MEMORANDUM OPINION AND ORDER REGARDING PLAINTIFF'S MOTION FOR NEW TRIAL AND DEFENDANTS' MOTION FOR ATTORNEYS' FEES AND COSTS

MARK W. BENNETT, District Judge.

                                   TABLE OF CONTENTS

   I.  INTRODUCTION ................................................................1284
       A.  Procedural Background ...................................................1284
       B.  Factual Background ......................................................1285
           1.  Evolution of the jury instruction on "addiction" ....................1285
           2.  Evidence on "addiction" .............................................1287
           3.  Juror access to dictionary definitions of "addiction" ...............1290

  II.  LEGAL ANALYSIS ..............................................................1292
       A.  Starbuck's Motion For New Trial .........................................1292
           1.  Overview of grounds for a new trial .................................1292
           2.  Juror misconduct ....................................................1294
               a.  Arguments of the parties ........................................1294
               b.  The need for an evidentiary hearing .............................1295
               c.  The merits of the contention ....................................1297
                    i.  Applicable standards .......................................1297
                   ii.  Analysis ...................................................1301
           3.  Verdict against the great weight of th e evidence ...................1303
               a.  Arguments of the parties ........................................1303
               b.  Applicable standards ............................................1303
               c.  Analysis ........................................................1304
           4.  Improper jury instruction on "addiction" ............................1307
               a.  Arguments of the parties ........................................1307
               b.  Analysis ........................................................1307
           5.  Summary .............................................................1308
       B.  The Defendants' Motion For Attorneys' Fees And Costs ....................1308

 III.  CONCLUSION ..................................................................1309

APPENDIX: VERDICT FORM

Is the plaintiff smoker in this "Engle progeny" case1 against cigarette makers entitled to a new trial where the jury answered "no" to the initial question of whether he was addicted to cigarettes containing nicotine on or before November 21, 1996, thus ending their deliberations? The plaintiff contends that he is entitled to a new trial, because the jurors were exposed to extrinsic evidence, in the form of dictionary definitions of "addiction"; because the jury's verdict was against the great weight of the evidence; and because my addition of a timeframe for his addiction in the jury instructions heightened his burden and likely confused the jury. The defendant cigarette makers deny that the plaintiff is entitled to a new trial on any of these grounds. They also seek an award of attorneys' fees and costs as prevailing parties and pursuant to a Florida offer-of-judgment statute.

This jury trial, really a courtroom battle extraordinaire, was exceptionally hard fought by extremely talented, industrious, skilled, and zealous trial lawyers for the plaintiff and the two tobacco company defendants. Many hundreds, if not thousands, of objections were lodged, primarily by the two tobacco companies, throughout the trial. Double-digit mistrial motions were made by the defense during the trial, and motions and other filings by the defendants were made virtually around the clock, with voluminous filings every evening after mere mortals went to sleep. This is partially explained by the high stakes in the case, as well as the thousands of other tobacco cases pending in the federal and state courts of Florida, and by the plethora of legal issues not yet resolved by appellate courts.

I. INTRODUCTION

A. Procedural Background

I was designated as a visiting judge for the December 2014 retrial of this "Engle progeny case," after a mistrial was declared in the first trial, in the early summer of 2014, because the jury could not reach a verdict. At trial, plaintiff William Starbuck sought damages for his lung cancer from defendants R.J. Reynolds Tobacco Company (RJR), individually and as successor by merger to The Brown & Williamson Tobacco Corporation, and Philip Morris USA Inc. (PM USA). Starbuck asserted two "product liability" claims: "negligence" and "strict liability"; and two "fraud" claims: "fraudulent concealment" and "conspiracy to fraudulently conceal." RJR and PM USA denied Starbuck's claims and asserted, as a specific defense to his "product liability" claims, that Starbuck was at fault and, thus, responsible for his injury.

The retrial was scheduled for December 1, 2014, and was to be bifurcated. In Phase 1, the jurors would hear evidence and decide whether or not Starbuck had proved his claims for damages, and, if so, what compensatory damages, if any, to award him. Also, if Starbuck had proved one or both of his "fraud" claims, the jurors would also decide whether punitive damages were justified on those claims. If the jurors decided that punitive damages were justified, then, in Phase 2, the parties would present additional evidence, and the jurors would decide what amount of punitive damages, if any, to award Starbuck.

The trial began on December 1, 2014, with jury selection. See Trial Minutes (Day 1) (docket no. 179), and continued with the presentation of evidence on December 2, 3, 5, and 8-12, 2014. The case was submitted to the jury on December 15, 2014, and the jury returned a defense verdict the following day, December 16, 2014. Trial Minutes (Day 11) (docket no. 226). Specifically, the jurors answered "no" to the initial question of whether Starbuck was addicted to cigarettes containing nicotine on or before November 21, 1996, thus ending their deliberations. See Verdict Form (docket no. 228). A blank copy of the Verdict Form is attached to this decision as an appendix.

By Order (docket no. 255), filed January 22, 2015, I denied as moot the defendants' December 12, 2014, Motion For Judgment As A Matter Of Law On Plaintiff's Fraudulent Concealment And Conspiracy Claims (docket no. 210) and the defendants' December 12, 2014, Motion For Judgment As A Matter Of Law [On All Claims] (docket no. 211). In that Order, I also denied, on the merits, Starbuck's December 30, 2014, Motion To Permit Juror Interviews (docket no. 252). On February 10, 2015, I entered Judgment (docket no. 257) on the jury's verdict.

On February 24, 2015, the defendants filed their Motion For Attorneys' Fees And Costs (docket no. 258), which is one of the motions now before me. The defendants amended that request on February 25, 2015. See Notice (docket no. 259). On March 13, 2015, Starbuck filed his Opposition (docket no. 261) to the defendants' Motion For Attorneys' Fees And Costs. By Order (docket no. 264), filed April 8, 2015, I granted the defendants' Motion For Leave To File A Reply In Support Of Defendants' Motion For Attorneys' Fees And Costs (docket no. 262). The defendants filed their Reply (docket no. 265) on April 10, 2015. On March 10, 2015, Starbuck filed his Motion For New Trial (docket no. 260), which is the second motion now before me. On March 24, 2015, the defendants filed their Opposition (docket no. 263) to that motion. Starbuck did not request leave to file any reply in further support of his motion.

I do not find that oral arguments are necessary on either motion, in light of the parties' briefing and submission of supporting documents. Therefore, I will resolve both motions on the parties' written submissions.

B. Factual Background

This factual background focuses on three aspects of the trial pertinent to Starbuck's Motion For New Trial. The first aspect is the evolution of the part of the jury instructions concerning "addiction." The second aspect is the evidence presented at trial concerning "addiction" and whether or not Starbuck was "addicted" to cigarettes. The third aspect concerns the jury foreperson's investigation of dictionary definitions of "addiction" and my subsequent attempt to determine whether other jurors had also been exposed to those dictionary definitions.

1. Evolution of the jury instruction on "addiction"

Prior to trial, on November 18, 2014, see Order (docket no. 138), and again on November 25, 2014, see Order (docket no. 164), I provided the parties with my draft jury instructions for Phase 1 of the trial and, on November 29, 2014, see Order (docket no. 173), I provided the parties with what I believed to be the "final" version of those jury instructions. In an annotation to the November 18, 2014, draft of the Phase 1 jury instructions, I advised the parties that I do not give separate preliminary and final instructions. Instead, I give "front-end loaded" instructions, which meant that, subject only to the rare instance when "supplemental" instructions might be appropriate, all instructions — except for instructions on deliberations, and, in this case, Phase 2 instructions on punitive damages — would be given to the jurors before opening statements.

Also in the November 18, 2014, draft of my Proposed Jury Instructions, I took up the matter of the initial elements that Starbuck had to prove to establish his membership in the "Engle class" and, consequently, his entitlement to rely on certain common liability findings in the "Engle class litigation." More specifically, in its decision otherwise decertifying the class for determination of individual claims, the Florida Supreme Court concluded that "the appropriate cut-off date" for class membership is November 21, 1996. See Engle, 945 So.2d at 1255. Thus, the Florida Supreme Court defined the class as all Florida residents, and their survivors, who have suffered, presently suffer, or who have died from diseases and medical conditions caused by their addiction to cigarettes that contain nicotine as of November 21, 1996. Id. at 1274 (stating the trial court's class description as "All United States citizens and residents, and their survivors, who have suffered, presently suffer or who have died from diseases and medical conditions caused by their addiction to cigarettes that contain nicotine."); id. at (concluding, "The class consists of all Florida residents fitting the class description as of the trial court's order dated November 21, 1996."). An individual who proves membership in this class is entitled to the res judicata effect of certain findings by the jury in Phase I of the class trial. Id. at 1254-55, 1269, 1276-77. Those findings included that smoking cigarettes causes lung cancer, among other diseases, and that cigarettes that contain nicotine are addictive or dependence producing. See id. at 1277; see also Engle v. R.J. Reynolds Tobacco Co., No. 94-08273 CA-22 (Fl. 11th Cir.Ct.), Verdict Form For Phase 1 (questions 1 (generic causation) and 2 (addiction/dependence)).

In the November 18, 2014, Proposed Jury Instructions, I instructed that Starbuck must first prove that he "was addicted to cigarettes containing nicotine on or before November 21, 1996." 11/18/14 Proposed Jury Instructions (docket no. 138-8), 16 (Instruction No. 6 — Starbuck's "Eligibility" To Assert His Claim). In a pertinent annotation, I explained,

Id. at n. 36 (emphasis in the original).

I also included in the pertinent Instruction a definition of "addiction," id. at 16-17, and I explained my reason for doing so, in a pertinent annotation, as follows:

Id. at 17 n. 37.

After receiving the parties' comments on the November 18, 2015, Proposed Jury Instructions, I made the changes that I deemed appropriate and provided the parties with the November 25, 2014, revised version of the Proposed Jury Instructions. In an annotation to the November 25, 2014, Proposed Jury Instructions, I explained,

11/25/14 Revised Proposed Jury Instructions (docket no. 164-4), 13 and n. 8 (Instruction No. 6 — Starbuck's Initial Elements). My positions on these two issues did not change through subsequent revisions of the jury instructions.

Thus, the jury instructions ultimately read to the jurors did not contain any definition of "addiction," but did include a required timeframe for "addiction," that is, on or before November 21, 1996. See Instructions To The Jury (docket no. 221) (Phase 1 Instructions), 11 (Instruction No. 6 — Starbuck's Initial Elements).2

2. Evidence on "addiction"

"Addiction" was a hotly contested issue during the trial. Starbuck called, as one of his experts, Dr. Michael Cummings, Ph. D., a behavioral scientist and addiction expert, who, among other things, opined on the definition of "addiction" and how "addiction" works. Based on his years of treating patients, research, and review of documents and literature, Dr. Cummings opined that the reason that people persist in smoking despite the knowledge of harm is nicotine addiction. Trial Transcript, December 2, 2014 (Afternoon) (docket no. 232), 28:14-21. He opined that the definition of nicotine addiction was the same as the definition of addiction for other drugs, see id. at 36:2-7, that is, the definition of "addiction" published by the National Institute on Drug Abuse (NIDA). According to Dr. Cummings, the NIDA definition is the following:

Trial Transcript, December 2, 2014 (Afternoon), 37:2-10. Dr. Cummings opined that this was the definition used in "modern science," in his practice, and in the NIDA textbook on drug abuse for health professionals. Id. at 39:11-23.

Dr. Cummings also opined, "[A]ddiction is not free will. Addiction is a constrained choice." Id. at 35:5-6. He explained that nicotine addiction is a "brain disease," because the receptors in the brain and the connections in the brain, which he described as "neuroplasticity," are altered by nicotine, and that the younger a person starts smoking, the more their neuroplasticity changes and the harder it is for them to stop smoking. Id. at 37:15-39:10. Furthermore, he explained how nicotine is absorbed into the body by smoking, the physiological mechanism that makes it both enjoyable and addictive, the symptoms and effects of withdrawal, that Starbuck's statements that he enjoyed smoking were "an indication probably he was addicted," and that Starbuck's negative symptoms when he quit smoking were consistent with him having nicotine withdrawal. Id. at 38:20-38:12, 39:19-10, 41:2-42:18, 43:1-46:19. Dr. Cummings also explained how nicotine enters the bloodstream and how quickly a smoker will feel the effects. Id. at 47:9-49:7.

Finally, Dr. Cummings discussed factors that are helpful in determining whether and to what degree an individual is addicted to nicotine, and he applied those factors to Starbuck. Id. at 49:11-74:8. More specifically, Dr. Cummings applied a "Heaviness of Smoking Index" to Starbuck's smoking. He found that Starbuck smoked first thing when he got up in the morning. Id. at 54:9-15. He found that Starbuck started smoking a pack a day, increased to about a pack and a half a day when he was in the Air Force, increased to a pack and a half to two packs a day when he moved to Florida, and was smoking two to two-and-a-half packs and sometimes 3 packs a day in the 1980s. Id. at 54:23-55:10. He concluded that Starbuck's scores using the Index placed Starbuck "in the high dependence category." Id. at 55:17-19. Dr. Cummings then looked at the following seven criteria to assess whether or not Starbuck was an "addicted" smoker: "whether the smoker is smoking on a persistent, daily basis"; the "amount smoked"; "time to first cigarette"; "age of initiation"; "difficulty controlling their smoking"; "[e]vidence of withdrawal"; "continued use, despite knowledge of harm." Id. at 55:20-56:15. He applied those criteria to Starbuck, finding that Starbuck's "age of initiation" was early, around 13 or 14, when his brain was still developing. Id. at 59:11-23. He also concluded that Starbuck was "a persistent daily user over decades, four decades"; that "[h]e smoked first thing in the morning"; he smoked "anywhere from one to three packs" per day, "[b]ut in the 1980s, he was up over two packs per day"; he had "difficulty refraining from smoking" and had "withdrawal symptoms"; and "he had some pretty serious health problems, including pneumonia and some other health problems, and yet he persisted in smoking despite, you know, those harms," so that "he met every one of these criteria in my assessment." Id. at 60:14-61:10. Ultimately, Dr. Cummings concluded that Starbuck "was addicted to nicotine in the cigarettes that he smoked, and I would rate him as "heavily addicted." Id. at 61:24-25.

Starbuck also called, as another expert witness, Dr. David Burns, who has worked on every United States Surgeon General's Report on Smoking and Health since 1975, to provide medical opinions on addiction. Starbuck points out that Dr. Burns testified that the public health community, and, in particular, physicians who diagnose and treat nicotine addiction, rely on the NIDA definition of "addiction" and use certain related criteria to assess nicotine addiction. See Trial Transcript, December 10, 2014 (Morning) (docket no. 241), 28:8-13; 33:21-36:25. On the other hand, Dr. Burns testified that the Diagnostic and Statistical Manual of Mental Disorders (DSM), either DSM-IV or DSM-V, is a manual for psychiatrists to classify behaviors, but not a manual providing definitions of addiction used by medical professionals for the purposes of diagnosing and treating addiction. Id. at 32:11-33:25. On cross-examination, Dr. Burns admitted that not all smokers are addicted and that even some daily smokers are not addicted. See id. at 129:22-130:2.

Starbuck also presented testimony by Dr. Jeffrey Gentry of RJR and Richard Jupe of PM USA, both of whom acknowledged that, over time, their companies had come to admit that cigarettes are addictive. Trial Transcript, December 11, 2014 (docket no. 243), 229:10-13; Trial Transcript, December 12, 2014 (Afternoon) (docket no. 246), 8:13-14, 22:15-18.

In addition to expert testimony, Starbuck presented his own testimony at trial concerning his cigarette smoking. Starbuck testified that he started smoking as an adolescent, at around age 13 or 14, see, e.g., Trial Transcript, December 5, 2014 (Afternoon) (docket no. 236), 43:21-25, which Dr. Cummings had explained was when Starbuck was particularly susceptible to nicotine's addictive effects. Trial Transcript, December 2, 2014 (Afternoon) (docket no. 232), 59:11-23. Starbuck became a regular smoker at age 14. See, e.g., Trial Transcript, December 5, 2014 (Afternoon), 43:21-25. Starbuck smoked thirty to forty cigarettes per day for over 40 years, smoking from when he woke up, throughout his waking hours while working, with breaks of no more than 20 to 30 minutes, until he went to bed at night. Id. at 60:14-61:22. The work in question was Starbuck's work as a "glass artist," making "glass sculpture of figurines" from "solid glass" heated over a flame, which is different from "glass blowing." Id. at 27:12-28:9. He also smoked when he woke up in the middle of the night. Id. at 65:8-15. Also, for years, Starbuck suffered withdrawal symptoms whenever he tried to quit. Id. at 66:19-22, 76:2-77:24. As to his efforts to quit smoking, Starbuck testified, "It was the hardest thing I probably — well, not harder than having cancer, but one of the hardest things I've ever had to do." Id. at 77:25-78:6. His inability to quit caused significant disruptions in his home life, resulting in fights with his wife and conflicts with his stepdaughter. Id. at 66:1-67:10. Despite trying numerous cessation aids, including patches, gum, candy, and nicotine gum, Starbuck was unable to quit smoking until doctors diagnosed him with lung cancer, removed a portion of his lung, and kept him in an induced coma for three weeks because of complications from his surgery. Id. at 74:1-78, 84:9-14.

All of this evidence was consistent with the jury's finding in Phase I of the "Engle class litigation" that smoking cigarettes causes lung cancer, among other diseases, and that cigarettes that contain nicotine are addictive or dependence producing. See Engle, 945 So.2d at 1277; see also Engle v. R.J. Reynolds Tobacco Co., No. 94-08273 CA-22 (Fl. 11th Cir.Ct.), Verdict Form For Phase 1 (questions 1 (generic causation) and 2 (addiction/dependence)).

The defendants point to specific portions of Starbuck's testimony as supporting their arguments against a new trial. First, the defendants point to Starbuck's testimony that he enjoyed smoking, the feeling it gave him, the taste, and being a part of a group of smokers. Trial Transcript, December 8, 2014 (Afternoon) (docket no. 238), 28:7-15. They also point to his testimony that smoking cigarettes helped him deal with stress and improved his mood. Id. at 28:16-19. The defendants cite Starbuck's testimony that, after his cancer surgery in 1995, he tried a cigarette, because he was curious, but he found that it tasted awful, that he got no euphoria from it, and that he has not had a cigarette since, that is, for nearly 20 years. Id. at 28:20-29:7. Finally, they cite testimony of Starbuck's family members about Starbuck's ability to refrain from smoking. Specifically, they point to testimony of Starbuck's stepdaughter that Starbuck was able to refrain from smoking while he was making glass items, for as much as an hour at a time, Trial Transcript, December 12, 2014 (Afternoon) (docket no. 245), 150:1-18. They also point to testimony of Starbuck's brother that Starbuck did not smoke while he was actually working glass, possibly for as long as two hours, and could go without smoking for as long as eight hours, when he was asleep. Id. at 148:20-149:13.

In response to Starbuck's expert testimony, the defendants presented the expert testimony of Dr. Christopher Ticknor, a board certified psychiatrist, based on his review of the evidence and his medical examination of Starbuck. Dr. Ticknor testified that the DSM-V, developed by the American Psychiatric Association, does not use the term "addiction," because of its negative connotation and the lack of an agreed-upon definition, but that "addiction" would be a "pretty extreme presentation" of "a tobacco use disorder," as defined in the DSM-V. Trial Transcript, December 12, 2014 (Afternoon), 65:23-66:9. Dr. Ticknor opined that, in his office practice, "well below 50 percent" of people met the criteria for tobacco use disorder and, in fact, "sometimes it's [only] as high as 20 or 30 percent of the smokers in my practice," while "probably closer to 10 percent or 20 percent" would fit a "description of truly an addiction." See id. at 66:22-67:5.

Dr. Ticknor also testified that, using the diagnostic criteria in the DSM-V, his review of the relevant literature, his review of the deposition testimony of Starbuck and members of his family, and his examination of Starbuck, it was his opinion that Starbuck "did not meet the criteria for a tobacco use disorder, that he was not addicted to nicotine." Id. at 67:6-24. Dr. Ticknor also testified that, in his opinion, "most people can quit smoking when they are motivated and determined to do so," although there were some who could not, even if motivated to do so. He identified examples of people who could not quit as people who were "mentally retarded," "severely mentally ill," "psychotic," "schizophrenic," "dependent," "emotionally fragile," or who had been "physically and sexually traumatized." Id. at 72:12-73:10. He also testified that, in his view, only about a third of the people who came to see him "really do want to quit, and they are motivated to do so," while another third are "ambivalent," and the remaining third "really have no desire to quit," because "for one reason or another — they like smoking." Id. at 73:13-74:3. On cross-examination, however, Dr. Ticknor testified that he understood that medical doctors — at least those not practicing psychiatric medicine — do not use the DSM-V for their determinations of nicotine dependence, that they use the NIDA definition, and that there is "nothing wrong" with using the NIDA definition of "addiction." Id. at 93:21-94:6; see also id. at 94:12-15 (stating, "[T]here are differences about what the word addiction means depending on your organization [and] I personally believe that the NIDA reference to addiction is a very good one," even if he did not fully agree with all of the elements of the NIDA definition).

3. Juror access to dictionary definitions of "addiction"

As mentioned, above, I read all of the Phase 1 jury instructions, except the last one on deliberations, to the jury before opening statements. The jurors each received a complete copy of the jury instructions, including the Verdict Form, just before I read them, so that they could follow along, and they were allowed to keep those copies in the courtroom or in the jury room throughout the trial. The jury instructions read to the jurors at the beginning of the trial included instructions that "[t]he law demands that you return a just verdict, based solely on the evidence, your individual evaluation of that evidence, your reason and common sense, and these Instructions," Instructions To The Jury (docket no. 221) (Phase 1 Instructions), 1 (Instruction No. 1 — Introduction); that "[e]vidence is not ... [a]nything that you see or hear about this case outside the courtroom," id. at 5 (Instruction No. 3 — Definition of Evidence) (emphasis in the original); and, more specifically, "[d]o not do any research — on the Internet, in libraries, in the newspapers, in dictionaries or other reference books, or in any other way — or make any investigation about this case, the law, or the people involved on your own," id. at 39 (Instruction No. 18 — Conduct Of The Jurors During Trial) (emphasis added).

The parties made their closing arguments on December 15, 2014, then I read the remaining instruction on "deliberations." That instruction included the following reminder: "Base your verdict solely on the evidence and on the law as I have given it to you in my Instructions." Id. at 42 (Instruction No. 19 — Deliberations). The case was submitted to the jury at 2:25 p.m., and the jurors began their deliberations. Approximately an hour later, the jurors adjourned their deliberations for the day.

Notwithstanding the various cautions in the jury instructions quoted above, on the morning of December 16, 2014, about half an hour before the jurors had indicated that they would resume their deliberations, I received a note from the foreperson of the jury asking the following question:

Juror Question (docket no. 224), 2 (formatting as found in the juror's note). After consulting the parties, I provided the jurors with a response to that note at about 9:32 a.m. That response stated, in pertinent part,

Response To Question From Jury (docket no. 224) at 1.3

At approximately 1:20 p.m., the jury returned to the courtroom and rendered a defense verdict. Trial Minutes (Day 11) (docket no. 226). Specifically, the jurors answered "no" to the initial question on the Verdict Form asking whether Starbuck was addicted to cigarettes containing nicotine on or before November 21, 1996, thus ending their deliberations. See Verdict Form (docket no. 228), blank copy attached as an Appendix to this ruling.

I excused the jurors to the jury room, before releasing them, so that I could consult with the parties about the issue of the possibility that the jury foreperson had looked up one or more dictionary definitions of "addiction," in light of the first question from the jury. After discussions with the parties, I interviewed the jury foreperson, individually, in open court. I read the foreperson the note quoted above, then had the following colloquy with him:

Trial Transcript, December 16, 2014 (docket no. 247), 38:21-22, 39:1-21. I then excused the foreperson back to the jury room.

Following my colloquy with the jury foreperson, I had an extended discussion with the parties concerning whether any further inquiry was appropriate, recognizing that I could address Starbuck's challenge to the verdict based on the foreperson's conduct on post-trial motions. I determined, over vociferous and prolonged objections by defendants, including that I was violating the rights of the jurors by detaining them after they had rendered a verdict, that a very limited further inquiry was permissible. Therefore, I asked each of the other jurors, separately, in open court, the following question or a slight variation of the question, "Did any juror describe a dictionary definition of addiction to the jury during deliberations?" Id. at 55:10-11, 56:1-2, 56:16-17, 57:7-8, 57:17-18; see also id. at 58:4-5 ("Did any juror describe a dictionary definition of addiction to you during deliberations?"); id. at 58:19-21 ("Did any juror describe a dictionary definition of addiction to you — sorry, to the jury during deliberations?"). The first juror answered, "I don't remember that," id. at 55:12, and each of the other jurors answered "No," id. at 56:3, 57:9, 57:19, 58:6, or "No, sir," id. at 58:17, or "No, sir. They did not," id. at 56:18. I excused the jury at approximately 2:07 p.m. without making any further inquiries.

II. LEGAL ANALYSIS

The disposition of Starbuck's Motion For New Trial may affect the disposition of the defendants' Motion For Attorneys' Fees And Costs. Therefore, I will begin my legal analysis with Starbuck's Motion, even though it was filed after the defendants' Motion.

A. Starbuck's Motion For New Trial

As mentioned at the outset of this opinion, Starbuck seeks a new trial because the jurors were exposed to extrinsic evidence, in the form of dictionary definitions of "addiction"; because the jury's verdict on "addiction" was against the great weight of the evidence; and because my addition of a timeframe for his addiction in the jury instructions heightened his burden and likely confused the jury. I will consider these three grounds in turn. First, however, I will briefly consider the authority of the district court to grant a motion for a new trial.

1. Overview of grounds for a new trial

Rule 59 of the Federal Rules of Civil Procedure expressly authorizes a new trial, on a party's motion, after a jury trial. FED.R.CIV.P. 59(a)(1)(A). The rule is less helpful on the grounds for doing so, however: The rule provides only that such a motion may be granted "for any reason for which a new trial has heretofore been granted in an action at law in federal court." Id. The Eleventh Circuit Court of Appeals has clarified matters, at least to the extent that it has recognized that a new trial may be appropriate on each of the grounds asserted by Starbuck. See, e.g., BankAtlantic v. Blythe Eastman Paine Webber, Inc., 955 F.2d 1467, 1471-72 (11th Cir.1992) (explaining the circumstances in which a juror's misconduct in considering extrinsic evidence requires a new trial (citing United States v. Rowe, 906 F.2d 654, 656 (11th Cir.1990))); Lamonica v. Safe Hurricane Shutters, Inc., 711 F.3d 1299, 1312-13 (11th Cir.2013) (explaining that a new trial may be granted if "the verdict is against the great — not merely the greater — weight of the evidence" (internal quotation marks and citations omitted)); Gowski v. Peake, 682 F.3d 1299, 1310 (11th Cir.2012) ("Motions for new trial on the basis of erroneous and prejudicial jury instructions are within the district court's discretion."). The specific standards for granting a new trial vary, depending upon the ground asserted, so I will address the standards for a new trial applicable to each of Starbuck's grounds in more detail, when I consider those claims in turn. Nevertheless, those specific standards seem to me to be variations on a theme that a new trial should not be granted, unless the error or circumstance at issue affected substantial rights or caused substantial prejudice, so that it was not merely harmless. See, e.g., Coquina Inv. v. TD Bank, N.A., 760 F.3d 1300, 1309 (11th Cir.2014) (stating this standard in reference to allegedly erroneous evidentiary rulings).

A district court's decision on a motion for a new trial is reviewed for abuse of discretion. Compare Finnerty v. Stiefel Labs., Inc., 756 F.3d 1310, 1322 (11th Cir. 2014) ("We review a district court's denial of a motion for new trial only for an abuse of discretion." (quoting Myers v. TooJay's Mgmt. Corp., 640 F.3d 1278, 1287 (11th Cir.2011))); with Aronowitz v. Health-Chem. Corp., 513 F.3d 1229, 1242 (11th Cir.2008) (stating that the appellate court reviews a district court's grant of a motion for new trial for "abuse of discretion"). Yet, in the Eleventh Circuit, the "abuse of discretion" standard varies in the degree of deference paid to the district court's resolution depending on the circumstances. Thus, the appellate court has stated that "deferen[tial]" review "for an abuse of discretion... is particularly appropriate where a new trial is denied and the jury's verdict is left undisturbed." St. Luke's Cataract and Laser Inst., P.A. v. Sanderson, 573 F.3d 1186, 1200 n. 16 (11th Cir. 2009) (quoting Rosenfield v. Wellington Leisure Prods., Inc., 827 F.2d 1493, 1498 (11th Cir.1987)). On the other hand, "[t]he range of discretion afforded to the district court is smaller ... when ... the district court orders a new trial because the jury verdict was contrary to the great weight of the evidence." Hardin v. Hayes, 52 F.3d 934, 938 (11th Cir.1995). As the court explained,

Hardin, 52 F.3d at 938 (footnote omitted); accord Aronowitz, 513 F.3d at 1242 ("Our review for abuse of discretion is `more rigorous when the basis' of the grant was the weight of the evidence." (quoting Williams v. City of Valdosta, 689 F.2d 964, 974 (11th Cir.1982))). Apparently, there is some middle level of deference, as well: While the district court's discretion is smaller to grant a motion for a new trial because the verdict was "contrary to the great weight of the evidence" than it is to deny a motion for new trial, "[a] more deferential standard of review is appropriate... if the district court's new trial order is precipitated by jury misconduct or other prejudicial trial events that contaminate the jury's deliberative process." Id. at 938 & n. 6.

Keeping in mind the varying degrees of deference that may be applicable to my ruling, depending on the issues presented and how I resolve them, I turn to specific consideration of each of the grounds on which Starbuck seeks a new trial.

2. Juror misconduct

Starbuck argues, first, that the jury's exposure to extrinsic evidence, in the form of the foreperson's consultation of dictionary definitions of "addiction," posed a reasonable probability of prejudice, such that he is entitled to a new trial. The defendants oppose a new trial on this ground.

a. Arguments of the parties

Starbuck argues that the meaning of "addiction" was not merely collateral, but highly material, because proof that he was "addicted" was a prerequisite to "Engle class membership." He also argues that the extrinsic influence, here, went straight to the heart of the trial and "appears" to have improperly influenced the jury's verdict that he was not addicted. Starbuck argues that the law presumes prejudice when a jury has come into contact with extraneous information, such as a dictionary definition, and that, unless the defendants can demonstrate that the jurors' consideration of that extrinsic information was harmless, a new trial is required. He argues that the defendants cannot show that the extrinsic evidence was harmless in this case, based on consideration of several pertinent factors. First, he argues that the meaning of "addiction" was undoubtedly important to resolution of the case. Next, he argues that the dictionary definition provided the only "neutral" definition of the contested term, so that the dictionary definition carried undue weight with the jurors. Starbuck also argues that he is entitled to a new trial if even a single juror's impartiality was overcome by an extraneous influence, so that, even if the foreperson did not share the specific dictionary definitions that he had looked up, a new trial is still warranted. Starbuck also argues that, here, the other jurors were aware that the foreperson had extrinsic evidence of the meaning of the key term, which may have influenced them to follow the foreperson's lead. Finally, Starbuck argues that the evidence did not strongly support a verdict in the defendants' favor on whether or not he was "addicted." Thus, he argues that these factors weigh in favor of granting his motion for new trial. In the alternative, Starbuck argues that I should now hold an evidentiary hearing to measure the level of prejudice resulting from the foreperson's consultation of extrinsic sources for a definition of "addiction."

In response, the defendants argue that I took prompt and proper actions to ensure that there was no prejudice to Starbuck. They argue that this is so, because I informed the jurors that they could not consider dictionary definitions of "addiction," the foreperson then destroyed the copies of the dictionary definitions that he had found, and the other jurors all stated on the record that no one had described a dictionary definition to them. The defendants contend that prejudice is not presumed in these circumstances, but, even if a rebuttable presumption of prejudice arises, no new trial is justified in this case. The defendants reiterate that the record shows that the dictionary definitions had no effect on any of the jurors, even the foreperson, where the other jurors did not hear any such definitions, and the foreperson ripped up his copies of the dictionary definitions before proceeding with deliberations. The defendants also point out that there is no evidence that the jurors were deadlocked prior to the foreperson's consultation of dictionary definitions, such that the extrinsic information might have effectively pushed the jurors' verdict one way or the other. The defendants also contend that no evidentiary hearing is required, because I have previously denied Starbuck's request for nearly identical relief — the opportunity to interview jurors — and I have already made the only proper inquiries. The defendants point out that Starbuck does not now offer any more evidence that the foreperson's conduct prejudiced him than Starbuck did when he sought leave to interview jurors.

b. The need for an evidentiary hearing

Although it is Starbuck's "alternative" argument, I will consider, first, his argument that I should conduct an evidentiary hearing to determine whether any prejudice arose from the foreperson's consultation of dictionary definitions of "addiction." The Eleventh Circuit Court of Appeals recently observed, albeit in a criminal case,

United States v. Alexander, 782 F.3d 1251, 1258 (11th Cir.2015).

Here, there is certainly "a colorable showing of extrinsic influence," because the foreperson admitted looking up dictionary definitions of "addiction" and "addict." Id. Even in these circumstances, however, an evidentiary hearing is not necessarily required, because the Eleventh Circuit Court of Appeals has stated that the required responses by the district court encompass alternatives, that is, that the district court must "make sufficient inquiries or conduct a hearing to determine whether the influence was prejudicial." Id. (emphasis added) (quoting Barshov, 733 F.2d at 851). Here, I have already made "sufficient inquiries," because, before releasing the jurors, I interviewed the foreperson and each of the other jurors, individually, to determine whether they had been exposed to an extrinsic influence, that is, dictionary definitions of "addiction." Id. Indeed, United States v. Gabay, 923 F.2d 1536 (11th Cir. 1991), one of the authorities on which Starbuck relies for his contention that I should conduct an evidentiary hearing, did not involve such an evidentiary hearing in response to a new trial motion based on juror misconduct. Rather, that case involved the district court's immediate investigation of juror misconduct, when that misconduct came to the district court's attention during deliberations. 923 F.2d at 1542-43.

Starbuck contends that my inquiry just after the jurors reached their verdict was inadequate, however. Starbuck contends that an adequate inquiry would result in detailed factual findings about what the foreperson said to his fellow jurors about his research into the definition of "addiction"; to whom he spoke; how other jurors responded, and which ones; how many and which jurors were present at the time of this exchange; how many and which jurors overheard his remarks; whether these comments occurred in the context of deliberations; how many jurors knew throughout the remainder of deliberations that their foreperson had performed outside research on the question of addiction; and any other related topics. As I explained in the part of my January 22, 2015, Order (docket no. 255), denying Starbuck's December 30, 2014, Motion To Permit Juror Interviews (docket no. 252),

Order (docket no. 255), 3. Thus, I concluded, and I now reiterate my conclusion, that I already made the only permissible inquiry under Rule 606(b).

Starbuck points out that I conducted a much broader inquiry concerning juror misconduct in United States v. Honken, 381 F.Supp.2d 936 (N.D.Iowa 2005), a federal death-penalty case. In that case, I examined applicable case law and Rule 606(b), and noted, inter alia, that there was "tension" between Rule 606(b) and the need to investigate alleged juror misconduct. 381 F.Supp.2d at 1039-46. In that case, I took the following steps:

Honken, 381 F.Supp.2d at 1041.

In Starbuck's case, however, I only questioned the jury foreperson to determine the nature of his contact with extrinsic information, that is, the dictionary definitions of "addiction," and then questioned the other jurors about whether anyone had described dictionary definitions of "addiction" to them. In Honken, one juror had conveyed extrinsic information or had described an improper contact to other jurors, but there was no need to make further inquiries in Starbuck's case, where none of the other jurors had any contact with the extrinsic information. In other words, there was no need to ask about the other jurors' ability to be fair and impartial, despite extrinsic information, because none of the other jurors had that extrinsic information. Also, my answer to the foreperson's note about use of dictionaries, explaining that the jurors could not use such definitions, was an effective curative procedure — at least to the extent of preventing other jurors from learning the dictionary definitions and causing the foreperson to destroy the copies of the definitions that he had discovered. This is so, because the other jurors' statements confirmed that no one had described any dictionary definitions to them. Thus, the circumstances at the time of my inquiry did not warrant further inquiry, and Starbuck has not demonstrated any basis, other than speculation, for believing that further inquiries are permissible or appropriate. See Alexander, 782 F.3d at 1258 (explaining that "there must be something more than mere speculation" to warrant inquiry into alleged juror misconduct (internal quotation marks and citations omitted)).

Starbuck's request for an evidentiary hearing on his juror misconduct/extrinsic information ground for a new trial is denied.

c. The merits of the contention

i. Applicable standards

The Eleventh Circuit Court of Appeals has observed, "In a criminal case, any private communication, contact, or tampering directly or indirectly, with a juror during a trial about the matter pending before the jury is ... deemed presumptively prejudicial." Alexander, 782 F.3d at 1258 (emphasis added) (quoting Remmer v. United States, 347 U.S. 227, 229, 74 S.Ct. 450, 98 L.Ed. 654 (1954)). Although Starbuck contends that this same presumption of prejudice applies in a civil case, Eleventh Circuit case law is to the contrary. Rather, as I noted, above, the Eleventh Circuit Court of Appeals explained in BankAtlantic v. Blythe Eastman Paine Webber, Inc., 955 F.2d 1467 (11th Cir.1992), the circumstances in which a juror's misconduct in considering extrinsic evidence in a civil case requires a new trial pursuant to Rule 59(b). 955 F.2d at 1471-72 (citing United States v. Rowe, 906 F.2d 654, 656 (11th Cir.1990)). Specifically, the court explained, "A juror's consideration of extrinsic evidence requires a new trial if the evidence poses a reasonable possibility of prejudice to the [aggrieved party]." BankAtlantic, 955 F.2d at 1471-72 (emphasis in the original) (quoting Rowe, 906 F.2d at 656). Still more specifically,

BankAtlantic, 955 F.2d at 1472. Because, in a civil case, the aggrieved party must first establish prejudice from alleged juror misconduct, involving contact with extrinsic information, by a preponderance of credible evidence, there is clearly no presumption of prejudice.

In BankAtlantic, a savings in loan, which had lost at jury trial on its claims against a financial advisor, sought a new trial, inter alia, on the ground that that "jury foreman Anthony Lippert had read, in violation of an express court order, an article that contain[ed] extraneous information about BankAtlantic and its chairman, Alan Levan." 955 F.2d at 1471. The court explained further,

BankAtlantic, 955 F.2d at 1471.

In BankAtlantic, the court considered one of its prior cases, United States v. Bolinger, 837 F.2d 436 (11th Cir.), cert. denied, 486 U.S. 1009, 108 S.Ct. 1737, 100 L.Ed.2d 200 (1988), in which a new trial had been sought on the basis of a jury's exposure to a newspaper article, and a Fifth Circuit case, United States v. Williams, 568 F.2d 464 (5th Cir.1978), in which several jurors had been exposed to a television news report that the criminal defendant in their case had previously been convicted of the same offenses. BankAtlantic, 955 F.2d at 1472. The court then resolved the question of whether a new trial was required in the case before it, on the ground that the jury foreperson had read a newspaper article, as follows:

BankAtlantic, 955 F.2d at 1472-73.

I will consider out-of-circuit authorities for guidance, as well. The only decision involving review by a federal court of appeals of a district court's ruling on a new trial motion in a civil case involving juror exposure to extrinsic dictionary definitions that the parties have cited is Mayhue v. St. Francis Hospital of Wichita, Inc., 969 F.2d 919 (10th Cir.1992).4 Mayhue was a "civil rights action," in which the Tenth Circuit Court of Appeals "examine[d] whether the district court's decision to grant the defendant's motion for a new trial because of the jury's unauthorized use of dictionary definitions during its deliberations constitutes reversible error." 969 F.2d at 920. The district court in that case had denied the jurors' request for a dictionary during their deliberations. After the jury returned its verdict, however, "the court's staff found a handwritten note in the jury room that contained definitions of the words `discriminate,' `p[re]judice,' `administer,' `clinical,' and `hypertension.'" Id. at 921.

"In accordance with Federal Rule of Evidence 606(b)," the district court in Mayhue held an evidentiary hearing to determine the effect of two of the dictionary definitions in the note, "p[re]judice" and "discriminate." Id. The district court then granted the defendant's motion for a new trial for the following reasons:

Mayhue, 969 F.2d at 921 (footnotes omitted).

In its review in Mayhue, the Tenth Circuit Court of Appeals stated, "The law in the Tenth Circuit is clear. A rebuttable presumption of prejudice arises whenever a jury is exposed to external information in contravention of a district court's instructions." 969 F.2d at 922. As I observed, above, however, the law in the Eleventh Circuit does not clearly require such a presumption. See, supra, beginning at page 1297. What is of greater interest, here, is that the court in Mayhue listed the following considerations as "relevant in an analysis of whether the presumption of prejudice ought to be rebutted when a jury consults a dictionary or dictionary definition without authorization":

Mayhue, 969 F.2d at 924. I find that these considerations are equally relevant to the ultimate determination of whether or not a juror's unauthorized consultation of a dictionary definition was prejudicial to the party moving for a new trial. See also Chums, Ltd. v. Snugz/USA, Inc., 64 F.3d 669 (10th Cir.1995) (unpublished op.) (concluding that the presumption of prejudice from a juror's unauthorized consultation of a dictionary had been rebutted, and the district court had properly denied a motion for a mistrial, based on considerations similar to those identified in Mayhue, where the district court had found "that (1) only one juror had seen the dictionary; (2) the dictionary was confiscated by a marshal; (3) the jury was re-instructed not to use or refer to extraneous sources of information; (4) the dictionary definition at issue comported with common-sense meaning; and (5) Chums' counsel argued only that the jury should be re-instructed to rely on ordinary and common meaning.").

In Mayhue, these considerations led the appellate court to affirm the district court's conclusion that the jury's unauthorized consultation of a dictionary had been sufficiently prejudicial to warrant a new trial. 969 F.2d at 924-26. Specifically, the court concluded, first, that "the meanings of `discriminate' and `prejudice' [we]re of crucial importance to the resolution of this section 1981 action." Id. at 924. "Second," the court concluded, "the dictionary definitions of `discriminate' and `prejudice' did not accurately reflect applicable law as set forth in the jury instructions." Id. at 925. "Third, the fact that the foreperson obtained and read the definitions might have caused those jurors who heard her to give the definitions undue emphasis at the expense of the jury instructions." Id. As to this consideration, the court explained, further, "the inflection in the foreperson's voice as she read the definitions and the discussion preceding their reading might have caused the jurors to place unmerited weight upon the definitions." Id. at 926. "Fourth, the jury had difficulty reaching a verdict prior to the introduction of the dictionary definition," and, "[f]inally, prejudice may be inferred from the timing of the verdict," because "[t]he jury was able to reach a verdict less than three hours after the foreperson read the definitions, despite having been plagued by `irreconcilable differences' the night before." Id.

ii. Analysis

There is no doubt that dictionary definitions of "addiction" and "addict" were "extraneous" or "extrinsic" information, because they plainly derived from sources — the dictionaries — "external" to the jury. See Warger, ___ U.S. at ___, 135 S.Ct. at 529. There is also no doubt that consultation of those dictionary definitions by the jury foreperson was unauthorized. Indeed, it was contrary to an explicit jury instruction, given to the jurors at the beginning of trial, that stated, "Do not do any research — on the Internet, in libraries, in the newspapers, in dictionaries or other reference books, or in any other way — or make any investigation about this case, the law, or the people involved on your own." Instructions To The Jury (docket no. 221) (Phase 1 Instructions), 39 (Instruction No. 18 — Conduct Of The Jurors During Trial) (emphasis added).

Nevertheless, while not free from doubt, the decision of the Eleventh Circuit Court of Appeals in BankAtlantic counsels denial of Starbuck's Motion For New Trial on the basis of the juror misconduct at issue, here. As in BankAtlantic, only one juror, here — the jury foreperson — was exposed to the extrinsic definitions of "addiction" and "addict," and none of the other jurors had any knowledge of the contents of those definitions. Cf. 955 F.2d at 1472. Also, notwithstanding that the meaning of "addiction" was also crucial in this case, it cannot be said that any dictionary definition of that term would have been more inflammatory than the contested definitions or other evidence presented by the parties in this case. Cf. id. It is true that I declined to ask any question of the foreperson about whether reading the dictionary definitions had any effect on his impartiality. Cf. id. Nevertheless, I find that the foreperson made a conscientious effort to put those definitions out of his mind, after being reminded, both by another juror and by me, that he could not consider dictionary definitions, because he destroyed his copies of the definitions without exposing any other jurors to them. Clearly, the dictionary definitions themselves had no effect on the impartiality of the other jurors, because the other jurors were not exposed to them. Again, I reject Starbuck's invitation to speculate that either the foreperson or the other jurors were affected by the dictionary definitions or the foreperson's investigation of such definitions. Cf. Alexander, 782 F.3d at 1258 (explaining that "there must be something more than mere speculation" to warrant inquiry into alleged juror misconduct (internal quotation marks and citations omitted)).

Contrary to Starbuck's contentions, consideration of the "Mayhue factors," postulated by the Tenth Circuit Court of Appeals, also counsels denial of Starbuck's Motion For New Trial on this ground. See 969 F.2d at 924. I do not disagree with Starbuck's contention that the definition of "addiction" was important — indeed, "crucial" — in this case. Cf. id. (finding that the dictionary definitions related to "crucial" terms in the case). Here, whether or not Starbuck was "addicted" to cigarettes containing nicotine was the threshold question, determining whether Starbuck was a member of the "Engle class" and entitled to rely on the "Engle class findings," and whether the jurors needed to proceed further in their deliberations. See id. (stating the first consideration as "[t]he importance of the word or phrase being defined to the resolution of the case," and concluding that, in that case, the dictionary definitions that a juror had improperly consulted were "crucial" to resolution of the case). That factor, standing alone, does not warrant a new trial, however, where other factors weigh against such relief.

As to the second "Mayhue factor" — "[t]he extent to which the dictionary definition differs from the jury instructions or from the proper legal definition," id. — the parties have not offered the precise content of the dictionary definitions that the jury foreperson consulted, so that I cannot tell precisely how those definitions might have differed from the parties' disputed definitions. Compare id. at 925 (concluding that the dictionary definitions "did not accurately reflect applicable law as set forth in the jury instructions"). Nevertheless, in my view, because the parties both urged me not to give a jury instruction on addiction, I find this factor impossible to apply and, therefore, neutral.

The third "Mayhue factor" is "[t]he extent to which the jury discussed and emphasized the definition." Id. at 924. In Mayhue, this factor carried considerable weight on the side of granting a new trial, because the jury foreperson had not only obtained the unauthorized dictionary definitions, but had read the definitions to the jurors. Id. at 925. In Starbuck's case, in contrast, the jury foreperson obtained the unauthorized dictionary definitions, but kept them to himself, did not describe them or read them to the other jurors, and destroyed them when the jurors received my response to his question in which I explained that the jurors could not use dictionary definitions. Thus, this case does not involve the possible influence of "the inflection of the foreperson's voice as [he] read the definitions." Compare id. at 926. There is also no evidence in Starbuck's case of any discussions about the meaning of "addiction" before the foreperson obtained the dictionary definitions that might indicate that "reading [the dictionary definitions] might have caused the jurors to place unmerited weight upon the definitions." Compare id. Starbuck argues that the other jurors might have given unmerited weight to the foreperson's opinion on "addiction" in the jury's discussions, because the other jurors knew that he had consulted dictionary definitions. I conclude, however, that such a concern was adequately addressed when I responded to the foreperson's note asking if the jurors could use dictionary definitions by expressly telling the jurors that they could not use a dictionary definition or other outside source. Response To Question From Jury (docket no. 224) at 1.

The last factors that the Mayhue court considered were "whether the jury had difficulty reaching a verdict prior to introduction of the dictionary definition," see id. at 924 (stating this question as part of the fourth factor), and "the timing of the verdict," presumably as one of "[a]ny other factors that relate to a determination of prejudice," id. (stating this as the fifth factor). In Mayhue, the court found that these factors weighed in favor of a new trial, because "the jury had difficulty reaching a verdict prior to the introduction of the dictionary definition" and because "[t]he jury was able to reach a verdict less than three hours after the foreperson read the definitions, despite having been plagued by `irreconcilable differences' the night before." Id. at 926. Here, however, there is no indication that the jurors were having difficulty reaching a verdict or that they were ever "plagued by `irreconcilable differences.'" Although the jurors met for only about an hour after the case was submitted to them before breaking until the following day, the jurors went home at about the same time that evidence had ended each of the prior days of the trial, with no indication that they were doing anything other than conforming to their prior schedule. Nor does the jury's rendering of a verdict just a few hours after I responded to their question about use of dictionary definitions indicate that the dictionary definitions were critical in resolving "irreconcilable differences." Again, there is no evidence that the jurors were having "irreconcilable differences," I instructed the jurors that they could not consider dictionary definitions, the foreperson then destroyed the copies of the dictionary definitions that he had obtained, and no one had described any dictionary definitions to the other jurors. The relative shortness of the jurors' deliberations is also reasonably explainable by the fact that the "addiction" question was the threshold question, and once the jurors resolved it, they did not have to proceed any further with their deliberations on other complicated issues.

Thus, notwithstanding that my exercise of discretion in the context of a juror misconduct claim might be given at least moderate deference, see Hardin, 52 F.3d at 938 n. 6 ("A more deferential standard of review is appropriate ... if the district court's new trial order is precipitated by jury misconduct or other prejudicial trial events that contaminate the jury's deliberative process."), I conclude that the jury foreperson's unauthorized consultation of dictionary definitions of "addiction," again, while not free from doubt, does not, standing alone, warrant a new trial. See United States v. Martinez, 14 F.3d 543, 551 (11th Cir.1994) ("Standing alone, the jury's use of the dictionary would not warrant a new trial."). Starbuck's Motion For New Trial on this ground is denied.

3. Verdict against the great weight of the evidence

As his second ground for a new trial, Starbuck asserts that the jury's verdict on "addiction" was against the great weight of the evidence. The defendants deny that Starbuck is entitled to a new trial on this ground.

a. Arguments of the parties

Starbuck contends that the "overwhelming" evidence was that he was addicted to nicotine in cigarettes. Starbuck contends that, notwithstanding the evidence of the addictive nature of cigarettes containing nicotine and the evidence of his own addiction that he presented at trial, the jury "inexplicably" found that he was never addicted to the nicotine in cigarettes. Starbuck argues, in essence, that the evidence and the verdict just do not add up, so that the jury's verdict is against the clear weight of the evidence. In contrast, the defendants argue that Starbuck cannot come close to demonstrating that the verdict was against the great weight of the evidence, because the jury heard ample evidence supporting its determination that Starbuck was not addicted to smoking. Specifically, they point to Dr. Ticknor's expert opinions; the testimony of Dr. Burns, one of Starbuck's own experts, that not all smokers, even daily smokers, are addicted; and Starbuck's testimony that he smoked for other reasons than addiction. Indeed, the defendants argue that the record reflects that Starbuck was able to quit smoking when he made up his mind to do so and that he has refrained from smoking for nearly twenty years. They also point to testimony from Starbuck's family members that he was able to control his smoking for "extended" periods of time. The defendants argue that the jury's verdict had a sufficient basis in the record and that their resolution of this disputed issue in the case should not be second-guessed.

b. Applicable standards

The Eleventh Circuit Court of Appeals has repeatedly stated that "`new trials should not be granted on evidentiary grounds unless, at a minimum, the verdict is against the great — not merely the greater — weight of the evidence.'" Lamonica, 711 F.3d at 1312-13 (quoting St. Luke's Cataract and Laser Inst., P.A., 573 F.3d at 1200 n. 16, in turn quoting Lipphardt v. Durango Steakhouse of Brandon, Inc., 267 F.3d 1183, 1186 (11th Cir.2001)). This is so, "[b]ecause it is critical that a judge does not merely substitute his judgment for that of the jury." Lipphardt, 267 F.3d at 1186. Indeed, the Eleventh Circuit Court of Appeals has explained, "If the jury's verdict is supported by the evidence, then it is immaterial that we or the district judge would have arrived at the same verdict because it is not our place to substitute our judgment for that of the jury." Auto-Owners Ins. Co. v. Southeast Floating Docks, Inc., 571 F.3d 1143, 1145 (11th Cir.2009). Although a district court may not grant a new trial simply because it would have reached a different verdict, a district court may grant a new trial, on the ground that the evidence is against the great weight of the evidence, "`even though there may be substantial evidence which would prevent the direction of a verdict.'" Lipphardt, 267 F.3d at 1186 (quoting Hewitt v. B.F. Goodrich Co., 732 F.2d 1554, 1556 (11th Cir.1984)). Again, exercising my discretion to grant a motion for new trial on this ground would be reviewed with the least deference. See Aronowitz, 513 F.3d at 1242; Hardin, 52 F.3d at 938 (footnote omitted).

c. Analysis

In the absence of contrary evidence, Starbuck's evidence would have required a verdict that he was addicted to cigarettes containing nicotine on or before November 21, 1996. This is so, because his experts testified that the appropriate definition of "addiction" was the NIDA definition, because that is the definition on which health professionals rely to diagnose and treat nicotine addiction. See Trial Transcript, December 2, 2014 (Afternoon), 39:11-23 (Dr. Cummings's testimony); Trial Transcript, December 10, 2014 (Morning), 28:8-13; 33:21-36:25 (Dr. Burns's testimony). Dr. Burns, one of Starbuck's experts, also explained that the DSM-V is not a proper source for a definition of "addiction" — even if it had one, and Dr. Ticknor, the defendants' expert concedes that it does not, Trial Transcript, December 12, 2014 (Afternoon), 65:23-66:9 — because it is a manual to classify behaviors, not to diagnose and treat addiction. Trial Transcript, December 10, 2014 (Morning), 28:8-13; 33:11-33:25 (Dr. Burns's testimony). That NIDA definition, as stated by Dr. Cummings, is that "[a]ddiction is defined as a chronic, relapsing brain disease that's characterized by compulsive drug seeking and use, despite harmful consequences" and that drugs "change [the brain's] structure and how it works." Trial Transcript, December 2, 2014 (Afternoon), 37:2-10.

Evidence of Starbuck's smoking and conduct related to smoking demonstrated unequivocally that he fit within the NIDA definition of "addiction," upon which his experts relied. Dr. Cummings testified — without contradiction — that the younger a person starts smoking, the harder it is for them to stop smoking, because of the changes that nicotine causes in their brains. Trial Transcript, December 2, 2014 (Afternoon), 37:15-39:10. Starbuck testified, without contradiction, that he began smoking regularly when he was fourteen or fifteen years old, see Trial Transcript, December 5, 2014 (Afternoon), 43:21-25, and that he continued to do so for the next four decades. Indeed, he testified that he smoked 30 to 40 cigarettes per day for over 40 years, smoking from when he woke up, throughout his waking hours, while working, with breaks of no more than 20 to 30 minutes, until he went to bed at night, and would even smoke when he woke up in the night. Id. at 60:14-61:22, 65:8-15. Starbuck also presented copious, largely unchallenged, evidence that he was unable to quit smoking, despite using a wide variety of cessation aids, because of the unpleasant symptoms, which are consistent with his expert's descriptions of the symptoms of withdrawal. See Trial Transcript, December 2, 2014 (Afternoon), 43:1-46:19 (Dr. Cumming's explanation that Starbuck suffered such symptoms, indicating that he was addicted); Trial Transcript, December 5, 2014 (Afternoon), 66:19-22, 76:2-77:24, 77:25-78:6, 74:1-78 (Starbuck's testimony about difficulties quitting, even with cessation aids). The only reasonable inference from this degree and history of smoking, based on Starbuck's evidence, is that he was engaged for forty years in compulsive nicotine seeking and use within the NIDA definition of "addiction." Trial Transcript, December 2, 2014 (Afternoon), 37:2-10. His evidence also demonstrated that he continued smoking, despite his awareness of the dangers of smoking, another part of the NIDA definition. Id. He pointed to messages from media, social interactions, and warnings on cigarette containers of the dangers of smoking and conflicts with his wife and stepdaughter over smoking, but he still did not quit. Id. at 66:1-67:10.

The weakness of the defendants' contrary evidence is apparent from the strained inferences that they attempt to draw from the evidence to which they point as sufficient to defeat a new trial motion. For example, Dr. Burns's concession that not all smokers, even daily smokers, are addicted to smoking and nicotine, see Trial Transcript, December 10, 2014 (Morning), 129:22-130:2, does absolutely nothing to counterbalance the evidence that Starbuck was addicted. Similarly, the defendants' identification of Starbuck's testimony that he enjoyed smoking, the feeling it gave him, the taste, and being part of a group of smokers, and that smoking helped him deal with stress and improved his mood, see Trial Transcript, December 8, 2014 (Afternoon), 28:7-19, does nothing to demonstrate that he was not addicted under the NIDA definition, or any other reasonable definition of addiction. Indeed, all of those statements are totally consistent with nicotine addiction. Dr. Cummings testified, inter alia, that the physiological mechanism triggered by nicotine makes it both enjoyable and addictive and creates unpleasant withdrawal symptoms when the smoker stops smoking. Trial Transcript, December 2, 2014 (Afternoon), 38:20-38, 39:19-10, 41:2-42:18, 43:1-46:19.

Even weaker is the defendants' evidence purportedly supporting their contention that Starbuck was able to control his smoking and "go for extended periods of time without smoking." Defendants' Opposition To Plaintiff's Motion For New Trial (docket no. 263), 14. The evidence cited by the defendants simply does not support that characterization. Starbuck's stepdaughter's testimony, to which the defendants point, is actually that Starbuck "wouldn't smoke for an hour" when he was working on a large glass project. Trial Transcript, December 12, 2014 (Afternoon), 150:1-18. Similarly, Starbuck's brother's testimony, to which the defendants also point in support of this proposition, is that Starbuck would not smoke "when he was working behind the torch... [b]ut there was an ashtray right there within his reach whenever he took a break," and

Trial Transcript, December 12, 2014 (Afternoon), 149:6-13. Starbuck's brother also testified that the longest Starbuck went without smoking when they were living together was about eight hours "when [Starbuck] was asleep." Id. at 48:25-149:3. In context, a maximum of one to two hours at a time during waking hours during days when Starbuck was otherwise smoking and for as much as eight hours while he was sleeping cannot reasonably be characterized as "extended periods of time without smoking."

The evidence cited by the defendants also does not support their contention that Starbuck "was able to quit smoking when he made up his mind to do so, and that he had refrained from smoking for nearly twenty years." Defendants' Opposition To Plaintiff's Motion For New Trial at 13. That characterization of the evidence relies on testimony from Starbuck that, after his cancer surgery in 1995 resulting in removal of part of his lung — and, incidentally, three weeks in a coma, as a result of complications — Starbuck did not smoke again, and, on the one occasion on which he had tried a cigarette after 1995, he had found it tasted awful. Trial Transcript, December 8, 2014 (Afternoon), 28:16-19. Absolutely nothing about that evidence is inconsistent with Starbuck being addicted to smoking cigarettes prior to his lung cancer surgery in 1995.

The defendants' attempt to generate a factual dispute on the definition of "addiction" is equally inadequate to preclude a new trial. This is so, for at least three independent reasons. First, Dr. Ticknor, the defendants' expert, conceded that the DSM-V, on which he relied, does not even define "addiction," so that he opined that "addiction" would be a "pretty extreme presentation" of "a tobacco use disorder," as defined in the DSM-V. Trial Transcript, December 12, 2014 (Afternoon), 65:23-66:9. Second, Dr. Ticknor admitted that there is "nothing wrong" with using the NIDA definition of "addiction." Id. at 93:21-94:6. Third, Dr. Ticknor's testimony that only about a third of the people who came to see him both wanted to quit and were motivated to do so, another third were "ambivalent," and that the last third "really have no desire to quit" because "for one reason or another — they like smoking," id. at 73:13-74:3, is entirely consistent with the NIDA definition of "addiction," which accounts for the lack of motivation that smokers have to quit smoking and for their liking to smoke, even if they were aware of the dangers of doing so. To put it another way, Dr. Ticknor's testimony simply directed the jurors' attention to smokers' motivations, without ever addressing the physiological effects on smokers of nicotine addiction and withdrawal that Starbuck's evidence demonstrated could overwhelm the exercise of free will. As Dr. Cummings explained — and the defendants' evidence does not refute — "[A]ddiction is not free will. Addiction is a constrained choice." Id. at 35:5-6.

In short, I conclude that the jury's determination that Starbuck was not addicted to cigarettes containing nicotine at some point during the relevant time period is outweighed by the great weight of the evidence, see Lamonica, 711 F.3d at 1312-13, "`even though there may be substantial evidence which would prevent the direction of a verdict'" in Starbuck's favor. Lipphardt, 267 F.3d at 1186 (quoting Hewitt, 732 F.2d at 1556).5 Starbuck's tsunami-like evidence of his addiction to nicotine cigarettes was stronger and more compelling than proof of any contested element of a claim in a civil case that I have presided over in the more than two decades that I have been a federal trial judge trying civil jury trials in five different federal courts spanning the District of the Northern Mariana Islands to the Middle District of Florida. On the other hand, the defense evidence that Starbuck was not addicted to nicotine cigarettes was weaker than any defense evidence attempting to counter an element of a plaintiff's claim that I can recall. The latter fact was obviously not owing to the lack of thoroughness, vigorousness, zealousness, monumental skill, or ability of defense counsel. While there were many contested issues later on in the verdict form on which a decision either way would not have surprised me, the jury's finding that Starbuck was not addicted is the most shocking "O'Henry ending" to any civil jury trial that I am familiar with.

Starbuck is entitled to a new trial on this ground.

4. Improper jury instruction on "addiction"

Although I concluded, just above, that Starbuck is entitled to a new trial on the second ground that he asserts, I will consider, in the alternative, whether he is also entitled to a new trial on the basis of an improper jury instruction. More specifically, Starbuck seeks a new trial on the ground that my addition of a timeframe for his addiction in the jury instructions heightened his burden and likely confused the jury. The defendants deny that Starbuck is entitled to a new trial on this ground.

a. Arguments of the parties

Starbuck concedes that the "Engle class" is defined as all Florida citizens and residents, and their survivors, who had suffered, were suffering, or had died from diseases and medical conditions caused by their addiction to cigarettes that contain nicotine as of November 21, 1996, the class cut-off date. Nevertheless, he argues that a putative class member need not prove addiction by a certain date — only that he or she developed a smoking-related illness on or before that date. He argues that the addition of a temporal element to the addiction element wrongly heightened his burden and, particularly in a case where he quit smoking before November 21, 1996, invited confusion by requiring the jury to find that he was addicted to nicotine "on or before" that date. The defendants counter that the definition of the "Engle class" required a member to be addicted to cigarettes containing nicotine before the class cut-off date, or they could not have suffered from a disease or medical condition legally caused by an addiction to cigarettes before the cut-off date. The defendants argue that the jurors could not have been confused by the addition of the temporal limitation in the jury instruction, precisely because Starbuck quit smoking for good before the cut-off date.

b. Analysis

As I noted, above, "[m]otions for new trial on the basis of erroneous and prejudicial jury instructions are within the district court's discretion." Gowski v. Peake, 682 F.3d 1299, 1310 (11th Cir.2012). The district court's ruling on such a motion is reviewed for abuse of discretion. Id. The Eleventh Circuit Court of Appeals has recognized that the jury instructions must not have any "tendency to confuse or to mislead the jury with respect to the applicable principles of law." SEC v. Yun, 327 F.3d 1263, 1281 n. 39 (11th Cir.2003) (internal quotation marks and citation omitted). Thus, "[i]f the instructions do not accurately reflect the law, and the instructions as a whole do not correctly instruct the jury so that [the appellate court is] left with a substantial and ineradicable doubt as to whether the jury was properly guided in its deliberations, [the appellate court] will reverse and order a new trial." Id. (internal quotation marks and citations omitted); accord Gowski, 682 F.3d at 1315 (explaining, in the context of a new trial motion based on erroneous jury instructions, "`If there is no basis in the record for the instruction given, such error may raise a substantial and ineradicable doubt as to whether the jury was properly guided in its deliberations, and reversal may be required'" (quoting Christopher v. Cutter Lab., 53 F.3d 1184, 1194 (11th Cir.1995))). Presumably, if the district court finds such circumstances, it may grant a new trial in the first instance.

I do not find that inclusion of a temporal limitation on Starbuck's addiction warrants a new trial. As I explained, when this issue arose during preparation of jury instructions,

11/18/14 Proposed Jury Instructions (docket no. 138-8), 16 n. 36 (Instruction No. 6 — Starbuck's "Eligibility" To Assert His Claim). Starbuck has done nothing to convince me that this conclusion was erroneous or that there was any error in the language I used to instruct the jury on the temporal limitation on addiction. Yun, 327 F.3d at 1281 n. 39. The pertinent element of the instruction in question stated that Starbuck must prove that he "was addicted to cigarettes containing nicotine on or before November 21, 1996." Instructions To The Jury (docket no. 221), 11 (No. 6 — Mr. Starbuck's Initial Elements). The instruction unambiguously did not require that Mr. Starbuck be addicted on November 21, 1996, or that he be addicted until November 21, 1996, only that he be addicted "on or before November 21, 1996." Where Mr. Starbuck quit smoking for good not later than January 1995, he clearly could not have been addicted to smoking cigarettes containing nicotine outside of the period specified in the jury instruction.

Starbuck is not entitled to a new trial on the ground that I improperly added a temporal limitation on when he was addicted in the pertinent jury instruction.

5. Summary

Starbuck is entitled to a new trial on the second ground that he asserts, that the jury's verdict that he was not "addicted" was against the great weight of the evidence. His Motion For New Trial is denied, however, as to his first and third grounds, juror misconduct in looking up dictionary definitions of "addiction" and a jury instruction imposing a temporal limitation on his "addiction."

B. The Defendants' Motion For Attorneys' Fees And Costs

The other motion now before me is the defendants' February 24, 2015, Motion For Attorneys' Fees And Costs (docket no. 258) pursuant to section 768.79, Florida Statutes; Rule 54(d)(2) of the Federal Rule of Civil Procedure; and Local Rule 4.18. Because I have granted Starbuck's Motion For New Trial, in part, however, the defendants' Motion is denied as moot.

III. CONCLUSION

Upon the foregoing,

1. Plaintiff Starbuck's March 10, 2015, Motion For New Trial (docket no. 260) is granted in part and denied in part, as follows:

2. The defendants' February 24, 2015, Motion For Attorneys' Fees And Costs (docket no. 258) is denied as moot.

3. A new trial in this matter will be set by separate order by a new judge. Due to the unreasonably contentious nature of this trial and the around-the-clock filings by the defense, I will not be available as a volunteer for the retrial, should one actually happen.

IT IS SO ORDERED.

IN THE UNITED STATES DISTRICT COURT

FOR THE MIDDLE DISTRICT OF FLORIDA

JACKSONVILLE DIVISION

VERDICT FORM

On Mr. Starbuck's claims and RJR's and PM USA's specific defense, we, the Jury, find as follows: