements of her claims, because she cannot show that any issue as to which she seeks preclusion was actually decided by the Engle jury." Defendants' Rule 16(c) Motion at 1, Waggoner, 835 F.Supp.2d 1244 (M.D. Fla. 2011) (No. 3:09-CV-10367). They cited our Brown II opinion to support their argument that Engle-progeny cases "are clearly governed by issue preclusion and not claim preclusion principles." Id. at 1 n.2.

Recognizing that the District Court, bound by Brown II, would require them to "flesh[] out" the Phase I findings,153 Brown II, 611 F.3d at 1335, the plaintiffs filed a "Brown [II] Proffer purporting to provide Engle record evidence" that elucidated what the Phase I jury decided. Waggoner, 835 F.Supp.2d at 1256. Like the Martin II Court, they relied heavily on the Omnibus Order — which evaluated what a properly instructed jury could have determined — as evidence of what the Phase I jury actually determined.154 Id.; Plaintiffs' Response to Defendants' Rule 16(c) Motion at 18, Waggoner, 835 F.Supp.2d 1244 (M.D. Fla. 2011) (No. 3:09-CV-10367). Adopting Martin II's reasoning, they argued that not only did their class representatives present enough evidence at the Phase I trial that a properly instructed jury could have found that all cigarettes were defective, unreasonably dangerous, and negligently produced, they presented enough evidence that "the jury necessarily determined that all the tobacco companies' cigarettes were defective." Waggoner, 835 F.Supp.2d at 1265 (emphasis added) (citation and quotation marks omitted). The defendants resisted this argument, reminding the Court "that the [Phase I] jury was presented with ... many differing and contradictory theories," id. at 1266, and that class representatives had specifically rejected as "[a] major fiction" the theory that all cigarettes were defective.155 Defendants' Rule 16(c) Motion at 18, Waggoner, 835 F.Supp.2d 1244 (M.D. Fla. 2011) (No. 3:09-cv-10367) (citation omitted).

The plaintiffs adopted Martin II's sufficiency-of-the evidence reasoning to identify what the Phase I jury decided, but they achieved an even greater level of precision than the Martin II Court. Martin II states that "the Phase I jury findings encompassed all the brands," 53 So.3d at 1068, but that opinion never actually identifies the defendants' unreasonably dangerous defect(s) or tortious act(s).156 The Waggoner plaintiffs figured out what the Martin II Court could not: it all came down to nicotine. Phase I "was, at bottom, a case about addiction to cigarettes [containing] nicotine." Waggoner, 835 F.Supp.2d at 1265 (citation and quotation marks omitted). "[I]t was the presence of [nicotine] in every cigarette ... that made all of Defendants' products defective," unreasonably dangerous, and negligently produced. Id. In sum, the plaintiffs interpreted a "year-long [Phase I] trial in which myriad defect, negligence and fraud theories were vigorously litigated," id. at 1276 (emphasis in original), as producing a record so one-sided and straightforward that a jury necessarily must have adopted a single theory of liability, one that the class representatives disavowed at trial and one that could have been litigated in days or weeks.

To the defendants, the Phase I record revealed a very different story, one of hopeless complexity. "[D]ue to the generality of the Phase I findings and the multiple theories of liability advanced in the trial record," the defendants argued, "plaintiffs simply can't get there (due process) from here (a Brown [II] Proffer)."157 Id. at 1266.

In light of the defendants' argument that even "the most thorough Brown [II] Proffer imaginable ... would not satisfy... federal due process," the District Court declined to review the plaintiffs' Proffer in detail. Id. at 1266-67. Evaluating the Proffer would not "decide the issue before the Court." Id. at 1267. Instead, the Court examined the threshold question of whether an actually decided inquiry was even required under the U.S. Constitution. The District Court in Brown I held that it was. Brown I, 576 F.Supp.2d at 1345 (citing De Sollar v. Hanscome, 158 U.S. 216, 221, 15 S.Ct. 816, 39 S.Ct. 956 (1895)). We avoided the question in Brown II because we assumed that Engle III had not sub silentio amended Florida's issue-preclusion doctrine to eliminate its actually decided requirement.158 Brown II, 611 F.3d at 1334. After Martin II and Jimmie Lee Brown II, our assumption appeared faulty,159 and the District Court thus felt the need to address the constitutional question.

The District Court observed that "[s]tate courts are generally free to develop their own rules for protecting against the relitigation of common issues or the piecemeal resolution of disputes." Waggoner, 835 F.Supp.2d at 1267 (quoting Richards v. Jefferson County, 517 U.S. 793, 797, 116 S.Ct. 1761, 1765, 135 L.Ed.2d 76 (1996)). Federal courts, pursuant to the Full Faith and Credit Act, 28 U.S.C. § 1738, must apply such law so long as it conforms to due process. Id. at 1260, 1267 (citing Richards, 517 U.S. at 797, 116 S.Ct. at 1765). Due process, the Court observed, "protects those rights `so rooted in the traditions and conscience of our people as to be ranked as fundamental.'" Id. at 1267 (quoting Snyder v. Massachusetts, 291 U.S. 97, 105, 54 S.Ct. 330, 332, 78 S.Ct. 674 (1934)). The question the District Court examined, then, was whether "a state's strict adherence to the boundaries of traditional preclusion law ... is a `fundamental federal right.'" Id. The Court thus focused its attention on a substantive due process question. In doing so, it neglected the procedural due process question with which defendants were primarily concerned: Are defendants denied their property without due process of law when a plaintiff is permitted "to use the [Phase I] findings to establish [for example] that the cigarettes she smoked were defective, in the face of a possibility that no jury ever found that fact"? Defendants' Rule 16(c) Motion at 4, Waggoner, 835 F.Supp.2d 1244 (M.D. Fla. 2011) (No. 3:09-CV-10367).

In tackling its substantive due process question, the Court found the defendants unable to identify any fundamental-right deprivations. The Court implicitly acknowledged that Florida courts had changed the state's preclusion law in a significant way for Engle-progeny cases.160 Specifically, it assumed, as we did in Brown II, that the Florida Supreme Court meant issue preclusion when it said "res judicata,"161 and it interpreted the Florida courts as eliminating issue preclusion's actually decided requirement. Eliminating the actually decided requirement did not constitute a due process deprivation because defendants did not have a fundamental right "to a strict application of traditional preclusion law." Waggoner, 835 F.Supp.2d at 1268-69.

Although the Court conceded that the defendants had fundamental rights to an opportunity to be heard and against arbitrary deprivations of property, it held that such rights had not been violated in Engle-progeny cases like Waggoner. Id. at 1272-77. In reaching this conclusion, it first observed that defendants had been afforded an opportunity to be heard at Phase I: "Defendants had every reason to litigate each potential theory of liability to the fullest extent possible." Id. at 1276. It acknowledged that Phase I did not afford defendants "their day in court on ... legal causation, comparative fault, and damages," id., but the Court assumed that such a day would come later because "the Phase I jury `did not determine whether the defendants were liable to anyone.'"162 Id. at 1272 (quoting Engle III, 945 So.2d at 1263).

After Phase I, "defendants continue to vigorously litigate each and every remaining issue in each and every progeny suit"; thus, "the preclusive application of the Phase I approved findings in no way [arbitrarily] deprives them of property." Id. What were the "remaining issues" to which the Court referred? First, plaintiff's "addict[ion] to one of Defendants' cigarettes containing nicotine"; second, "that such addiction was the legal cause of [plaintiff's harm]"; third, "that [plaintiff's harm] manifested before the class membership cut-off date"; and fourth, "that no other procedural bar prevents any aspect of her claim." Id. at 1273 (emphasis added). Whether defendants' tortious conduct caused plaintiffs' harm did not make the District Court's list of remaining issues on which defendants deserved an opportunity to be heard.163

Returning to its core substantive due process reasoning, the Waggoner Court again emphasized that defendants did not have a fundamental right to traditional preclusion law. Florida courts needed "flexibility to accommodate the due process interests of both the Defendants and the thousands of Engle progeny plaintiffs." Id. at 1277. The Court did not expound on the limits of such flexibility and did not mention any concerns regarding Florida courts' changing the state's preclusion doctrines with respect to only one group of unpopular defendants after the parties had already litigated part of their lawsuit.

* * *

The Waggoner Court failed to directly engage with the defendants' due process arguments. It reframed as substantive their procedural due process concerns, and refused to endorse the proposition that due process prevents state courts from changing their preclusion doctrines. Naturally, courts sometimes deviate from or change established procedures and "not all [such] deviations ... result in constitutional infirmity." Honda Motor Co. v. Oberg, 512 U.S. 415, 430, 114 S.Ct. 2331, 2339, 129 L.Ed.2d 336 (1994). But query whether courts deny due process when they abrogate common-law practice so as to descend significantly below the level of protection afforded at common law. Id. at 430-33, 114 S.Ct. at 2331, 2339-42. Query also whether state courts can change preclusion law with respect to a few unpopular defendants in a lawsuit that has already been partially litigated. Waggoner does not say.

The Waggoner Court's silence on these matters may, in part, be explained by its thorough misunderstanding of Florida preclusion law. The Court assumed it was dealing with issue preclusion. It was not.164 It interpreted Martin II as requiring Engle-progeny plaintiffs to prove legal causation separately from class membership. That interpretation was incorrect.165 Finally, the Court assumed the Florida Supreme Court meant what it said when it noted in Engle III that "the Phase I jury `did not determine whether the defendants were liable to anyone.'" Id. at 1272 (quoting Engle III, 945 So.2d at 1263). The Florida Supreme Court later disavowed that statement in Douglas III. See Douglas III, 110 So.3d at 433-34 ("Respectfully, the Engle judgment was a final judgment on the merits.").

The upshot of all this confusion is that the Waggoner Court never evaluated whether defendants were deprived of property without due process of law when they were held liable despite plaintiffs never proving, and defendants never having an opportunity to contest, that defendants' tortious conduct caused plaintiffs' harm.166 The answer to that question, in my estimation, is simple.

E. The Second District Court of Appeal in Douglas II Accepted Martin II's Reasoning, But Certified the Due Process Question to the Florida Supreme Court

As noted earlier, Waggoner was the lead case among the cases awaiting trial in the Middle District of Florida. With the Waggoner preclusion decision in hand, the judges presiding over the remaining cases were free to proceed. Moving forward, progeny courts would accord the Phase I findings preclusive effect in accordance with the Martin II formulation of res judicata. The threshold question for progeny juries would be whether the plaintiff proved addiction to a defendant's cigarettes; if so, the trial would focus, not on the causal connection between the defendant's tortious conduct and the plaintiff's injury, but on damages arising from the plaintiff's smoking-related disease and the comparative negligence of the plaintiff and any other Engle defendant involved.

Earl Graham's case against RJR and Philip Morris was one of the lead progeny cases handled under the Waggoner umbrella. Whether it would be tried in accordance with preclusion law as set forth in Martin II, though, would depend on the Florida Supreme Court's ruling on a question the Second District Court of Appeal had certified in Douglas II, 83 So.3d at 1011, regarding the due process implications of the way in which progeny courts had been applying the Phase I findings.

The case had been brought in the Circuit Court for Hillsborough County, Florida, by James Douglas as the representative of the estate of his late wife, Charlotte Douglas. Its complaint presented the six Engle III-approved claims and, as in Brown I, Martin I, Jimmie Lee Brown I, and Waggoner, cited the Engle III opinion and the Common Liability Findings as conclusive proof of the elements of the claims.167 Third Amended Complaint at 7-8, Douglas v. Philip Morris USA Inc. (Douglas I), No. 08-CA-008108 (Fla. Cir. Ct. Mar. 12, 2010). The defendants answered the complaint, denying liability and raising the same due process issues they had raised in those cases. Answer to Amended Complaint, Douglas I, No. 08-CA-008108.

Douglas I was tried on the plaintiffs' claims of strict liability and fraudulent concealment.168 The plaintiff produced evidence of his wife's addiction to the defendants' cigarettes169 and that smoking caused her death. After the defendants rested, the Court instructed the jury, first, that the plaintiff had to prove membership in the Engle class; that is, that the decedent was addicted to and injured by cigarettes containing nicotine. Douglas II, 83 So.3d at 1004. Next, as in Martin II and Jimmie Lee Brown II, the Court informed the jury that it had to "accept the eight Phase I Engle findings as proven fact." Id. at 1005. Accordingly, if it found that the decedent's death was caused by cigarettes (rather than the defendants' tortious conduct), the jury had only to determine the percentage of the cigarettes she smoked that were of the respective defendants' brands. Id. at 1003.

Id. The Court entered judgment in accordance with the jury's verdict, and the tobacco companies appealed, claiming once again that allowing a class plaintiff to recover against an Engle defendant solely on proof of addiction to the defendant's cigarettes and resulting injury constituted a denial of due process. Id. at 1010.

In addressing the due process issue, the Second District reviewed Brown II, Martin II, and Jimmie Lee Brown II. The Court noted that we had concluded in Brown II that Engle III's "res judicata" statement meant issue preclusion. Id. at 1006 (citing Brown II, 611 F.3d at 1333). However, the Court also noted that we had "pointed out that the parties disagree as to what issue preclusion meant"170 and quoted the following passage from Brown II to substantiate:

Id. at 1006-07 (quoting Brown II, 611 F.3d at 1333). The Court then said that, we

Id. at 1007.

The Second District then compared the First District's decision in Martin II, which interpreted res judicata as claim preclusion, with the Fourth District's decision in Jimmie Lee Brown II, which interpreted res judicata as issue preclusion. Id. at 1007-11. The Court found a substantial difference in the two Courts' analyses.

The First District, citing the statements in the Engle trial judge's Omnibus Order "as conclusive on each of the elements of the [plaintiff's] causes of action," id. at 1008, concluded that the Circuit Court "properly relied on the Phase I findings and that there was no need for the plaintiff class members to `independently prove up those elements or demonstrate the relevance of the findings to their lawsuits.'" Id. at 1008 (quoting Martin II, 53 So.3d at 1069).

The Fourth District, on the other hand, believed that the Phase I findings "preclusively establish the conduct elements of the strict liability and negligence claims," but not the causation element of those claims. Id. at 1009 (quoting Jimmie Lee Brown II, 70 So.3d at 715) (emphasis added). "Legal causation," the Fourth District concluded, "and damages, must be proven in the second phase of trial. Additionally, the Fourth District gave lip service to the idea that legal causation for negligence and strict liability should be distinguishable from the causation that proves addiction resulting from class membership." Id. (citing Jimmie Lee Brown II, 70 So.3d at 715). Specifically, the Jimmie Lee Brown II Court reasoned, rather than merely proving a causal connection between cigarettes and injury, progeny plaintiffs should be required to show a causal connection between a defendant's tortious conduct and the plaintiff's injury: "post-Engle III plaintiffs must show `(i) [that] the defendant's failure to exercise reasonable care was a legal cause of decedent's death[ ] and (ii) [that] the defective and unreasonably dangerous cigarettes were a legal cause of decedent's death.'" Id. (quoting Jimmie Lee Brown II, 70 So.3d at 715).

After concluding its review of Martin II and Jimmie Lee Brown II, the Douglas II Court opted for Martin II's reasoning, finding no violation of due process.171 The Court concluded, however, that the due process issues the Engle defendants had been raising were of such significance that the Florida Supreme Court should address them. The Court therefore certified the following question to the Florida Supreme Court:

Id. at 1011.

IV.

THE FLORIDA SUPREME COURT IN DOUGLAS III HELD THAT THE ENGLE III COURT HAD (1) IMPLICITLY DETERMINED THAT THE PHASE I FINDINGS WERE FULL-BLOWN LIABILITY DETERMINATIONS AND (2) IMPLICITLY ENTERED JUDGMENT AGAINST ALL DEFENDANTS ON BEHALF OF ALL CLASS PLAINTIFFS

The Florida Supreme Court accepted the certification and entertained the certified question wearing two hats. First, it wore the hat of the rendering Engle III Court, attempting to recall what it had in mind when it decided Engle III seven years before. By invoking "res judicata," did it intend to invoke claim preclusion? In other words, did it implicitly enter judgment pursuant to the Phase I findings, which it interpreted as establishing the Engle defendants' liability to all class members? Or did it intend to invoke issue preclusion, interpreting the Phase I findings as factual findings upon which future courts would enter judgment?

Second, it wore the hat of a recognizing court, applying state preclusion law and evaluating whether doing so denied the defendants of their property without due process of law. I consider each of these two hats in turn.

* * *

The Second District, in its opinion certifying the constitutional question, explained the preclusive effect the Phase I findings would have if they represented factual determinations of whether the defendants engaged in tortious conduct, on the one hand, or full-blown liability determinations, on the other. It pointed to our opinion in Brown II and the Fourth District's in Jimmie Lee Brown II as examples of courts interpreting the Phase I findings as factual findings and the First District's opinion in Martin II as an example of a court portraying the Phase I findings as liability determinations as to each of the Engle III-approved tort claims. Douglas II, 83 So.3d at 1006-11.

Under Brown II's reasoning, the Phase I findings were factual determinations that foreclosed litigation in progeny lawsuits over whether each of the defendants engaged in six kinds of tortious conduct.172 The findings did not represent a jury determination that the defendants were liable to all class plaintiffs for all six torts. Thus, litigation of other essential facts — causation (which requires identifying tortious conduct)173 and damages — would not be foreclosed or limited. Indeed, consistent with issue preclusion's actually decided requirement,174 the findings would be "given effect to the full extent of, but no farther than, what the jury found." Brown II, 611 F.3d at 1333. This approach was consistent with what the Phase I findings said on their face and Engle III's holding that "individualized issues such as legal causation [and] comparative fault" would be litigated in progeny trials. Engle III, 945 So.2d at 1268. The Fourth District, in Jimmie Lee Brown II, agreed with this limitation (even if it was "constrained" not to implement it): the findings clearly were not, and should not be treated as, liability determinations; rather, they should operate to foreclose litigation only as to facts the Phase I jury actually found. 70 So.3d at 714-15. Thus, every progeny plaintiff should be required to prove, and defendants should be able to contest, that the conduct deemed tortious in Phase I caused the plaintiff's harm.175 Id. at 715.

In Martin II, the First District purported to "generally agree[ ]" with the Brown II approach, but concluded that determining what the Phase I jury actually decided, as issue preclusion requires, was entirely unworkable and would "undercut" the Engle III plan.176 The First District interpreted the Engle III plan as one designed to enhance class members' chances of prevailing against Engle defendants in progeny cases; issue preclusion was inconsistent with that plan because progeny courts would be unable to determine exactly what conduct the Phase I jury identified as tortious.

What would be consistent, though, is if the Phase I findings could somehow be portrayed as liability determinations rather than mere conduct findings. The Martin II Court accomplished such a portrayal simply by proclaiming it, insisting that the Phase I findings established the elements of the progeny plaintiffs' tort claims such that they "need not independently prove up those elements or demonstrate the relevance of the findings to their lawsuits." Martin II, 53 So.3d at 1069. "No matter the wording of the findings on the Phase I verdict form." Id. at 1067. And never mind that the Phase I jury was instructed not to consider liability,177 and that Engle III makes clear that "Phase I yielded no determination as to the defendants' liability to any individual class member." Id. at 1064 (citing Engle III, 945 So.2d at 1263). Never mind, also, that the Phase I Circuit Court rejected the plaintiffs' proposed verdict-form questions about whether the defendants' wrongful conduct was a "legal cause of damage, injury or death" so as to preserve litigation of legal/proximate cause for later phases of the Engle litigation.178 On the basis of its counterfactual portrayal of the Phase I findings, the Martin II Court concluded that progeny plaintiffs needed only to prove class membership — harm caused by smoking a defendant's cigarettes; no need to identify particular tortious conduct or prove that such conduct caused harm.

The Florida Supreme Court agreed. It endorsed Martin II's reasoning that interpreting "res judicata" as issue preclusion — and thus interpreting the Phase I findings merely as establishing facts that might assist the class members in proving their tort claims — was unacceptable because it would render those findings "useless":

Douglas III, 110 So.3d at 433 (emphasis added). Portraying the findings as liability determinations was essential because proving a causal connection between tortious conduct and injury in progeny cases would necessarily require identifying the defendants' tortious conduct; simply proving that a defendant's conduct, which may or may not have been tortious, caused harm would be insufficient.179 Despite a yearlong trial, the Phase I findings provided no information about the particular conduct the jury had deemed tortious. Thus, merely allowing those findings to stand would leave plaintiffs in about the same position in which they would have been had Phase I never taken place — "defendants [would be] permitted to relitigate matters pertaining to their conduct." Id. at 429.

Whereas issue preclusion left class members empty handed, claim preclusion assisted them because it necessarily reinterpreted the Phase I findings as "conclusive not only as to every matter which was offered and received to sustain or defeat the claim, but as to every other matter which might with propriety have been litigated and determined in that action."180 Id. at 432 (quoting Kimbrell v. Paige, 448 So.2d 1009, 1012 (Fla. 1984)). Assisting the plaintiffs, the Court concluded, is what it had in mind when it wrote its opinion in Engle III.

Justice Canady, in dissent, rejected the Douglas III Court's reinterpretation of the Phase I findings as establishing the defendants' liability to hundreds of thousands of class plaintiffs who were unknown and not present at Phase I. He pointed to Question 3181 from the Phase I verdict form as an example of a finding that was "a much too slender reed to support the imposition of liability on the defendants." Id. at 436 (Canady, J. dissenting)

Id. at 436-37. Justice Canady felt that all the findings were too general to identify the defendants' tortious conduct, let alone establish that such conduct caused every class members' harm.

His counterparts in the majority disagreed: "[T]he Phase I jury already determined that the defendants' conduct subjects them to liability to Engle class members under [strict liability and negligence] theor[ies]." Id. at 430. To bolster its portrayal of the Phase I findings, the Douglas III majority, like the First District in Martin II, focused not on the Phase I findings themselves, but on the evidence that was before the Phase I jury. Its inquiry centered on whether that evidence was sufficient to withstand a motion for directed verdict.182 Because, Douglas III reasons, the Phase I record contained "more than sufficient evidence" upon which a properly instructed jury could have found the defendants liable to all plaintiffs, "[t]hat the ... jury did not make detailed findings ... [was] immaterial." Id. at 423, 433 (emphasis added).

By adopting such a portrayal, the Court disavowed its earlier statement in Engle III in which it made clear that "the Phase I jury `did not determine whether the defendants were liable to anyone.'" Engle III, 945 So.2d at 1263. It also disregarded its Engle III holding that "individualized issues such as legal causation" would be litigated in progeny trials. Id. at 1268. Such issues were now "immaterial." Douglas III, 110 So.3d at 433. Although the majority agreed with Justice Canady that the findings are "useless" for the purposes of identifying the conduct the Phase I jury deemed tortious and proving that such conduct caused harm183 — the majority conceded that Phase I jury "did not make detailed findings about what evidence it relied upon to make the Phase I common liability findings" — such uselessness did not matter because the jury had determined liability instead. Id.

With defendants' liability to all class members established, all that remained for progeny plaintiffs to prove was (1) their class membership (by proving addiction to a defendant's cigarettes and a smoking-related injury184) and (2) their damages. Engle defendants, therefore, could contend that a plaintiff was not a class member because her injury was not caused by smoking,185 but they were precluded by "res judicata" from contending that a plaintiff's smoking-related injury was not caused by tortious conduct — when a plaintiff "prov[es] that addiction to the Engle defendants' cigarettes containing nicotine was a legal cause of the injuries alleged," "injury as a result of the Engle defendants' conduct is assumed." Id. at 429 (emphasis added).

In sum, the Florida Supreme Court, wearing its rendering-court hat in Douglas III, insisted that it knew when it wrote Engle III that the Phase I findings were "useless" under an issue-preclusion regime, which is why it invoked claim preclusion instead.186 See 110 So.3d at 433 ("[W]e used the correct term when we gave the Phase I findings `res judicata effect,' signifying that relitigation of the class's cause of action established by the Phase I findings would be barred."). The Florida Supreme Court thus knew all along that progeny plaintiffs would be unable to decipher what the Phase I jury had decided — "the Engle jury did not make detailed findings for which evidence it relied upon to make the Phase I ... findings."187 Id. Though the Court always knew that the Phase I findings were useless for issue-preclusion purposes, it nevertheless intended those findings to be binding for claim-preclusion purposes. The same findings that admittedly could not be relied upon to identify something the Phase I jury had been asked to determine — tortious conduct — could be relied upon to establish something that jury was asked not to determine — causation.188

In addition to taking issue with the Court's preposterous portrayal of the Phase I findings, Justice Canady also argued that the Court could not apply claim preclusion "[b]ecause the judgment that emerged from Engle was not a final judgment on the merits." Id. at 436 (Canady, J. dissenting).

Id. at 438-39 (emphasis added except for "not").

Again, the Court disagreed: "[T]he Engle judgment was a final judgment on the merits." Id. at 433. By invoking claim preclusion, the Douglas III Court reasoned, the Engle III Court implicitly entered judgment pursuant to the Phase I "common liability findings" and "necessarily decided that the approved Phase I findings are specific enough." Id. at 429 (citing Engle III, 945 So.2d at 1255). Because the defendants' liability had been properly adjudicated, the Court reasoned, it was not unusual for "the jury's findings in the first trial [to be] binding in the second even if the first trial does not result in a money judgment." Id. at 434 (citing 3 A. Conte & H. Newberg, Newberg on Class Actions § 9:53 (4th ed. 2012), which points out that "[n]ot infrequently, actions filed as class actions present predominating common issues of liability, while proof of damages may remain as individual issues for the several class members").

* * *

After interpreting its opinion in Engle III, the Florida Supreme Court moved on to its recognizing-court tasks of applying state preclusion law and evaluating whether doing so would deprive the defendants of their property without due process of law. Progeny courts had read Engle III as predetermining the res judicata effect of the Phase I findings, thereby foreclosing them from carrying out their constitutional duty to examine the Engle procedures to determine whether those procedures had denied the tobacco companies due process and whether the party asserting preclusion had established its elements. See, e.g., Martin II, 53 So.3d at 1067 (refusing to perform recognizing-court tasks because Engle III had predetermined the res judicata question, and that was good enough for the First District, "[n]o matter the wording of the findings on the Phase I verdict form."). Prodded by the defendants' briefs, e.g., Reply Brief for Petitioners at 14-15, Douglas III, 110 So.3d 419 (No. SC12-617), the Douglas III Court reintroduced the recognizing court's duties under Florida preclusion law and the U.S. Constitution in progeny cases, observing that

Douglas III, 110 So.3d at 431 (quoting Hansberry, 311 U.S. at 40, 61 S.Ct. at 117). In conducting such an examination, Douglas III explains, a recognizing court must ascertain whether the litigant was denied "the basic common law protection against an arbitrary deprivation of property ... due process [requires]." Id. at 431 (citing Honda Motor Co., Ltd. v. Oberg, 512 U.S. 415, 432, 114 S.Ct. 2331, 2340-41, 129 L.Ed.2d 336 (1994)).

The Florida Supreme Court's first step in its examination was to evaluate whether defendants were afforded adequate notice. It found that the original trial plan provided some notice: "The class action trial plan put the Engle defendants on notice that if the Phase I jury found against them, the conduct elements of the class's claims would be established, leaving only plaintiff-specific issues for individual trials." Id. at 429 (emphasis added). The Court was right, of course, that the defendants had notice that conduct elements of the plaintiffs' claims would be decided in Phase I.189 The Court did not, however, identify any point at which the defendants were afforded notice that causation and liability would be decided in Phase I or, as it were, in the Engle III review of the issues decided in Engle II. The Court also did not comment on whether the defendants were afforded notice that it was considering overruling its holding in Engle III that "individualized issues such as legal causation" would be litigated, rather than presumed, in progeny trials. Engle III, 945 So.2d at 1268.

Next, the Florida Supreme Court examined whether the defendants were given an opportunity to be heard. The Court answered that question with a resounding "yes" — with respect to conduct:

Douglas III, 110 So.3d at 431.

The Court also found that the defendants had been afforded an opportunity to be heard, in a generic sense, in the Douglas I trial and on appeal below:

Id. at 432 (emphasis added). Though the Court was satisfied that the defendants had an opportunity to be heard on whether they had committed tortious acts, it was silent as to whether the defendants were ever, at any stage, afforded an opportunity to be heard on the causal connection between their tortious acts and class members' injuries. The Court was also silent as to whether the defendants were afforded an opportunity to be heard on the matters it said were decided in Engle III, or on whether it should overrule its holding from Engle III that "individualized issues such as legal causation" would be litigated in progeny trials. Engle III, 945 So.2d at 1268.

The defendants felt that the Florida Supreme Court, in considering whether they had been given notice and opportunity to be heard with respect to conduct in Phase I, had entirely missed the point of their concerns. They were concerned about notice and opportunity to be heard with respect to causation, not conduct. They contended that progeny plaintiffs were being "improperly excuse[d] ... from having to prove that the Engle defendants' conduct was a legal cause of their injuries." Douglas III, 110 So.3d at 430. They were also concerned about the lack of notice and opportunity to be heard afforded to them in Engle III. In that case, they argued, the Florida Supreme Court acted on its own initiative, without notifying the parties or allowing them an opportunity to be heard.191 For that reason, after the Engle III opinion issued, the defendants moved the Court for rehearing. The Court denied that motion in a one-line order, which it chose not to publish.

To its credit, the Court did engage, at least in part, with one of the defendants' due process concerns. It examined the procedures afforded to the defendants to determine whether they had been denied any "basic procedural protections of the common law." Oberg, 512 U.S. at 430, 114 S.Ct at 2340. Specifically, it examined the defendants' claim that the Constitution requires that issues be actually decided before they may be given preclusive effect. As mentioned above, the Court responded to that argument by insisting that recognizing progeny courts apply claim preclusion to preclude causes of action not issues. "[C]laim preclusion, unlike issue preclusion," the Court reasoned, "has no `actually decided' requirement.'" Douglas III, 110 So.3d at 435. Although the Court addressed the absence of the actually decided protection, it failed to address whether the defendants had been denied basic common-law protections when the rendering Engle III Court dictated the preclusive effect of the Phase I findings to recognizing progeny courts; when class plaintiffs were allowed to assert claim preclusion — an affirmative defense under Florida law192 — offensively; or when class plaintiffs had been spared their common-law burden of proof on the elements of both preclusion and causation.

* * *

In wearing both rendering- and recognizing-court hats in Douglas III, the Florida Supreme Court found itself in a conflict-of-interest position, which called into question its ability to be impartial. By pronouncing the "res judicata effect" of the Phase I findings in Engle III, the Court had signaled that those findings would be useful to class plaintiffs such that they would not have to relitigate the defendants' conduct. Under the reasoning of Brown II and Jimmie Lee Brown II, it had not turned out that way — in order to prove that the defendants' tortious conduct caused a plaintiff's injury, the plaintiff would have to identify the tortious conduct. The useless Phase I findings provided no way to do that, so litigation over conduct would begin anew. The Florida Supreme Court was thus faced with a choice: stick with what Engle III said, engage head on with the defendants' due process concerns, and admit mistake; or insist that Engle III meant something different from what it said, dodge the defendants' due process concerns, and pretend the mistake never happened. Unfortunately, the Court chose the latter option.

This conflict, in turn, gave rise to another. In defending its portrayal of the Phase I findings, the Douglas III Court proffered evidence, on behalf of the plaintiff, from the Phase I trial record. It then evaluated that evidence using a lower standard of review than the law requires193 to support the conclusion it had presupposed. But juries make findings by answering questions, not by looking at evidence, so the very premise of the Douglas III Court's sufficiency-of-the-evidence evaluation was flawed.

The Douglas III Court's conflicts caused it to disregard Engle III while purporting to interpret it. Engle III says that the Phase I findings are entitled to res judicata effect. Those findings, that Court made clear, "did not determine whether the defendants were liable to anyone." 945 So.2d at 1262-63 (quoting Engle II, 853 So.2d at 450). That liability, according to the Engle trial plan, would be established in the Phase III trials, where "the remaining issues, including individual causation and apportionment of fault among the defendants," would be litigated. Id. at 1254. That the Phase I Circuit Court rejected proposed verdict-form questions about proximate causation corroborates the obvious proposition that it was both inappropriate and impossible for the Phase I jury to determine whether the defendants were liable to hundreds of thousands of absent class members.194

Consistent with the reality of the Phase I trial and the necessary implications of Engle III's no-liability statement, Brown II and Jimmie Lee Brown II interpreted Engle III's res judicata statement as an instruction to recognizing courts to give issue preclusive effect to the Phase I factual findings. As Brown II had discovered, and as Douglas III acknowledged, however, those factual findings were "useless." Douglas III, 110 So.3d at 433. Had the Florida Supreme Court in Engle III made an embarrassing mistake by signaling otherwise?

Martin II's reasoning seemed to provide an escape. The Engle III Court had clearly intended to throw the plaintiffs a bone, so a bone they would get. If progeny courts could preclude the defendants from defending their entire cause of action, Martin II reasoned, the uselessness of the findings would be rendered irrelevant. Claim preclusion and reinterpreting the Phase I findings were the key.

So Douglas III — faced with the prospect of an embarrassing mea culpa — adopted Martin II's reasoning along with its conclusive presumptions.195 But, as Justice Canady observed, that reasoning is only half baked — claim preclusion requires a final judgment, and Engle III makes it clear that plaintiffs do not have one. In fact, Engle III holds that "individualized issues such as legal causation" had yet to be litigated. 945 So.2d at 1268. Douglas III simply overrules that holding through the sheer force of its own authority as the state's highest court: "[t]he Engle judgment" is "a final judgment on the merits" because "the class jury resolved ... the Engle defendants' common liability to the class under several legal theories."196 Id. at 433-34.

But even this stunning reversal did not get the Florida Supreme Court out of Dodge. The defendants had some compelling due process concerns that were made even more compelling by the Court's reversal and reliance on claim preclusion. Why, for example, were they not entitled to an opportunity to be heard as to whether their tortious conduct caused the plaintiffs' injuries? Why were they never given notice before the Phase I trial or before the Engle III decision was handed down that the Phase I findings would hold them liable to all class plaintiffs? Why was it okay to relieve plaintiffs of their common-law burden of proof and allow them to use an affirmative defense offensively?

Luckily, for the Florida Supreme Court, it was in a position to avoid such uncomfortable questions. When the defendants said "no notice and opportunity to be heard on causation," the Court simply redirected: "notice and opportunity to be heard on conduct." When the defendants said "issues must be actually decided," the Court said, "What issues? All we can see are causes of action litigated to completion."

Thus, to the Douglas III Court, Engle III was nothing but a code. Where Engle III says "no final judgment," it means "final judgment." Where Engle III says "res judicata to factual findings," it means "res judicata to causes of action litigated to completion." Where Engle III holds that "legal causation would be litigated in progeny trials," it means "legal causation would be presumed in progeny trials." The Martin II Court understood the code and ran with it. The Jimmie Lee Brown II Court begrudgingly accepted it. We, in Brown II, naively believed that Engle III meant what it said.

Why did the Florida Supreme Court resort to a code in Engle III as opposed to simply saying what it meant? Was it attempting to punish unpopular defendants and benefit sympathetic plaintiffs while concealing the constitutional shortcuts it took to do so? Was it attempting to legislate a ban on cigarettes while cloaking the resulting preemption problems?

Consider, for a moment, what may have been running through the minds of the justices that comprised the Engle III majority as they contemplated how to draft their majority opinion. Douglas III tells us that the Engle III majority197 knew that the Phase I findings were "useless." See id. at 433 (insisting that the Engle III majority deliberately opted for claim preclusion over issue preclusion because it knew the Phase I findings would be "useless in individual actions"). Despite the reams of evidence that plaintiffs had presented against the defendants, the jury had not been properly instructed, the Phase I verdict form had not been properly formulated, and the yearlong Phase I trial had accordingly been a waste. But why, the majority may have thought, should plaintiffs suffer for the trial judge's incompetence?198 Defendants should be held accountable, and we should do what we can to avoid such an embarrassing failure in a high-profile Florida state-court case.

But animating litigation that had so badly faltered was not an easy task. The majority did not even try, as some progeny courts would, to breathe life into the useless Phase I findings by creatively interpreting them in light of the Omnibus Order199 or the Phase I trial record.200

Further, the Phase I jury, in accordance with the trial plan, was instructed not to evaluate the defendants' liability to all class members.201 To nevertheless hold that the jury's findings established the defendants' liability to all class members would too obviously run roughshod over the defendants' jury-trial rights.202 It would also too obviously implicate due process concerns: The parties had filed their briefs and orally argued their positions on the issues the Supreme Court accepted for review in Engle III. Those issues did not include adjudicating liability at that stage of the case, and parties did not argue, nor did they have an opportunity to argue, whether adjudicating liability to all class members was appropriate. Requesting briefs on the question would only shed light on obvious constitutional barriers to the majority's desired outcome.

So whether they did so intentionally or not, the Engle III majority achieved surreptitiously what they could not have achieved openly. Their res judicata instruction, which they issued without input from the parties,203 expertly toed the line between subtlety and clarity. To start, "res judicata" is an ambiguous term that embraces both issue and claim preclusion.204 Using that term thus obscured that the Court had invoked claim preclusion and had thereby implicitly adjudicated defendants' liability to all class members.205 The Court further obfuscated this idea by observing that "the Phase I jury `did not determine whether defendants were liable to anyone.'"206 Engle III, 945 So.2d at 1263 (citing Engle II, 853 So.2d at 450). Moreover, its instruction was dicta,207 and this allowed the class representatives to successfully argue in their brief in opposition to the Engle defendants' petition for a writ of certiorari that the defendants' "due process concerns are premature and not ripe for review." Brief in Opposition, R.J. Reynolds Tobacco Co. v. Engle, 552 U.S. 941, 128 S.Ct. 96, 169 L.Ed.2d 244 (2007) (No. 06-1545), 2007 WL 2363238, at *1. As the class representatives argued, "[N]o court has yet [applied] the findings" and recognizing courts ordinarily decide for themselves whether to give res judicata effect to a rendering court's findings.208 Id.

Amidst all of its ambiguity and doublespeak, Engle III was still clear enough to carry a binding message to progeny courts: adjudicate class members' claims without requiring them to relitigate the issues litigated in Phase I. In other words, hold defendants liable to all class members.209 The Fourth District felt "constrained" by Engle III's message, Jimmie Lee Brown II, 70 So.3d at 715, and the First District embraced it, Martin II, 53 So.3d at 1066-67. Even our court deferred to it. Walker II, 734 F.3d at 1289.

Did the Engle III majority really intentionally mastermind such a deviously clever opinion? In doing so, did they deliberately forego asking the parties for briefs because they did not want to shed light on the common-law and constitutional obstacles that stood in the way of their objective? If they did — and Douglas III invites that conclusion — the Engle III Court, in a troublingly calculated way, deprived the defendants of their property without due process of law and entered a final judgment in a controversy that did not exist, one the Court itself contrived. Had the Court acknowledged this in Engle III, is there any doubt the U.S. Supreme Court would have granted certiorari review? If the Engle III majority had the Douglas III result in mind, they could not let it see the light of day. As Justice Brandeis remarked, "Sunlight is said to be the best of disinfectants; electric light the most efficient policeman."210

Whatever result Engle III had in mind, Douglas III ushered in a new Florida law211 under which tobacco manufacturer's liability would now be established differently from that of all other tort defendants.212 The "highly individualized" differences between smokers, which were not amenable to "class action treatment," Engle III, 945 So.2d at 1263, and were to be dealt with in the class members' individual Phase III actions, would no longer matter. Nor would it matter what brands the smoker used. It would not matter which of the defendants' advertisements she had seen or not seen. In fact, it would not matter if she had never seen a manufacturer's advertisement. It would not matter if she grew up in a house of smokers or smoked because her friends always smoked. Most simply, it would not matter why the smoker chose to smoke. Likewise, it would not matter if the defendants breached a duty of care to the smoker. And it would not matter whether the defendants' (unidentified) breach of duty caused the smoker any harm.

Now, under Florida tort law as set out in Douglas III, a plaintiff can recover damages from an Engle defendant by merely by proving her status as a class member by establishing that she contracted a smoking-related illness.213 She does not have to prove that the cigarettes she smoked were defective or negligently sold, nor that the defect or the act of negligence caused her harm. That cause is assumed conclusively. Douglas III, 110 So.3d at 429 (When a plaintiff "prov[es] that addiction to the Engle defendants' cigarettes containing nicotine was a legal cause of the injuries alleged," any "injury as a result of the Engle defendants' conduct is assumed." (emphasis added)). Thus, Douglas III's conclusive presumption treats all class members as one, relieving all of the burden of proving that the defendant's tortious conduct caused their injury.

In sum, Douglas III did three things. First, it established an unconstitutional214 conclusive presumption215 that all of the cigarettes smoked by the class members were defective and unreasonably dangerous and negligently sold.216 The Court did this not on the basis of the useless Phase I findings, but on the basis of the evidence adduced during the Phase I trial (as depicted in the Engle trial judge's Omnibus Order).217 Indulging this presumption had the same effect as deciding that all cigarettes were defective and unreasonably dangerous and negligently sold as a matter of law. However, since the evidence on those issues was at variance, the Court deprived the defendants of their constitutional right to a jury trial. See supra note 142 and accompanying text.

Second, Douglas III utilized this conclusive presumption as the foundation for a second unconstitutional218 conclusive presumption — that the Engle defendants' tortious conduct (presumed present in all cases) caused the class members' injuries — in order to enable the class members to establish the causation element of their tort claims.219 Without the presumption, only the Phase I findings — useless in resolving the causation issue — supported the claims. Why? Because the Phase I jury was instructed that its role was not to decide whether the defendant's tortious conduct, if any, caused a class member's injury. Rather, its role was limited to answering the special interrogatories, all of which dealt with the conduct alleged in the class action complaint. A progeny jury would determine whether the defendant's tortious conduct caused a class member's injury. And as the defendants noted over a decade ago, it is impossible to properly determine causation because the findings do not identify the defect or tortious conduct that the jury found. In approving Douglas II's affirmance of the Circuit Court's use of a conclusive presumption to establish the cause of Ms. Douglas' death, the Florida Supreme Court denied the defendant's right to submit the causation issue to a jury.

Third, in creating these conclusive presumptions, Douglas III materially altered the laws of products liability and negligence as they relate to the manufacture and sale of cigarettes, ultimately making the sale of cigarettes unlawful. Douglas III also overruled Engle III's holding that "individualized issues such as legal causation [and] comparative fault" would be litigated in progeny trials. Engle III, 945 So.2d at 1268.

The effect of Douglas III's holdings is that tobacco companies now have a tort-law duty not to sell cigarettes in the state of Florida. As sanctions for breaching this duty, an Engle defendant must pay damages — and possibly punitive damages — to any class member who can satisfy a jury that she is addicted to its cigarettes220 and suffered a smoking-related disease.221

V.

THE WALKER PANEL EFFECTIVELY REWROTE AND THEN GAVE FULL FAITH AND CREDIT TO DOUGLAS III BEFORE ISSUING A NEW OPINION THAT GAVE FULL FAITH AND CREDIT TO ENGLE III, YET LEFT THE ORIGINAL OPINION'S INAPPOSITE REASONING INTACT

Like the progeny plaintiffs in Waggoner, the plaintiffs in Walker I222 — suing on behalf of deceased relatives — amended their complaints in the immediate aftermath of Martin II to capitalize on its claim-preclusion holding. Mr. Walker's second amended complaint characterized Engle III's res judicata dicta223 as a "mandate" that "provided Plaintiff the opportunity to complete unresolved damages claims."224 Second Amended Complaint at 2, 3, Walker v. R.J. Reynolds Tobacco Co. (Walker I), No. 3:09-cv-10598-RBD-JBT (M.D. Fla. Mar. 3, 2011) (emphasis added). The plaintiffs assumed, in accordance with Martin II, that all that remained to be resolved in his case was damages; the defendants were precluded by the "mandate" in Engle III from litigating both (1) that the cigarettes smoked by the decedents were defective, unreasonably dangerous, and negligently produced and (2) that the defendants' tortious conduct caused the decedents' death. Id. Accordingly, they pleaded only class membership — addiction and a smoking-related injury. Id. at 4. To support their assertion of claim preclusion, they proffered only the Engle III opinion and the "Common Liability Findings" approved by the Engle III Court. Id. at 5-9.

In their answers, the defendants took issue, as they had in previous progeny cases, with the plaintiffs' meager pleadings. RJR insisted that the Phase I findings were not liability findings at all; rather, they "are so generalized and non-specific that they are inadequate to support an individualized determination of essential issues such as liability, legal causation, and damages in this or any other subsequent individual action." Answer, Defenses and Jury Demand of Defendant R.J. Reynolds Tobacco Co at 2, Walker I, No. 3:09-cv-10598-RBD-JBT. In particular, RJR argued that the Phase I finding related to product defect(s) was deficient because it provided no basis on which a

Id. at 3. Similarly, with respect to the Phase I negligent-conduct finding, RJR contended that

Id. at 5-6. Philip Morris and Lorillard echoed such sentiments, maintaining that proving a connection between tortious conduct and injury on the basis of the Phase I findings was "impossible." Philip Morris USA Inc.'s Answer to Plaintiff's Second Amended Complaint and Demand for Trial by Jury at 16, 18, Walker I, No. 3:09-cv-10598-RBD-JBT; Lorillard Tobacco Co.'s Answer to Plaintiff's Second Amended Complaint and Demand for Jury Trial at 12, 14, Walker I, No. 3:09-cv-10598-RBD-JBT.

As in previous cases, these objections triggered the District Courts' Hansberry obligation under Florida law and the U.S. Constitution to examine the proceedings to ensure that applying Florida claim preclusion would not violate the tobacco companies' due process rights.225 See supra Part II.B. Relying on Waggoner's pretrial ruling, the District Courts concluded that an additional examination was unnecessary.226 They gave preclusive effect to Engle III and accepted the plaintiffs' claim-preclusion proffer of Engle III and its approved findings as adequate to establish that the Engle defendants were liable to all class members, including Walker and Duke. Philip Morris and Lorillard later settled, but the plaintiffs' damages cases against RJR went to a jury. In accordance with Martin II and the plaintiffs' pleadings, juries in both cases were instructed to hold RJR liable if the plaintiffs proved their membership in the Engle class — "addiction to smoking the defendant's cigarettes and resulting injury." Martin II, 53 So.3d at 1069. The jury in Walker I "found in favor of Walker on the claims of strict liability and negligence" and the jury in Duke "found in favor of Duke only on the claim of strict liability." Walker II, 734 F.3d at 1286. RJR appealed the District Court judgments to our Court.

While these appeals were pending, the Florida Supreme Court decided Douglas III, which, as explained above, portrayed the Phase I findings as liability determinations and endorsed Martin II's retroactive substitution of claim preclusion for issue preclusion as the Engle litigation model.227 Before it addressed RJR's arguments, the Walker panel thus had to decide between two basic portrayals of the Phase I findings — Brown II's depiction of the findings as factual determinations regarding the defendants' conduct and Douglas III's depiction of the findings as full-blown liability determinations.228 Acknowledging clear language on the Phase I verdict form and in Engle III that contradicted Douglas III's portrayal, the panel adopted Brown II's basic depiction.229 See Walker v. R.J. Reynolds Tobacco Co. (Walker 0) at 18, Nos. 12-13500, 12-14731 (11th Cir. Sept. 6, 2013) vacated and superseded by Walker II, 734 F.3d 1278 ("During [Phase I], the jury considered only `common issues relating exclusively to the defendants' conduct... but did not decide whether the tobacco companies were liable to any of the class representatives or members of the class." (citing Engle III, 945 So.2d at 1256, 1263)).230

Upon rejecting Douglas III's portrayal of the Phase I findings as liability determinations, the Walker panel could no longer apply claim preclusion in accordance with Florida law. It could, however, reach Douglas III's outcome under the issue-preclusion framework set forth in Brown II, if it could portray the Phase I conduct findings as "specific enough to apply in favor of every class plaintiff." Id. at 21. In other words, if "the tobacco companies acted wrongfully toward all plaintiffs" because all cigarettes are defective, unreasonably dangerous, and negligently produced because they "contain nicotine are addictive and produce dependence,"231 then any error that might have occurred in the Walker I and Duke trials could arguably be characterized as harmless. Id. That the District Courts precluded RJR from contesting whether the cigarettes Walker and Duke smoked were defective, unreasonably dangerous, and negligently produced did not seem problematic because those issues would have been actually decided by a previous fact finder. Likewise, that the District Courts precluded RJR from contesting whether its tortious conduct caused Walker's and Duke's injuries seemed acceptable because Walker and Duke had to prove their status as class members by proving that (supposedly defective and negligently produced) cigarettes caused their injuries.

The Walker panel did not interpret the "ambiguous [Phase I] jury verdict[s]" as "specific enough" factual findings. Id. at 23. Rather than simply look at the questions the Phase I jury was instructed to answer, it relied on the Douglas III Court to interpret the findings. The Douglas III Court, the panel insisted, "looked past the ambiguous jury verdict[s] to decide [a] question of fact"; namely, the Douglas III Court decided that "the approved Phase I findings are specific enough." Id. at 23-24. Granted, the panel "disagree[d] with [Douglas III's] holding about what the jury in Phase I decided." Id. at 18. Nevertheless, the panel concluded that it "could not refuse to give full faith and credit" to the Douglas III Court's supposed "merely erroneous" factual determination. Id. (citation omitted).

In its petition for rehearing, RJR rejected the panel's portrayal of Douglas III.232 It maintained that the Court in Douglas III never searched the record to determine what the Phase I jury actually decided. Id. In fact, RJR argued, the Court conceded that performing such a search was futile: "[A]pplying issue preclusion — and its `actually decided' requirement — `would effectively make the Phase I findings... useless in individual actions.'" Id. at 12 (quoting Douglas III, 110 So.3d at 423 (Fla. 2013)). RJR elaborated, "Precisely because the court could not determine exactly what the Engle jury had decided, the court went on to develop — and defend at length — its novel version of claim preclusion." Id. Not only did Douglas III make clear that the "useless" Phase I findings lack evidentiary value, it distinguished RJR's "lead due-process precedent as a case about issue preclusion, and it emphasized the `specific importance to this case' of the fact `that claim preclusion, unlike issue preclusion, has no "actually decided" requirement.'" Id. at 13 (quoting Douglas III, 110 So.3d at 435). The Walker panel disregarded these concerns.

By portraying Douglas III as determining that the Phase I findings were "specific enough" factual findings, the panel also disregarded Douglas III's reliance on claim preclusion and the plaintiffs' corresponding claim-preclusion proffer, opting for issue preclusion instead. In light of RJR's due process objections, the Walker panel, as a recognizing court, had a duty under both Florida law and the U.S. Constitution to "examine the course of procedures in both" (1) the Walker I and Duke trials and (2) the case to which it gave full faith and credit, Douglas III. Douglas III, 110 So.3d at 430-31 (quoting Hansberry, 311 U.S. at 40, 61 S.Ct. at 117). The panel found no due process deprivation in Walker I and Duke because, as explained above, it believed any errors that occurred in those trials — including the misinterpretation of the Phase I findings and the application of claim preclusion instead of issue preclusion — were harmless. The panel also found that the Douglas III decision "did not arbitrarily deprive R.J. Reynolds of property without due process of law" because the Douglas III Court "looked through the jury verdict entered in Phase I to determine what issues the jury decided."233 Walker 0, at 19. According to the panel, the Douglas III Court "concluded that the jury was asked only to `determine all common liability issues for the class,' not brand specific defects,"234 and the panel held that the Douglas III Court was "entitled to look beyond the jury verdict[s] to determine what issues the jury decided." Id. (citations omitted). The panel did not explain whether the Douglas III Court afforded the parties before it with notice that it was going to "look beyond the jury verdict[s]" to interpret them. Id. Nor did the panel comment on whether the Douglas III Court gave the parties an opportunity to be heard on the appropriate interpretation of Phase I findings. Rather, in a cursory response to RJR's due process argument, the panel noted that "if due process requires a finding that an issue was actually decided, then the [Douglas III Court] made the necessary finding." Walker 0, at 24.

After its portrayal and inspection of Douglas III, the Walker panel noted its clever reconciliation of the disparate Brown II and Douglas III approaches. It observed that in Brown II, "we stated that, although the jury verdict in Phase I was ambiguous on its face, members of the Engle class should be allowed an opportunity to establish that the jury in Phase I actually decided particular issues in their favor." Id. at 20. According to the panel, the Douglas III Court took the plaintiffs' burden upon itself and decided erroneously, but not arbitrarily, when it concluded the jury findings were specific enough. Id.

For all its cleverness, though, the panel lost track of fundamentals. It reached its creative solution sua sponte,235 never affording the parties an opportunity to brief it. As RJR argued in its petition for rehearing, neither party had "argued that the panel was bound by Douglas [III] on any matter: not on its legal reasoning ... and certainly not on its supposed factfinding." RJR Petition for Panel Rehearing at 29, Walker 0 (Nos. 12-13500, 12-14731).

The panel's failure to solicit input from the parties had consequences. As RJR put it, the panel's full-faith-and-credit analysis was "demonstrably erroneous." Id. at 11. RJR explained,

Id. Because Alvin Walker and James Duke were not in privity with James L. Douglas, they could not enforce the judgment entered in Douglas III. E.C. v. Katz, 731 So.2d 1268, 1269 (Fla. 1999) ("[U]nless both parties are bound by the prior judgment, neither may use it in a subsequent action."). Nor could they rely on the opinion's factual findings. Forman v. Florida Land Holding Corp., 102 So.2d 596, 598 (Fla. 1958) ("Stare decisis relates only to the determination of questions of law, [and] ... has no relation whatever to the binding effect of determinations of fact."). In sum, Douglas III was not a rehearing of Engle III. It was merely another recognizing court interpreting the preclusive effect of the Phase I findings to one specific plaintiff. Thus, not only did the panel err in giving full faith and credit to Douglas III, its evaluation of the process afforded to the parties in that case was totally irrelevant.

"After [RJR's] first petition for rehearing [in response to Walker 0] explained that Florida law requires mutuality of parties, which Douglas [III] and [Walker] lack," the panel had no choice but to correct its error. Petition for Writ of Certiorari at 18 n.2, R.J. Reynolds Tobacco Co. v. Walker, 134 S.Ct. 2727 (No. 13-1193). But rather than abandon its analysis, which had been premised upon a fundamental misunderstanding of Florida law, the panel stuck to its inapposite guns. Instead of "giv[ing] full faith and credit to the decision in Douglas [III]," Walker 0, at 17, the panel "[gave] full faith and credit to the decision in Engle [III], as interpreted in Douglas [III]," Walker II, 734 F.3d at 1287. Never mind that the panel had determined that the Douglas III Court, not the Engle III Court, had been the one to determine the Phase I findings were specific enough "based on [the Douglas III Court's] review of the class action trial plan and the jury instructions."236 Walker 0, at 19 (citing Douglas III, 110 So.3d at 423). And never mind that the panel had conducted its Hansberry inquiry on Douglas III, not Engle III.237 The panel simply changed a few words in its opinion, and then reissued its opinion with all of its original analysis intact. In fact, the panel made substantive changes to only four sentences in its twenty-six page opinion.238 For ease of reference, the changes are identified in Table 1.

 Table 1

            Alterations From Walker 0 to Walker II

           Walker 0                    Walker II

"These principles require that we give       "These principles require that we give
full faith and credit to the decision in     full faith and credit to the decision in
Douglas [III] so long as it `satisf[ies]     the  Engle, as interpreted in Douglas [III], so
minimum procedural requirements' of          long as it `satisf[ies] the minimum
due process." Walker 0, at 17 (quoting       procedural requirements' of due
Kremer, 456 U.S. at 481, 102 S. Ct. at       process." Walker II, 734 F.3d at 1286
1897).                                       (quoting Kremer, 456 U.S. at 481, 102
                                             S. Ct. at 1897).

"Our inquiry is a narrow one: whether        "Our inquiry is a narrow one: whether
giving full faith and credit to the          giving full faith and credit to the
decision in Douglas [III] would              decision in Engle, as interpreted in
arbitrarily deprive R.J. Reynolds of its     Douglas [III], would arbitrarily
property without due process of law."        deprive R.J. Reynolds of its property
Walker 0, at 18.                             without due process of law." Walker
                                             II, 734 F.3d at 1287.

"And we cannot refuse to give full           "And we cannot refuse to give full
faith and credit to the decision in          faith and credit to the decision in
Douglas [III] because we disagree            Engle because we disagree with the
with its holding about what the jury in      decision in Douglas [III] about what
Phase I decided." Walker 0, at 18.           the jury in Phase I decided." Walker
                                             II, 734 F.3d at 1287.

"Nor does R.J. Reynolds identify any         "Nor does R.J. Reynolds identify any
other court that has declined to give        other court that has declined to give
full faith and credit to a judgment of a     full faith and credit to a judgment of a
state court about what issues were           state court as later interpreted by the
actually decided in a prior litigation       same state court on the ground that
on the ground that the state court           the later state court decision was so
decision was so wrong that it                wrong that it amounted to a violation
amounted to a violation of due               of due process." Walker II, 734 F.3d
process." Walker 0, at 22-23.                at 1289.

Aside from these changes, every aspect of the panel's opinion remained precisely the same. For example, it was still, in the panel's estimation, the Court in Douglas III — in its role as a recognizing court — that "was entitled to look beyond the jury verdict to determine what issues the jury decided." Walker 0, at 19; Walker II, 734 F.3d at 1287. And it was still, according to the Walker panel, the Court in Douglas III that fulfilled its role as a recognizing court, and "looked past the ambiguous jury verdict to decide this question of fact." Walker 0, at 23; Walker II, 734 F.3d at 1289. Further, it was still the Court's supposed factual decision in Douglas III that the Walker panel concluded was not arbitrary. Walker 0, at 20; Walker II, 734 F.3d at 1288. And, "if due process requires a finding that an issue was actually decided," then it was still, in the panel's opinion, the Court in Douglas III that "made the necessary finding."239 Walker 0, at 20; Walker II, 734 F.3d at 1289.

RJR took issue with the panel's "incomplete" and "semantic" revisions to its initial opinion. RJR Petition for Panel Rehearing or Rehearing En Banc at 4, 15, Walker II (Nos. 12-13500, 12-14731). In its petition for rehearing Walker II, RJR explained, "The panel's cosmetic revisions to the original opinion" "do not even fix the original decision's sua sponte — and obviously erroneous — application of the Full Faith and Credit Act to Douglas [III]." Id. at 14. RJR continued, "By revising its decision at all, the panel presumably recognized that Douglas [III] itself is entitled to no preclusive effect in these cases." Id. That is so because

Id. Yet, despite the panel's apparent understanding of this fact, RJR noted, it still "defer[ed] to the decision in Douglas [III]" and still failed to conduct even a cursory review of the decision made by the rendering Engle III Court.240 Id. at 15 (quoting Walker II, 734 F.3d at 1289). Moreover, in purporting to defer to a factual finding Douglas III never made, it also rejected Douglas III's portrayal of the Phase I findings as nonspecific liability determinations that are "useless" under an issue-preclusion framework. Douglas III, 110 So.3d at 433. Thus, according to RJR, the panel "concot[ed] and then `deferr[ed]' to an indefensible reading" of Douglas III. Petition for Writ of Certiorari at 29-30, R.J. Reynolds Tobacco Co. v. Walker, 134 S.Ct. 2727 (No. 13-1193).

* * *

The Walker panel erred in at least eight ways.241 First, it disregarded the plaintiffs' claim-preclusion proffers, replacing them with its own issue-preclusion proffer. In doing so, the panel not only shifted the plaintiffs' evidentiary burden,242 it improperly carried that burden for the plaintiffs.243 The panel neglected to consider whether its abrogation of two common-law protections rooted in fundamental fairness — the party asserting preclusion's burden of proof and the presumption against preclusion — violated RJR's rights to due process.

Second, after injecting its new theory of preclusion into the case on behalf of the plaintiffs, the panel failed to provide RJR with an opportunity to be heard on the theory's applicability to its case before entering judgment. The panel failed to consider the due process implications of denying RJR its right to be heard on a dispositive issue in the case against it.

Third, as the panel only hinted at, but did not commit to, what it believed is the defect that taints all cigarettes,244 it sanctioned an unconstitutional conclusive presumption that all smoking-related injuries are caused by the manufacturer's tortious conduct.245

Fourth, in sanctioning this conclusive presumption, the panel denied RJR's Seventh Amendment right to a jury trial on a contested and material element of the claims against them.

Fifth, the panel appeared to act as both advocate and arbiter. In doing so, it failed to consider whether this denied RJR of its due process right to an impartial decision maker

Sixth, in advocating for the plaintiffs, the panel effectively rewrote Douglas III in a strained attempt to reconcile it with our Brown II precedent.246 To its credit, the panel rejected Douglas III's preposterous portrayal of the Phase I findings as liability determinations. Yet, the panel then adopted an equally preposterous portrayal of the Phase I findings based, in turn, on a preposterous portrayal of Douglas III.247 As interpreted by the panel, when the Douglas III Court said that the Phase I findings would be "useless" under issue preclusion, 110 So.3d at 433, it really meant that the Phase I findings were dispositive under issue preclusion, like special verdicts that establish all the elements of liability under various tort theories.248 When the Douglas III Court said that "the Engle jury did not make detailed findings for which evidence it relied upon to make the Phase I common liability findings," id., it really meant that it had "looked past the ambiguous jury verdict[s]" to identify (but not disclose) precisely the evidence upon which the Phase I Engle jury relied.249 Walker 0, at 21; Walker II, 734 F.3d at 1289. After its absurd portrayal of Douglas III was in place, the panel then "disagree[d]" with Douglas III's supposed factual determinations, because such determinations were "erroneous."250 Walker 0, at 18; Walker II, 734 F.3d at 1287. The panel nevertheless gave full faith and credit to them. Walker 0, at 18; Walker II, 734 F.3d at 1287. The panel failed to evaluate whether its absurd depiction of Douglas III amounted to an arbitrary denial of RJR's right to due process.251

This, in turn, gave rise to the panel's seventh error — giving full faith and credit to an opinion the plaintiffs did not proffer and upon which they could not rely under Florida law. This error thoroughly tainted the panel's Walker 0 opinion. In that opinion, the panel not only erroneously gave full faith and credit to Douglas III, it conducted a totally irrelevant Hansberry inquiry on Douglas III, concluding that the Douglas III Court appropriately "look[ed] beyond the [Phase I] jury verdict[s]" to interpret them. Walker 0, at 19.

When the panel attempted to hastily correct its seventh error by vacating Walker 0 and issuing Walker II, it committed its most indefensible error of all. The panel left intact all of its tainted and inapposite reasoning from Walker 0. Walker II, therefore, gives full faith and credit to Engle III, yet it inexplicably reviews the process afforded to litigants in Douglas III, Walker II, 734 F.3d at 1287, while conducting no inquiry whatsoever on what was decided in Engle III or the process afforded to litigants in that proceeding.252 For example, the panel does not say whether the issues it precluded the defendants from litigating were decided in Engle III. Nor does it indicate whether any such issues were before the Court and decided with adequate notice and opportunity to be heard.253

Moreover, Walker II still claims that it was the Douglas III Court that "look[ed] beyond the [Phase I] jury verdict[s]" to interpret them. Id. Thus, the Walker II panel gave full faith and credit to Engle III, yet the relevant factual determination was made by the Douglas III Court. Thus, Walker II, issued by a federal appellate court tasked with interpreting the U.S. Constitution, holds that "if due process requires a finding that an issue was actually decided, then the [Douglas III] Court made the necessary finding." Walker II, 734 F.3d at 1289. Federal courts cannot discharge their constitutional duties by deferring to inapplicable state-court opinions.

Why did the panel choose to publish such a transparently nonsensical opinion? Was it because it had already backed itself into a corner with Walker 0? Walker 0 makes clear that the panel saw only "ambigu[ity]" when it looked at the Phase I findings. Walker 0, at 23. It "disagree[d] with [Douglas III's] [supposed] holding about what the jury in Phase I decided" and believed that holding was "erroneous." Id. at 18. The panel thus could not, in the immediate aftermath of Walker 0, reasonably contend in Walker II that it "looked beyond the jury verdict" for itself "to determine what issues the jury decided."254 Walker 0, at 19. Nor could the panel claim, with a straight face, that the rendering Engle III Court had been the one to "look beyond the jury verdict" after it had just finished explaining that the recognizing Douglas III Court had been the one to do so.255 Did the panel therefore resign to change a few words, muddy the waters a bit by referring to "Engle [III], as interpreted in Douglas [III]," and hope that this Court would not notice, or at least not fully grasp the implications of, its inapposite reasoning when faced with the defendants' petition for rehearing en banc? Walker II, 734 F.3d at 1286.

VI.

THE MAJORITY REPEAT AND ADD TO THE WALKER PANEL'S ERRORS

Mr. Graham's complaint was nearly identical to the plaintiffs' complaints in Walker I and Duke. Just like those complaints,256 Mr. Graham's alleged that Engle III was a "mandate" that "provided Plaintiff the opportunity to complete unresolved damages claims." Second Amended Complaint at 2, 3, Graham v. R.J. Reynolds Tobacco Co., No. 3:09-cv-13602-MMH-HTS (M.D. Fla. Dec. 5, 2011) (emphasis added). Just like Walker and Duke, Mr. Graham assumed, in accordance with Martin II, that all that remained to be resolved in his case was damages; the defendants were precluded by the "mandate" in Engle III from litigating both (1) that the cigarettes smoked by Ms. Graham were defective, unreasonably dangerous, and negligently produced and (2) that the defendants' tortious conduct caused Ms. Graham's death. Id. Accordingly, Mr. Graham pleaded only class membership — Ms. Graham's addiction and a smoking-related injury. Id. at 4. To support his assertion of claim preclusion, he proffered only the Engle III opinion and the "common liability findings" approved by the Engle III court. Id. at 5-9.

In response, the defendants contended that Mr. Graham failed to sufficiently plead claim preclusion. Philip Morris USA Inc.'s Answer to Plaintiff's Second Amended Complaint at 9, Graham I, No. 3:09-cv-13602-MMH-HTS; Answer, Defenses and Jury Demand of Defendant R.J. Reynolds Tobacco Co at 23, Graham I, No. 3:09-cv-13602-MMH-HTS. Moreover, they argued that Mr. Graham could not rely on either claim or issue preclusion because the Phase I findings are "inadequate to support an individualized determination of liability, legal causation, and damages in this or any other subsequent individual action." Answer, Defenses and Jury Demand of Defendant R.J. Reynolds Tobacco Co at 2, Graham I, No. 3:09-cv-13602-MMH-HTS; see also Philip Morris USA Inc.'s Answer to Plaintiff's Second Amended Complaint at 10, Graham I, No. 3:09-cv-13602-MMH-HTS (making the same argument).

The District Court, relying on Waggoner's pretrial ruling, accepted Mr. Graham's claim-preclusion proffer of Engle III and its approved findings as adequate to establish that RJR and Philip Morris were liable to all class members, including Mr. Graham. Accordingly, the District Court precluded RJR and Philip Morris from contesting, and did not instruct the jury to determine, (1) that the cigarettes Ms. Graham smoked were defective, unreasonably dangerous, and negligently produced and (2) that the defendants' tortious conduct caused Ms. Graham's death. Ante at 1180. The jury thus ruled in Mr. Graham's favor because he proved, to their satisfaction, his status as a class member. In sum, the Graham trial played out in much the same way as the Walker I and Duke trials.

Thus, we are confronted with the same question with which the Walker panel was confronted: Did the District Court deprive RJR and Philip Morris of property without due process of law by applying claim preclusion so as to preclude the defendants from litigating (1) that the cigarettes Ms. Graham smoked were defective, unreasonably dangerous, and negligently produced and (2) that the defendants' tortious conduct caused Ms. Graham's death?

In answering this question, the Majority implicitly acknowledge that the District Court erred by applying claim preclusion. They "reaffirm" Walker II, Ante at 1174, in which the panel rejected Douglas III's portrayal of the Phase I findings as liability determinations. See Walker II, 734 F.3d at 1280 ("[T]he [Phase I] jury did not decide whether the tobacco companies were liable for damages to individual members of the class.").

Nevertheless, following the Walker panel's basic analytical framework, the Majority conclude that the District Court's error was harmless. Like the Walker panel, the Majority reach this conclusion by portraying the Phase I findings as "specific enough" factual findings — a portrayal Mr. Graham neither advanced nor proffered evidence to support. Ante at 1182. That is, they view the Phase I findings as establishing that all cigarettes are defective, unreasonably dangerous, and negligently sold. See id. ("The Florida Supreme Court rejected [the] argument" that "the jury did not necessarily find that all cigarettes the defendants placed on the market were defective and unreasonably dangerous."). In adopting such a portrayal, they attempt to correct for the District Court's barring the defendants from litigating whether the cigarettes Ms. Graham smoked were defective, unreasonably dangerous, and negligently sold. The Majority also suggest that the unreasonably dangerous defect that taints all cigarettes is that they "cause disease and are addictive."257 Id. at 1182. In hinting at this portrayal,258 the Majority attempt to correct for the District Court's barring the defendants from litigating whether their tortious conduct caused Ms. Graham's death.259

Whereas the Walker panel imputed its portrayal of the Phase I findings as "specific enough" factual findings to Douglas III, the Majority attribute their identical portrayal to Engle III.260 They assert, without offering evidence to support their assertion, that the Engle III Court was the one to "interpret[ ] [the Phase I] findings to determine what the jury actually decided."261 Id. at 1185. In doing so, they sidestep the mutuality obstacle that doomed the Walker 0 opinion. See supra Part V.

In support of their portrayal, which they attribute to Engle III, the Majority mine the Phase I trial record and proffer excerpts for the plaintiff to establish a theory of preclusion the plaintiff did not advance. Id. at 1175-77, 1182-83. On the basis of their own evidentiary proffer, they appear to conclude that Engle III's supposed portrayal of the Phase I findings is not arbitrary262 because a properly instructed jury could have made such findings.263 See, e.g., id. at 1182 ("[T]he jury's answers on the verdict form, when read together with the entire record, were consistent with the general theories that the tobacco companies' cigarettes are defective and the sale of their cigarettes is negligent because all of those cigarettes cause disease and are addictive." (emphasis added)).

Though the Majority give full faith and credit to Engle III, id. at 1174, 1180, they do not perform their recognizing-court inquiries as Florida law and the U.S. Constitution require.264 Accordingly, the Majority do not evaluate whether the Engle III Court violated the defendants' due process rights by making a dispositive determination about the Phase I findings secretly — that is, without revealing that it had done so — and without affording the parties notice or opportunity to be heard.265

* * *

Sitting in the same posture as the Walker panel, the Majority replicate and add to the constitutional violations the panel committed in its blind affirmance of judgments entered in Duke and Walker I. First, like the Walker panel, the Majority disregard the plaintiff's claim-preclusion proffer, replacing it with their own issue-preclusion proffer.266 In doing so, the Majority not only shift the plaintiffs' evidentiary burden,267 they improperly carry that burden for the plaintiffs.268 Just as the Walker panel did, the Majority leave unanswered whether their abrogation of two common-law protections rooted in fundamental fairness — the party asserting preclusion's burden of proof and the presumption against preclusion — deny the defendants' rights to due process.

Second, following the Walker panel's lead, after injecting their own theory of preclusion into the case in place of the plaintiff's, the Majority fail to provide the defendants with an opportunity to be heard on the theory's applicability to their case before entering judgment. Despite the Majority's demonstrated willingness to allow the parties to file supplemental briefs,269 they refuse to provide the defendants their constitutional right to be heard on the accuracy of the Majority's novel factual determination that the Engle III Court determined that the Phase I findings were "specific enough" factual findings.270

Third, in a notice-and-opportunity-to-be-heard double whammy, the Majority conduct no Hansberry examination into Engle III to evaluate whether the "interpret[tation]" of the Phase I findings they attribute to Engle III, and to which they give full faith and credit, was rendered with adequate notice and opportunity to be heard.271 They do not, for example, mention the proceedings in Engle II or describe what issues were before the Court in Engle III.272 Had the Majority conducted such an evaluation, they would have recognized that the parties in that case were not on notice that the Engle III Court was going to make a determination about how progeny courts should interpret the Phase I findings.273 They would have recognized that the parties had no opportunity to be heard on such a determination, and it is therefore not entitled to preclusive effect.274

Fourth, in acting both as advocate and arbiter and attributing to the Engle III Court a portrayal of the Phase I findings that it expressly declined to make, the Majority violate the defendants' due process right to an impartial decision maker.

Fifth, as the Majority only hint at, but do not commit to, what they believe is the defect that taints all cigarettes,275 they sanction an unconstitutional conclusive presumption that all smoking-related injuries are caused by the manufacturer's tortious conduct.276

Sixth, in sanctioning this conclusive presumption, the Majority deny the defendants' Seventh Amendment right to a jury trial on a contested and material element of the claims against them.

In engaging in their interpretational endeavors and causing or sanctioning the constitutional violations described above, was the Majority attempting to reach a particular outcome? Was that outcome to ensure that Engle-progeny plaintiffs secure the same result in federal courts as they would achieve in Florida courts so as to avoid unequal treatment resulting from the accident of diversity jurisdiction? If so, the Majority fail to recognize the many constitutional impediments to their desired outcome. They also fail to consider — as the Walker panel did when it issued its Walker 0 opinion — the impact of Florida's complete-mutuality requirement. The Majority's conclusion that the District Court did not deny RJR's and Philip Morris's due process rights by precluding them from contesting (1) that the cigarettes Ms. Graham smoked were defective, unreasonably dangerous, and negligently produced and (2) that their tortious conduct caused her harm hinges on the Majority's (demonstrably false) factual determination that the Engle III Court determined that the Phase I findings of fact were "specific enough" to establish that Ms. Graham's death was caused by unreasonably dangerous defects in RJR's and Philip Morris's cigarettes and by RJR's and Philip Morris's negligent conduct. Class plaintiffs who are not present in this case will be unable to rely on the Majority's fact findings.277 Future district courts attempting to implement the Majority's reasoning will thus need to make their own determination that the plaintiff before them made a sufficient issue-preclusion proffer, including proof to "a reasonable degree of certainty" that the Engle III Court determined that the Phase I findings are "specific enough" to preclude the defendants from contesting essential elements of the claims against them. As recognizing courts with Hansberry duties of their own, they will also need to make their own determination about whether the Engle III Court made such a factual determination with adequate notice and opportunity to be heard. In short, the only proposition for which the Majority's opinion stands is that the judgments rendered against RJR and Philip Morris and in favor of Mr. Graham are affirmed.

VII.

THE FUNCTIONAL BAN ON CIGARETTES IS PREEMPTED BY FEDERAL LAW

In addition to holding that the Engle defendants' due process rights were not violated, the Majority also hold that federal law does not preempt Florida's functional ban on cigarettes. Consistent with its due process analysis, the Majority "constru[e] the [Phase I] findings as embracing a theory that all cigarettes manufactured by the tobacco companies are defective and the sale of all cigarettes is negligent." Ante at 1186. Nevertheless, the Majority hold that the "six tobacco-specific [federal] laws that are relevant to this appeal" do not "reflect[ ] a federal objective to permit the sale or manufacture of cigarettes." Id. at 1186, 1188. Thus, because banning cigarettes would not be preempted, Florida law "regulat[ing] cigarette sales" by "impos[ing] tort liability on cigarette manufacturers" is not preempted. Id. at 1191.

Not only do the Majority's and the Douglas III's Court's preclusion regimes both engender severe due process violations, both are preempted by federal law. Under the Majority's regime, every cigarette is defective and unreasonably dangerous and the very act of selling cigarettes is a breach of a duty of care. Under the Douglas III Court's regime, tobacco manufacturers are presumed liable for any smoking-related injury. Either way, under Douglas III's claim-preclusion framework or the Majority's issue-preclusion framework, cigarettes have effectively been banned. Though the Majority and I agree on this point, we disagree about whether such a ban is preempted by federal law. I believe that it is.

Our constitutional system contemplates "that both the National and State governments have elements of sovereignty the other is bound to respect." Arizona v. United States, 567 U.S. 387, 398, 132 S.Ct. 2492, 2500, 183 L.Ed.2d 351 (2012). When state and federal law "conflict or [otherwise work] at cross-purposes," id., the Supremacy Clause commands that federal law "shall be the supreme Law of the Land ... any Thing in the Constitution or Laws of any State to the Contrary notwithstanding." U.S. Const. art. VI, cl. 2. That is, state laws that "interfere with, or are contrary to," federal law cannot hold sway — they "must yield." Gibbons v. Ogden, 22 U.S. (9 Wheat.) 1, 211, 6 S.Ct. 23 (1824).

Federal law may preempt state law in three ways. First, Congress has the authority to expressly preempt state law by statute. Crosby v. Nat'l Foreign Trade Council, 530 U.S. 363, 372, 120 S.Ct. 2288, 2293, 147 L.Ed.2d 352 (2000). Second, even in the absence of an express preemption provision, "[t]he scheme of federal regulation may be so pervasive as to make reasonable the inference that Congress left no room for the States to supplement it." Rice v. Santa Fe Elevator Corp., 331 U.S. 218, 230, 67 S.Ct. 1146, 1152, 91 S.Ct. 1447 (1947). Third, federal and state law may impermissibly conflict, for example, "where it is impossible for a private party to comply with both state and federal law," Crosby, 530 U.S. at 372, 120 S.Ct. at 2294; or when the state law at issue "stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress," Hines v. Davidowitz, 312 U.S. 52, 67, 61 S.Ct. 399, 404, 85 S.Ct. 581 (1941).278 It is this last subcategory of conflict preemption — obstacle preemption — the Court faces here.279

A. Obstacle Preemption

Obstacle preemption leaves the tobacco companies with a tough row to hoe. Supreme Court precedent teaches that "a high threshold must be met if a state law is to be preempted for conflicting with the purposes of a federal Act." Chamber of Commerce v. Whiting, 563 U.S. 582, 607, 131 S.Ct. 1968, 1985, 179 L.Ed.2d 1031 (2011) (quotation marks omitted). Indeed, "[i]mplied preemption analysis does not justify a freewheeling judicial inquiry into whether a state statute is in tension with federal objectives." Id. (quotation marks omitted). That is because "such an endeavor would undercut the principle that it is Congress rather than the courts that preempts state law." Id. (quotation marks omitted).

In addition to overcoming this "high threshold," the tobacco companies must also confront the presumption against preemption — namely, that "we start with the assumption that the historic police powers of the States were not to be superseded by [federal law] unless that was the clear and manifest purpose of Congress." Rice, 331 U.S. at 230, 67 S.Ct. at 1152.280 The presumption is a "cornerstone[] of our pre-emption jurisprudence."281 Wyeth v. Levine, 555 U.S. 555, 565, 129 S.Ct. 1187, 1194, 173 L.Ed.2d 51 (2009). And its logic carries particular force when, as here, "federal law is said to bar state action in fields of traditional state regulation." N.Y. State Conference of Blue Cross & Blue Shield Plans v. Travelers Ins. Co., 514 U.S. 645, 655, 115 S.Ct. 1671, 1676, 131 L.Ed.2d 695 (1995). We must recognize, therefore, "the historic primacy of state regulation of matters of health and safety," which can be enforced through state statutes and state tort law alike.282 Medtronic, Inc. v. Lohr, 518 U.S. 470, 485, 116 S.Ct. 2240, 2250, 135 L.Ed.2d 700 (1996). Given the "great latitude" that states possess "under their police powers to legislate as to the protection of the lives, limbs, health, comfort, and quiet of all persons," id. at 475, 116 S.Ct. at 2245 (quotation marks omitted), we will not ascribe to Congress the intent "cavalierly [to] pre-empt state-law causes of action," id. at 485, 116 S.Ct. at 2250. To do otherwise would ignore altogether that "[t]he allocation of powers in our federal system preserves the integrity, dignity, and residual sovereignty of the States." Bond v. United States, 564 U.S. 211, 221, 131 S.Ct. 2355, 2364, 180 L.Ed.2d 269 (2011).

Finally, the lodestar of any preemption inquiry is congressional intent. Retail Clerks Int'l Ass'n v. Schermerhorn, 375 U.S. 96, 103, 84 S.Ct. 219, 223, 11 L.Ed.2d 179 (1963). In assessing the extent to which state law "stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress," Hines, 312 U.S. at 67, 61 S.Ct. at 404, "[w]hat [constitutes] a sufficient obstacle is a matter of judgment, to be informed by examining the federal statute as a whole and identifying its purpose and intended effects," Crosby, 530 U.S. at 373, 120 S.Ct. at 2294. To begin, then, "we must first ascertain the nature of the federal interest." Hillman, 569 U.S. at ___, 133 S.Ct. at 1950.

B. Federal Regulation of Tobacco is Premised on Consumers' Ability to Choose

By my count, Congress has enacted at least seven statutes regulating tobacco products in the past fifty years.283 I examine their text and structure, which provide the most reliable indicia of what Congress has resolved itself to achieve. CTS Corp. v. Waldburger, 575 U.S. ___, ___, 134 S.Ct. 2175, 2185, 189 L.Ed.2d 62 (2014). This amounts to the "classic judicial task of reconciling many laws enacted over time, and getting them to `make sense' in combination." United States v. Fausto, 484 U.S. 439, 453, 108 S.Ct. 668, 676-77, 98 L.Ed.2d 830 (1988).

I start with first principles. Congress possesses the constitutional authority to ban cigarettes. See U.S. Const., art. I, § 8, cl. 3. It has never done so. This, despite an ever-growing body of research documenting the health risks associated with smoking. In 1964, for example, the Surgeon General issued a report concluding that "[c]igarette smoking is a health hazard of sufficient importance in the United States to warrant appropriate remedial action." Advisory Comm. to the Surgeon Gen. of the Public Health Serv., U.S. Dep't of Health, Educ., & Welfare, Smoking and Health 33 (1964), available at http://profiles.nlm.nih.gov/ps/access/NNBBMQ.pdf. The report warned "that cigarette smoking contributes substantially to mortality from certain specific diseases and to the overall death rate." Id. at 31.

These findings spurred legislative action. Congress's first attempt to address cigarette smoking and its consequences came in the Federal Cigarette Labeling and Advertising Act (the "Labeling Act"), Pub. L. No. 89-92, 79 Stat. 282 (1965) (codified as amended at 15 U.S.C. §§ 1331-1341). The Labeling Act aimed to "establish a comprehensive Federal program to deal with cigarette labeling and advertising." Id. § 2. Central to this comprehensive program was a requirement that all cigarette packages display the warning statement, "Caution: Cigarette Smoking May Be Hazardous to Your Health." Id. § 4.

For our purposes, the Labeling Act is instructive because it encapsulates the competing interests Congress has sought to reconcile when regulating cigarettes. On one hand, Congress has recognized that smoking can cause serious physical harm, even death. On the other hand, Congress has also acknowledged the important role tobacco production and manufacturing plays in the national economy. Congress has carefully calibrated these policy considerations by promoting full disclosure to consumers about the attendant risks tobacco products carry, thereby permitting consumers to make to a free but an informed choice. The plain language of the Labeling Act summarizes well this approach:

Id. § 2284

Since the Labeling Act's passage, Congress's basic goals have remained largely unchanged. For example, Congress has tinkered with the text of the warning labels affixed to cigarette packages in an effort to arm consumers with more complete and accurate information. Public Health Cigarette Smoking Act of 1969, Pub. L. No. 91-222, § 4, 84 Stat. 87 (codified as amended at 15 U.S.C. § 1333); Comprehensive Smoking Education Act, Pub. L. No. 98-474, § 4, 98 Stat. 2200 (1984) (codified at 15 U.S.C. § 1333). To promote transparency, Congress has required the Secretary of Health and Human Services to issue a report to Congress every three years regarding the "addictive property of tobacco." Alcohol and Drug Abuse Amendments of 1983, Pub. L. No. 98-24, 97 Stat. 175. Congress has stepped in also to regulate smokeless tobacco products. Comprehensive Smokeless Tobacco Health Education Act of 1986, Pub. L. No. 99-252, 100 Stat. 30. And Congress has even incentivized states to prohibit the sale of tobacco products to minors by conditioning block grants on the creation of programs "to discourage the use of ... tobacco products by individuals to whom it is unlawful to sell or distribute such ... products." Alcohol, Drug Abuse, and Mental Health Administration Reorganization Act, Pub. L. No. 102-321, § 202, 106 Stat. 323 (1992) (codified at 42 U.S.C. § 300×22).

All this, but Congress has enacted no ban on the sale of cigarettes to adult consumers. No ban even though over the last fifty years a scientific consensus has emerged that smoking can kill. The Surgeon General has reaffirmed this, at least twice. Office of the Surgeon Gen., U.S. Dep't of Health & Human Servs., The Health Consequences of Smoking: Nicotine Addiction (1988), available at http://profiles.nlm.nih.gov/ps/access/NNBBZD. pdf; Office of the Surgeon Gen., U.S. Dep't of Health & Human Servs., The Health Consequences of Smoking — 50 Years of Progress (2014), available at http://www.surgeongeneral.gov/library/reports/50-years-of-progress/full-report. pdf. The Environmental Protection Agency has classified secondhand smoke as a known human carcinogen. Office of Health & Envtl. Assessment, Respiratory Health Effects of Passive Smoking: Lung Cancer and Other Disorders 4 (1992), available at http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=2835. The Food and Drug Administration (the "FDA") has published research indicating that "[t]he pharmacological processes that cause [nicotine addiction] are similar to those that cause addiction to heroin and cocaine." FDA, Jurisdictional Determination, 61 Fed. Reg. 44619, 44631 (Aug. 28, 1996). These are, of course, but a few examples.

In short, Congress has known about the dangers of cigarettes for many years. Congress has regulated cigarettes for many years. But it has never banned them. Indeed, regulation of cigarettes rests on the assumption that they will still be sold and that consumers will maintain a "right to choose to smoke or not to smoke." H.R. Rep. No. 89-449 (1965), reprinted in 1965 U.S.C.C.A.N. 2350, 2352.

The Supreme Court has so concluded, holding that the FDA lacked jurisdiction to regulate cigarettes because it would have otherwise been required by statute to prohibit their sale. FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 161, 120 S.Ct. 1291, 1315-16, 146 L.Ed.2d 121 (2000). This result, the Court determined, would have contravened the intent of Congress, given that "the collective premise of these statutes is that cigarettes and smokeless tobacco will continue to be sold in the United States." Id. at 139, 120 S.Ct. at 1304.

And although Congress has since overruled this decision, granting the FDA regulatory authority over cigarettes in 2009, Congress nonetheless stated that the FDA "is prohibited from" "banning all cigarettes" or "requiring the reduction of nicotine yields of a tobacco product to zero." Family Smoking Prevention and Tobacco Control Act ("the TCA"), Pub. L. No. 111-31, § 907(d)(3)(A)-(B), 123 Stat. 1776 (2009) (codified at 21 U.S.C. § 387g). To be sure, the TCA does not "affect any action pending in Federal ... court" prior to its enactment — including this one. Id. § 4(a)(2); see Engle III, 945 So.2d at 1277 (noting that Engle-progeny cases must be filed within one year of the issuance of the case's mandate). It merely makes textually explicit what was already evident by negative implication: Congress never has intended to prohibit consumers from purchasing cigarettes. To the contrary, it has designed "a distinct regulatory scheme" to govern the product's advertising, labelling, and — most importantly — sale. Brown & Williamson, 529 U.S. at 155, 120 S.Ct. at 1312.

C. Florida Has Imposed a Duty Not to Sell Cigarettes Contrary to Federal Law

I now turn to how these federal objectives interact with state law. Federal law can expressly or impliedly preempt a state tort suit. E.g., Geier v. Am. Honda Motor Co., 529 U.S. 861, 120 S.Ct. 1913, 146 L.Ed.2d 914 (2000) (finding implied preemption of state tort suit); Cipollone v. Liggett Grp., Inc., 505 U.S. 504, 112 S.Ct. 2608, 120 L.Ed.2d 407 (1992) (plurality opinion) (finding express preemption of certain state tort suits); see generally Williamson v. Mazda Motor of Am., Inc., 562 U.S. 323, 330, 131 S.Ct. 1131, 1136, 179 L.Ed.2d 75 (2011) (collecting cases). A tort is "a breach of a duty that the law imposes on persons who stand in a particular relation to one another." Tort, Black's Law Dictionary 1626 (10th ed. 2014). As such, successful tort actions "are premised on the existence of a legal duty." Cipollone, 505 U.S. at 522, 112 S.Ct. at 2620 (plurality opinion); see also Geier, 529 U.S. at 881, 120 S.Ct. at 1925 (characterizing a successful tort action as "a state law — i.e., a rule of state tort law imposing ... a duty"). Strict-liability and negligence claims like those at issue here are no exception. Mut. Pharm. Co. v. Bartlett, 570 U.S. ___, ___, 133 S.Ct. 2466, 2474 n.1, 186 L.Ed.2d 607 (2013) ("[M]ost common-law causes of action for negligence and strict liability ... exist ... to ... impose affirmative duties."); Samuel Friedland Family Enters. v. Amoroso, 630 So.2d 1067, 1068 n.3 (Fla. 1994) (recognizing, in the strict-liability context, that "[o]ne who sells any product in a defective condition unreasonably dangerous to the user or consumer ... is subject to liability for physical harm thereby caused" even though "the seller has exercised all possible care in the preparation and sale of his product" (quoting Restatement (Second) of Torts § 402A)); Curd v. Mosaic Fertilizer LLC, 39 So.3d 1216, 1227 (Fla. 2010) (noting that a negligence claim requires identification of "[a] duty, or obligation, recognized by the law, requiring the [defendant] to conform to a certain standard of conduct, for the protection of others against unreasonable risks" (citation omitted) (second alteration in original)).

These duties, moreover, can stand as just as much of an obstacle to the purposes and objectives of Congress as a state statute or administrative regulation. E.g., Williamson, 562 U.S. at 329, 131 S.Ct. at 1136; Geier, 529 U.S. at 886, 120 S.Ct. at 1928. That is because, like any statute, common-law duties amount to "either affirmative requirements or negative prohibitions." Cipollone, 505 U.S. at 522, 112 S.Ct. at 2620 (plurality opinion). This Court's job, then, is to determine whether the legal duties underpinning Graham's strict-liability and negligence claims stand impermissibly as an obstacle to the achievement of federal objectives — here, regulating, but not banning, the sale of cigarettes. To accomplish this task, we must once again return to Engle and its progeny.

State laws, like that created by Douglas III, are broadly applicable, not restrained by mutuality rules or class membership. Cf. Palsgraf v. Long Island R.R. Co., 248 N.Y. 339, 162 N.E. 99 (1928) (expounding as a matter of state law the rule that in tort a duty is owed only to foreseeable plaintiffs). In other words, under Douglas III, anyone who is addicted to cigarettes can hold any tobacco company liable for damages simply by proving addiction and injury — "injury as a result of ... conduct is assumed."285 Douglas III, 110 So.3d at 429.

That private litigants, rather than executive agencies, are enlisted to enforce the ban does not diminish its potency. Although no executive agency intervenes to prevent tobacco companies from continuing to sell cigarettes while paying the resulting damages, "pre-emption cases do not ordinarily turn on such compliance-related considerations as whether a private party in practice would ignore state legal obligations — paying, say, a fine instead — or how likely it is that state law actually would be enforced." Geier, 529 U.S. at 882, 120 S.Ct. at 1926; cf. Cipollone, 505 U.S. at 521, 112 S.Ct. at 2620 (plurality opinion) (noting that state regulation "can be as effectively exerted through an award of damages as through some form of preventive relief. The obligation to pay compensation can be, indeed is designed to be, a potent method of governing conduct and controlling policy." (quotation marks omitted)).

Admittedly, how compliance-related considerations should factor into preemption analysis — if at all — remains something of an open question. "The Court has on occasion suggested that tort law may be somewhat different, and that related considerations — for example, the ability to pay damages instead of modifying one's behavior — may be relevant for pre-emption purposes." Geier, 529 U.S. at 882, 120 S.Ct. at 1926.286 We do not write on a blank slate, however. Justice Blackmun's opinion for himself and two other Justices in Cipollone forcefully contends that tort law should be treated differently from positive enactments for preemption purposes. 505 U.S. at 536-37, 112 S.Ct. at 2627-28 (Blackmun, J., concurring in part and dissenting in part) ("The effect of tort law on a manufacturer's behavior is necessarily indirect.... The level of choice that a defendant retains in shaping its own behavior distinguishes the indirect regulatory effect of the common law from positive enactments such as statutes and administrative regulations."). But Justice Blackmun lost that argument: his opinion did not command a majority. And critically, his logic was called into question by a majority of the Court in Geier. 529 U.S. at 882, 120 S.Ct. at 1926 ("[T]his Court's pre-emption cases ordinarily assume compliance with the state-law duty in question."). Absent more specific guidance from the Supreme Court, this Court must follow Geier's lead in assuming that the tobacco companies will comply with whatever state-law duties Florida may impose.

Nor is it convincing to argue that Congress, well aware of state tort litigation against the tobacco companies, would not have intended to preempt state-law claims similar to the two at issue here. See Wyeth v. Levine, 555 U.S. 555, 574-75, 129 S.Ct. 1187, 1200, 173 L.Ed.2d 51 (2009) ("If Congress thought state-law suits posed an obstacle to its objectives, it surely would have enacted an express pre-emption provision.... Its silence on the issue, coupled with its certain awareness of the prevalence of state tort litigation, is powerful evidence that Congress did not intend [to preempt state tort suits.]"); cf. Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U.S. 141, 166-67, 109 S.Ct. 971, 986, 103 L.Ed.2d 118 (1989). That proposition may be true at a high level of generality. But as I have explained in great detail, Graham's is not a run-of-the-mill tort suit. If it were, our analysis would be radically different. Make no mistake: this opinion should not be taken to mean that I believe Congress intended to insulate tobacco companies from all state tort liability. To the contrary, there is nothing in the text, structure, or legislative history of the federal statutes examined above to support such a far-reaching proposition.

I merely conclude that, having surveyed both federal and state law, it is clear that Congress would have intended to preempt Graham's strict-liability and negligence claims, rooted as they are in a broadly applicable state law set forth by the Florida Supreme Court that deems all cigarettes defective, unreasonably dangerous, and negligently produced. I therefore express no opinion as to other state-law suits that may rest on significantly narrower theories of liability.

D. The Majority Misinterpret the Statutory Framework of Tobacco Regulation

At bottom, the Majority and I disagree over how to understand the federal statutory framework regulating tobacco in place at the time of Engle III and Douglas III and how to understand the Supreme Court's decision in Brown & Williamson interpreting that framework. Though the importance of our disagreement should not be minimized — and given the uncertainty surrounding this particular issue and preemption generally, I would once again urge the Supreme Court to clarify the hazy state of preemption law — I end by highlighting the narrow scope of our disagreement. Like the Majority, I agree that the Tenth Amendment guarantees that "State governments retain their historic police powers to protect public health." See ante at 1190. Also like the Majority, I fully embrace the "happy incident" of Our Federalism that States are generally free to serve as the proving grounds for "novel social and economic experiments" of how best to further public health. See id. (quoting New State Ice Co. v. Liebmann, 285 U.S. 262, 311, 52 S.Ct. 371, 386-87, 76 S.Ct. 747 (1932) (Brandeis, J., dissenting)). And like the Majority, I see no reason to question the constitutionality of Tennessee's ability to ban the sale of cigarettes at the end of the 19th century, long before Congress began regulating tobacco nationally in any significant fashion. See id.; Austin v. Tennessee, 179 U.S. 343, 21 S.Ct. 132, 45 S.Ct. 224 (1900).

Our disagreement is simply this. I understand the federal statutory framework regulating tobacco in place at the time of Engle III and Douglas III, as the Supreme Court's decision in Brown & Williamson confirms, to allow states wide leeway to concurrently regulate tobacco while prohibiting states from banning the sale of cigarettes outright. As a result, we cannot give effect to the Florida Supreme Court's decisions in a manner that operates as a ban on the sale of cigarettes without elevating state law over federal law, which the Supremacy Clause forbids. The Majority come to the opposite conclusion. Because I believe the Majority err by doing so, I must dissent from the Majority's preemption holding as well.

CONCLUSION

In 2003, the Third District Court of Appeal effectively ended the Engle litigation. It decertified the Engle class because the Phase I proceeding had failed to achieve its purpose — the Phase I jury was not instructed to make "specific findings as to any act by any defendant at any period of time," much less "determine whether defendants were liable to anyone." Engle II, 853 So.2d at 450, 467 n.8. With no useful findings on which to rely, plaintiffs could sue tobacco manufacturers in individual lawsuits, but they would have to start from scratch.

Three years later, the Florida Supreme Court, lamenting the demise of the Engle litigation, sua sponte crafted a "pragmatic solution" designed to rejuvenate it. Engle III, 945 So.2d at 1269. Its solution entailed "retaining [most of] the jury's Phase I findings" — the only things that were left standing after the Court's decertification of the class — and puzzlingly declaring, in dicta, that those findings "will have res judicata effect in [progeny] trials." Id. In retaining these useless findings and directing the hundreds of thousands of class members to file claims within the year, the Court sent a signal to progeny courts that they should attempt to develop some rationale for propping up the plaintiffs' cases and allowing them to recover.

The result, to put it mildly, "was some confusion among the courts." R.J. Reynolds Tobacco Co. v. Marotta, 214 So.3d 590, 593 (Fla.2017). One Judge compared litigating and adjudicating a progeny case to "play[ing] legal poker, placing ... bets on questions left unresolved by Engle [III]." Jimmie Lee Brown II, 70 So.3d at 720 (May, J. concurring). When this Court reviewed an Engle-progeny case in Brown II, we placed our bet on a belief that the Florida Supreme Court had not attempted to secretly transform the useless Phase I findings into a hefty jackpot for Engle class members. See supra Part III.B.

Recognizing that our bet effectively ended the game for progeny plaintiffs, the First District Court of Appeal, in Martin II, made a different bet. It sua sponte interpreted Engle III as implicitly holding sua sponte that the Phase I findings were far from useless — "[n]o matter the wording of the findings on the Phase I verdict form." Martin II, 53 So.3d at 1067. Rather than being useless, the findings established the defendants' liability to all class members. Engle III's holding to this effect (to say nothing of its secret and sua sponte nature) was not a problem, the First District reasoned, because sufficient evidence was presented at trial upon which a properly instructed jury could have found that the defendants were liable to all class members.

Hoping to avoid stepping into the fray and upsetting a rationale advanced by the First District, the Florida Supreme Court denied certiorari review of Martin II. But it soon became clear that the Florida Supreme Court would need to reenter the scene to sort out the reckless betting it set in motion. Shortly after Martin II was issued, the Second and Fourth Districts placed slightly different bets. They understood, as Martin II did, that they were supposed to hold the defendants liable by "some mechanism." They acquiesced in that respect, as they felt "constrained," Jimmie Lee Brown II, 70 So.3d at 715, to achieve the result the Engle III Court seemed to desire. However, they had serious reservations about doing so, and could not coherently explain how "the smokers would prove causation in individual cases," Ante at 1178, when the Phase I findings do not even reveal the basis for the supposed defect in the defendants' cigarettes. In the midst of such bewilderment, the Second District asked the Florida Supreme Court for help with a certified question.

To settle "the confusion among the courts," Marotta, at 593, the Douglas III Court reluctantly accepted the certified question, and chose the rationale it liked best. Preferring the Martin II outcome, it declared the First District's bet a winner and ours a loser. Endorsing Martin II's analysis, the Florida Supreme Court concluded that it had sua sponte secretly ruled in 2006 that the Phase I findings were nonspecific liability determinations — "useless" under an issue-preclusion framework, but dispositive under a claim-preclusion framework. Douglas III, 110 So.3d at 433. Progeny plaintiffs thus had nothing to do but plead claim preclusion, proffer the Engle III opinion and the Phase I findings, prove their class membership and damages, and defend against claims of comparative fault. Granted, it was outlandish for the Douglas III Court to suggest that Engle III adjudicated the claims of all the absent and unidentified class members, but in the Engle-progeny poker parlor, the house always wins. Under Martin II's claim-preclusion rationale, progeny courts could be saved from the embarrassing and impossible task of explaining how the unidentified defect in the Phase I findings caused each class plaintiff's harm, because this aspect of the plaintiff's claim would be treated as having been established in Engle III. Eager to stop the progeny courts from asking questions, Douglas III endorsed the rationale.

When the Walker panel entered the parlor, it was faced with the question of whether such a preclusion regime violated the Engle defendants' due process rights. In tackling this question, the panel rejected the Douglas III Court's portrayal of the Phase I findings as nonspecific liability determinations, noting language in Engle III that contradicted that portrayal. Instead, it adopted a portrayal of the Phase I findings that the plaintiffs before it had not advanced and that no court had adopted previously: the Phase I findings were factual determinations that were "specific enough" to identify the conduct the Phase I jury deemed tortious. Walker 0, at 1287-88. The panel then imputed to the Douglas III Court that portrayal and sua sponte gave full faith and credit to it, not realizing that Florida law forbade giving full faith and credit to Douglas III. Upon realizing its mistake, the panel was forced to vacate its opinion and issue another one. But instead of rethinking its counterfactual rationale, the panel simply gave full faith and credit to Engle III instead of Douglas III and left all of its inapposite reasoning intact. The panel, therefore, never evaluated what Engle III decided or whether the Engle III Court violated the defendants' due process rights by making a dispositive determination about the Phase I findings secretly and without affording the parties notice or opportunity to be heard.

The Majority now double down on the Walker panel's misplaced bet. Like the Walker panel, the Majority reject the Douglas III Court's portrayal of the Phase I findings as liability determinations. Instead, echoing the Walker panel, they adopt a portrayal of the Phase I findings that Mr. Graham neither advanced nor proffered evidence to support. This time, instead of imputing their "specific enough" portrayal to Douglas III, they impute it to Engle III. In support of their portrayal, the Majority mine the trial record and proffer excerpts for the plaintiff, concluding that Engle III's supposed portrayal of the Phase I findings is not arbitrary because a properly instructed jury could have made such findings. Although the Majority did not make the mistake of giving full faith and credit to Douglas III, they still fail to evaluate, as Florida law and the U.S. Constitution require, whether the Engle III Court violated the defendants' due process rights by making a dispositive determination about the Phase I findings secretly and without affording the parties notice or opportunity to be heard.

In short, Engle III sent a signal to progeny courts to develop a rationale for holding the defendants liable to class plaintiffs. The Florida courts, most explicitly in Douglas III, then developed a rationale that the Walker panel and the Majority, correctly, albeit implicitly, recognize is unconstitutional. Yet, instead of simply refusing to apply the Florida courts' unconstitutional rationale, the Walker panel and now the Majority, develop their own rationale that is similarly sullied with constitutional errors.

If one lesson can be learned from this chaotic poker game it is that we should stick to our day jobs. Rather than act as advocates for the plaintiff, we should saddle him with the burden the law tasks him with carrying, and assess, impartially, whether the plaintiffs have established the elements of proving preclusion in the manner the law demands. On the record before us now, the plaintiff clearly has not, and the District Court's judgment should be reversed.

Appendix

WILSON, Circuit Judge, dissenting:

At its most fundamental level, the Due Process Clause guarantees an aggrieved party notice and "the opportunity to present his case and have its merits fairly judged." See Logan v. Zimmerman Brush Co., 455 U.S. 422, 428, 433, 102 S.Ct. 1148, 1153, 1156, 71 L.Ed.2d 265 (1982). The defendants have no doubt been provided notice and some degree of opportunity to be heard in court, but like Judge Tjoflat, I am not content that the use of the Engle jury's highly generalized findings in other forums meets "the minimum procedural requirements of the ... Due Process Clause in order to qualify for ... full faith and credit." See Kremer v. Chem. Constr. Corp., 456 U.S. 461, 481, 102 S.Ct. 1883, 1897-98, 72 L.Ed.2d 262 (1982). Because such a violation requires remand, I see no need to determine whether the Engle findings are preempted by federal law.