OPINION
PER CURIAM.
Debra Lynn Henley, individually and on behalf of her minor child, is a named representative in a class action in which the named class members filed suit in the United States District Court for the Southern District of West Virginia against FMC Corporation, alleging that they were injured by a cloud of phosporous trichloride that escaped an FMC chemical plant because of a gas leak. There were a total of
forty named representatives,
but after the class was divided into sub-classes, only fourteen were chosen as initial “trial plaintiffs” (the trial plaintiffs are collectively referred to as the Appellants).
After the jury awarded the Appellants compensatory and punitive damages, the district court granted a new trial based upon newly discovered evidence in the form of affidavits submitted by a witness who had been unavailable at trial. The Appellants now appeal the jury’s unfavorable verdict at the second trial, arguing that the district court erred in granting a new trial based upon the new affidavits, that the district court erred in concluding that the jury’s findings at the second trial are binding on the entire class, and that the district court erred in refusing to grant the Appellants a new trial following the second jury verdict. We conclude that the district court did not abuse its discretion in granting the second trial based upon newly discovered evidence. However, because at the second trial, the district court erred in fading to instruct the jury as to the bifurcation of the class-wide issue of causation from the individual issues of causation and damages, we conclude that yet another trial is necessary. We therefore affirm in part, reverse in part, and remand for further proceedings consistent with this opinion.
I.
On December 5, 1995, FMC’s chloride unit ruptured at its Nitro, West Virginia plant and released a hazardous cloud. The rupture was caused, at least in part, by failures in FMC’s monitoring devices and safety valves. After the release of the cloud, hundreds of people alleged that they had suffered injuries as a result of cloud exposure. On December 11, 1995, the Appellants sued FMC in the United States District Court for the Southern District of West Virginia. The Appellants sought class certification, with a proposed class consisting of “all persons or other entities, who or which sustained damage as a result of the leak of toxic gas from the Nitro, West Virginia facility of [Defendant] on December 5, 1995.” (J.A. at 226 (alteration in original).) On January 22, 1997, the district court issued a memorandum opinion and order granting a conditional certification of the class pursuant to Federal Rules of Civil Procedure 23(b)(3) and (c)(4)(A).
On March 5, 1998, the district court issued a case management order certifying six subclasses.
On March 13, 1998, the parties agreed to a joint plan under which each side would select an equal number of named representatives to serve as “trial plaintiffs .” (J.A. at 322.) On September
18, 1998, the first trial began with these fourteen representatives. The case was to be tried in two phases. In the first phase of the trial, the jury was to decide the class-wide issues of negligence and strict liability, as well as compensatory damages for the Appellants. The same jury would decide whether FMC’s conduct warranted punitive damages, which would be awarded in a single lump sum and then allocated, after adjudication or settlement of the individual claims, to all class members successfully proving actual damages stemming from the cloud.
The second phase of trial, “if necessary, would involve the disposition of individual claims of class members,” under which “[rjesolution [could] occur either by a series of mini-trials or disposition by a special master or mediator.” (Case Management Order of Sept. 4, 1998.) “If mini-trials [were to be used], the juries would be instructed as to the findings on common issues of the original jury.”
(Id.)
At trial, FMC offered the testimony of its expert, Dr. Tony Eggleston, to show that the wind direction had carried the cloud across a largely uninhabited portion of the state. Eggleston based his opinion partially upon data gathered from a wind monitoring station run by the West Virginia Division of Environmental Protection (DEP). Charles Spann worked at that DEP station and Eggleston spoke with Spann before trial to confirm the accuracy of the data and the equipment used to gather that data." Spann was listed as a potential witness by both parties, but Spann was unavailable to testify because he was out of the country.
After Eggleston was excused by the district court and left town, the Appellants offered Steve Drake as a rebuttal witness to undermine Eggleston’s testimony. Drake was an employee of the DEP who was supervised by Spann. The Appellants had not previously disclosed Drake as a witness to either the district court or to FMC, and Drake had not been a subject of discovery. Although counsel for the Appellants represented that Drake had been discovered as a witness only on the previous day, the district court found that counsel actually had become aware of Drake before Eggleston left town. Nevertheless, Appellants’ counsel did not disclose Drake as a witness until after Eggleston was no longer available. The Appellants offered Drake’s testimony to show that the equipment upon which Eggleston relied was defective and therefore that Eggleston’s expert opinion was flawed.
The district court was concerned about the surprise nature of Drake’s testimony, as well as the questionable manner in which the Appellants had proceeded in offering the testimony:
[Y]ou had access to this type of information as early as Wednesday. You went ahead and crossed the witness Eggle-ston with a view toward coming up with refuting type of information, and then you disclosed to counsel, once Dr. Eg-gleston left the stand and left the jurisdiction, that you had a witness who would significantly attack or erode the effect of Mr. or Dr. Eggleston’s testimony. So we get back to that point, which is one that the Court remains concerned about, and that is, as everyone here concedes, trials like this are not to be
conducted by way of ambush but rather following the rules of discovery.
(J.A. at 2247.) After hearing Drake’s testimony outside of the presence of the jury, the district court noted the “immense probative value” of the evidence and its “possible real significance as to the validity of the testimony of some of the witnesses.” (J.A. at 2281.) The district court also continued to express reservations as to whether to admit the evidence:
[T]he evidence comes about and comes before this Court and a jury ... under highly questionable circumstances.... I suggest that plaintiffs’ counsel quite obviously knew of the information, knew what plaintiffs’ counsel intended to do and made no disclosure to the defense.
(J.A. at 2281-82.) Recognizing the “very difficult balancing function for any trial judge in a situation like this,” the district court nevertheless permitted Drake to testify before the jury as to certain documents, including some authored by Spann, but not as to others. (J.A. at 2281-82.) The district court based its decision in part upon the assumption that Drake was authorized by the DEP to testify and speak as to the documents:
I note my concern that Mr. Drake says he is not the keeper of the records, and he made that clear. I assume, on the other hand, that the DEP section chief and the counsel for DEP have given him the authority to come down here, whether he has the authority or not, and to speak to those records. And I assume that if it were necessary to authenticate these records as business records, it could be done. So that’s one other reason that I will allow it in, even aware of the fact that this man is not the keeper of the records. And I just think that the evidence may have a significance or probative value that slightly outweighs the prejudicial effect.
(J.A. at 2283.)
During his testimony, Drake told the jury, based upon several documents, including some authored by Spann, that the weather equipment was defective. Drake’s testimony critically undermined the basis of Eggleston’s opinion and, therefore, fatally damaged FMC’s defense. The jury thereafter returned a verdict in favor of the Appellants, awarding compensatory damages in a total amount of $83,000 to ten of the fourteen Appellants, in amounts ranging from $6,000 to $17,500. After FMC stipulated its net worth at $776.6 million, the jury awarded punitive damages in the amount of $38.8 million.
Upon Spann’s return from his travels, he submitted two affidavits contradicting the testimony of Drake and affirming the accuracy of the DEP equipment, and, on January 6, 1999, FMC filed a Notice of Newly Discovered Evidence based upon Spann’s affidavits. Spann’s affidavits stated that, contrary to Drake’s testimony, the equipment upon which Eggleston relied likely was not defective and that Drake had been asked to “testify at trial without having all of the information necessary to do so.” (J.A. at 2850.) The district court, in an opinion extensively describing its reasons and the history of the Drake/ Spann issue, granted FMC a new trial. Among other things, the district court found that “the actual trial testimony from Drake effectively dismantled FMC’s defense. ... New evidence offered by FMC post-trial has laid waste to, not simply impeached or impugned, Drake’s trial testimony. To ignore this newly discovered evidence would amount to a miscarriage of justice.” (J.A. at 3660-01.) The district court recognized that “[p]ractically, FMC’s entire defense rested on the foundation laid by Dr. Eggleston,” and that the Ap
pellants relied extensively on Drake to rebut Dr. Eggleston’s opinion. (J.A. at 3662.) The district court reemphasized that it was “disturbed greatly by the misrepresentations, innocent or intentional, Plaintiffs’ counsel made concerning Drake’s proposed testimony.” (J.A. at 3667-68 n. 6 .)
At the second trial, EMC accepted responsibility for the cloud itself, and framed the issue as whether the Appellants had been exposed to and injured by the cloud. The jury found in favor of EMC. Under the jury form, the jury could not decide punitive damages absent a finding of compensatory damages for one or more of the Appellants. Thus, the jury did not reach the question of punitive damages in rendering its verdict in favor of FMC.
On November 22,1999, the district court issued a judgment order in favor of FMC, directing the Clerk to remove the case from the active docket of the court. On December 2, 1999, the Appellants filed a post-trial memorandum seeking a new trial, to set the remaining cases of the other members of the class action for jury trial, and to reinstate the verdict of the first trial. On April 18, 2000, the district court denied the Appellants’ post-trial motions, concluding that the jury’s finding that the Appellants had not been injured by the cloud and its failure to reach the punitive damages question was binding on the entire class. On April 18, 2000, the Appellants noted their appeal.
The Appellants raise several issues on appeal. First, the Appellants argue that the district court erred in granting a new trial after the first jury verdict. Second, the Appellants argue that the second jury’s findings should not bind the entire class. Third, the Appellants argue that the district court erred in declining to grant a new trial after the second jury verdict. We address each issue in turn.
II.
Plaintiffs first challenge the district court’s decision to grant a new trial on the basis of newly discovered evidence. The district court, in granting the new trial, found that “FMC was excusably ignorant and exercised appropriate diligence upon Spann’s return to expose Plaintiffs’ calculated, late, surprise and successful effort to crush the cornerstone of its defense.” (J.A. at 3684.) The Appellants argue that the evidence was not newly discovered because FMC could have tried to depose Spann earlier to verify Dr. Eggleston’s opinion and that FMC also could have found other means to verify the accuracy of the DEP data on which its entire case relied. The Appellants also argue that Spann’s testimony was presented only to impeach Drake’s testimony and that the evidence, even if new, was not material. FMC responds that the district court did not abuse its discretion in granting a new trial to offset the Appellants’ use of an eleventh-hour surprise witness who testified incorrectly about a subject on which he had little knowledge, i.e., the accuracy of the DEP instruments and the data gathered therefrom, and which had not been an issue that previously had been in dispute.
Under Federal Rule of Civil Procedure 59, the district court may grant a new trial based upon newly discovered evidence.
In this circuit, the standard governing relief on the basis of newly discovered evidence is the same whether the motion is brought under rule 59 or rule 60, and rule 60 requires that a party demonstrate: (1) the evidence is newly discovered since the judgment was entered; (2) due diligence on the part of the movant to discover the new evidence has been exercised; (3) the evidence is not
merely cumulative or impeaching; (4) the evidence is material; and (5) the evidence is such that is likely to produce a new outcome if the case were retried, or is such that would require the judgment to be amended.
Boryan v. United States,
884 F.2d 767, 771 (4th Cir.1989) (internal citation omitted).
“Thus, in order to support a motion for reconsideration, the movant is obliged to show not only that this evidence was newly discovered or unknown to it until after the hearing, but also that it could not with reasonable diligence have discovered and produced such evidence at the hearing.”
Id.
(internal quotation marks omitted). The district court made extensive findings to support its decision to grant a new trial, and we review those findings for an abuse of discretion.
Cline v. Wal-Mart Stores, Inc.,
144 F.3d 294, 301 (4th Cir.1998). Under the abuse of discretion standard, we may not “substitute [our] judgment for that of the district court; rather, we must determine whether the court’s exercise of discretion, considering the law and the facts, was arbitrary or capricious.”
United States v. Mason,
52 F.3d 1286, 1289 (4th Cir.1995).
A.
The first two elements of the newly discovered evidence test are whether the evidence is, “in fact, newly discovered,” and whether there are facts alleged “from which the court may infer diligence on the part of the movant.”
United States v. Custis,
988 F.2d 1355, 1359 (4th Cir. 1993). The Appellants argue that “FMC could have taken Mr. Spann’s deposition at any time before trial, but chose not to do so,” (Appellant’s Br. at 29) and that FMC had access to Spann before trial, as demonstrated by the fact that Eggleston verified with Spann the data that formed the predicate of his expert opinion before Spann left the country. The Appellants also point to the fact that Spann was on both parties’ witness lists before trial, although he subsequently could not be called at trial because of his travels. The district court found that the Appellants “waited until the waning days of a lengthy trial to spring a surprise witness and unexpected evidence upon FMC’s lawyers and then only after Dr. Eggleston left town.” (J.A. at 3683.) Accordingly, the district court found that “FMC was excusably ignorant and exercised appropriate diligence upon Spann’s return to expose Plaintiff’s calculated, late, surprise and successful effort to crush the cornerstone of its defense.” (J.A. at 3684.)
We cannot conclude that the district court abused its discretion in finding
that Spann’s affidavits satisfied the first two elements of the newly discovered evidence test. First, Spann’s testimony was not within FMC’s possession prior to or during trial because he was out of the country in a remote location. Second, although FMC was aware of Spann prior to trial, the accuracy of the equipment itself was not in dispute until late in the trial, when the Appellants offered Drake as a surprise witness after Spann was known to be unavailable and Eggleston had already left the jurisdiction. We do not believe that FMC, even exercising the height of diligence, could have anticipated that it would have to develop evidence to respond to inaccurate testimony by a surprise witness, Drake, who was called to testify about matters outside of his knowledge.
(See
J.A. at 2709, 2707 (Spann affidavit of December 23, 1998, stating that he believed that “Drake was placed in a difficult position by being asked to testify at trial without having all of the information necessary to do so” and that Drake did not consult with Spann prior to testifying).) As stated by the district court, “Rule 59 requires a moving lawyer to be duly diligent, not a prophet.” (J.A. at 3684 n. 9.) Third, we do not believe that FMC’s failure to question Drake before trial as to the accuracy of the data reflects a lack of diligence; indeed, prior to trial, FMC’s expert had verified the data with Spann, who was Drake’s supervisor and who had authored the most important document about which Drake testified.
(See
J.A. at 2539 (statement by FMC’s counsel to district court that “we have evidence from [Eggleston] that he spoke with [Spann] and [Spann] was the source of the information that he got on which he relied, and he checked with him and was told it was good”); 3666 (district court’s characterization of Spann as “the author of the most hotly contested of the three documents”).) We simply do not agree that FMC had to question every DEP employee duplicatively as to matters about which Spann had already spoken and about which Spann clearly had the best knowledge, particularly where there was no indication that the accuracy of the equipment would be in dispute. Certainly, had Spann been available to testify at trial, FMC would have had little reason to question Drake before trial as to matters already verified by Spann. We do not believe that FMC’s failure to question Drake somehow reflects a lack of diligence simply by virtue of the fact that Spann subsequently became unavailable for trial. For these reasons, and in light of the unique factual setting of this case, we have no difficulty concluding that the district court did not abuse its discretion in finding that Spann’s affidavits were new evidence and that there were facts from which the district court could infer due diligence on the part of FMC.
B.
The third element of the newly-discovered-evidence standard is whether the evidence is “not [ ] merely cumulative or impeaching.”
Custis,
988 F.2d at 1359. The Appellants contend that Spann’s testimony “was offered for no other purpose than to contradict the testimony of Mark Drake,” (Appellants’ Br. at 32), and, therefore, it was merely impeachment evidence. The district court found that “[t]he new evidence here ... is much more than grist for the mill of impeachment” because “[i]t not only serves to seriously deprecate Drake’s trial testimony and his interpretation of the documents, but also supplants that testimony entirely, replacing it with his supervisor’s contrary account of the critical wind direction evidence.” (J.A. at 3686.) We agree.
Although the Appellants argue that the only purpose of Spann’s affidavits was to impeach Drake’s testimony, the district
court did not abuse its discretion in concluding that Spann’s affidavits do not simply create a conflict in evidence or offer an alternative version of the facts.
See United States v. Estabrook,
774 F.2d 284, 290 (8th Cir.1985) (affirming district court’s denial of new trial based on affidavits asserting the falsehood of a government witness’ testimony because the affidavits served only to refute the witness’s testimony by attacking the witness’s credibility and offering an alternative version of the facts). Drake was offered as a witness by the Appellants, and reluctantly accepted by the district court, only because Spann was unavailable for trial. Spann, however, was self-evidently more knowledgable than Drake as to the subject of Drake’s testimony, both as Drake’s supervisor and as author of the most important notations and documents about which Drake sought to speak. Spann, in other words, was clearly in a better position than Drake to testify and the jury should have heard Spann, rather than Drake. We therefore agree with the district court that Spann’s affidavits did not merely impeach Drake’s testimony or create a simple conflict in the evidence, but rather that Spann’s affidavits wholly “supplant[ed]” Drake’s testimony. (J.A. at 3686.) Accordingly, the district court did not abuse its discretion in concluding that Spann’s affidavit could serve as a proper basis for a new trial.
C.
The fourth and fifth elements of the test for newly discovered evidence are whether it is “material to the issues involved,”
Custis,
988 F.2d at 1359, and whether the “evidence is such that it is likely to produce a new outcome if the case were retried, or is such that would require the judgment to be amended,”
Boryan,
884 F.2d at 771. The district court found that the timing and nature of Drake’s testimony were crucial to the case because Drake’s testimony undermined Eggleston’s testimony, which formed the backbone of FMC’s defense, and “Drake was the last witness the jury heard and, judging by their expressions, the members were riveted by his not-to-be-misunderstood, matter-of-fact testimony.” (J.A. at 3685.) The district court therefore concluded that “the newly discovered evidence is both material and of a nature that could likely lead to a different result at a second trial.” (J.A. at 3686.)
The district court did not abuse its discretion in so finding. First, it is notable that presented with Spann’s testimony, the jury at the second trial found in favor of FMC, clearly signaling the outcome-determinative and material nature of the new evidence. Second, although the Appellants now assert that the DEP wind direction data and the accuracy of the machines was immaterial because wind direction data is inaccurate when measured over short time intervals, we note that the Appellants argued the opposite when seeking to admit Drake’s last-minute testimony.
(See
J.A. at 2246 (arguing to the district court that “[the fact that] this may not be accurate information from Nitro is very important”).) Notably, only after considering the assertedly highly probative nature of
Drake’s testimony did the district court decide, reluctantly, to admit Drake’s testimony notwithstanding the prejudicial nature of the Appellants’ conduct, stating that the “immense probative value” of Drake’s testimony “slightly” outweighed the prejudicial nature of the testimony. (J.A. 2281, 2283.) Accordingly, we agree with the district court, which saw Drake’s testimony firsthand and witnessed the devastating effect of his testimony on FMC’s defense, that Spann’s new evidence was both material and was such that it was likely to produce a new outcome on retrial.
For these reasons, we conclude that the district court did not abuse its discretion in granting a new trial based on newly discovered evidence.
III.
The Appellants next challenge the district court’s decision to deny relief to the other class members once the Appellants’ claims failed at the second trial. As discussed further below, the jury form conflates the issue of whether the cloud released by FMC was located in the vicinity of the Appellants with the issue of whether each Appellant individually proved that he had been injured by the cloud. Thus, the jury’s verdict in favor of FMC demonstrates that the Appellants are not entitled to relief, but it does not indicate whether the Appellants are not entitled to relief because they were not located in an area covered by the cloud, which was the primary issue common to the entire class. If we could determine with confidence that the jury’s verdict indicates that it resolved this class-wide issue in favor of FMC, we would have no difficulty affirming the district court’s decision to deny relief to the other class members once the claims against the Appellants failed. Based upon the jury form, however, we cannot make this determination. Accordingly, while we affirm the entry of judgment in favor of FMC with respect to the Appellants, we must remand for a new trial involving a different group of class representatives.
The district court conditionally certified the class action pursuant to Federal Rules of Civil Procedure 23(b)(3) and 23(c)(4)(A).
The primary common issue that was to be litigated class-wide was
whether the cloud was capable, by virtue of its placement, of causing the injuries asserted by the class.
After resolution of this class-wide issue, each Appellant was required to prove on an individualized basis that he was harmed as a result of the cloud.
Because recovery would be authorized only upon resolution of these individual issues, the district court apparently contemplated that individualized trials on causation and damages would be necessary after the class-wide issue had been resolved. (J.A. at 324r-27; Case Management Order, at 1-2 (envisioning a bifurcated trial that separated common issues from individual issues, under which the first phase of trial would resolve class-wide issues, punitive liability, and compensatory damages to the trial plaintiffs, and the second phase of trial would resolve the individual claims of other class members).)
During the trial, conflicting evidence was introduced regarding the placement of the cloud. Eggleston testified that the cloud traveled only over largely uninhabited areas, and, thus, that it could not have caused the injuries alleged by the Appellants. (J.A. at 2100 (testifying that, in his opinion, none of the Appellants were within the area of the cloud)). However, several Appellants testified that they were in or near Nitro at the time of the release, saw the cloud, and experienced serious injuries as a result of the exposure. (J.A. at 3878, 3888 (testimony by JoAnn Gaylor that she saw “puffs of clouds that were coming out of FMC” and that she experienced respiratory problems afterwards).)
Upon the close of the evidence, despite FMC’s objection, the district court submitted interrogatories to the jury that conflated the class-wide issue of placement of the cloud with individual issues of whether the cloud caused actual and proximate harm to each Appellant.
The jury form stated, “Did any one or more of the Plaintiffs prove by a preponderance of the evidence
that on December 5, 1995 they were (a) exposed to chemicals from the FMC plant; and (b) that such exposure was the proximate cause of either physical injury to them or damage to their property or both.” (J.A. at 4404 .) It then listed each Appellant separately, with boxes beside each name for the jury to check ‘Yes” or “No.” (J.A. at 4404.) The jury form further provided that, if the jury answered, ‘Yes,” with respect to any individual Appellant, the jury should proceed to determine the amount of compensatory damages with respect to each. (J.A. at 4404-05.) Because the placement of the cloud was disputed and the district court failed to bifurcate the class-wide issue of placement of the cloud from the individualized issues, there is no means to discern whether the jury denied relief to the Appellants due to the jury’s conclusion that the cloud was not located in the vicinity of the class members. Without this determination, we cannot resolve whether the jury’s denial of relief is binding on the entire class.
If the Appellants were denied relief solely because of their failure to prove individual harm and damages and not because the jury rejected evidence indicating that they were within the vicinity of the cloud, the other class members should not have been denied individualized trials. If, on the other hand, the jury’s verdict as to the Appellants resulted from the Appellants’ failure to prove that they were within the area covered by the cloud, that failure of proof, as an issue common to the entire class, would apply to all class members equally, and the district court would have acted properly by denying individual trials to the other class members. Accordingly, while we affirm the entry of judgment against the Appellants, we must remand for the district court to conduct a new trial before a jury properly instructed to resolve the common issues separately from and prior to the individual issues.
IV.
In sum, we cannot conclude that the district court abused its discretion in granting a new trial based upon newly discovered evidence. However, because the district court erred in failing to bifurcate the common issues from the individual issues, and because we therefore cannot determine whether the denial of relief as to the Appellants was due to the jury’s conclusion that the Appellants failed to prove that they were within the vicinity of the cloud rather than that the Appellants were not individually injured by the cloud, we remand for further proceedings consistent with this opinion.
AFFIRMED IN PART, REVERSED IN PART, AND REMANDED.