Is limited remand required if the district court admitted or excluded evidence without a Daubert analysis?
How a federal court of appeals disposes of a case after ruling on the merits depends on two federal statutes. Under 28 U.S.C. [section] 2111, a federal court of appeals may not reverse in the absence of harmful error. (1) Under 28 U.S.C. [section] 2106, once the court has determined that harmful error occurred, it may dispose of the appeal by directing the trial court to hold any further proceedings that may be appropriate. (2)
This article discusses the interplay between those statutes when the trial court has admitted or excluded evidence without first making the proper threshold finding on admissibility. In that situation, the question arises whether an appellate court must, instead of ordering a new trial, remand with instructions to the trial court to (i) determine whether the evidence was inadmissible, or if the court originally excluded the evidence, whether the evidence was instead admissible, and (ii) order a new trial if admitted evidence should have been excluded or excluded evidence should have been admitted. This article takes the position that the federal courts of appeals should retain the power to decide this question on a case-by-case basis.
II. DAUBERT GATEKEEPING ERRORS ARE NOT ALWAYS GROUNDS FOR REVERSAL
Recently, an en banc panel of the Ninth Circuit considered the question of how to dispose of an appeal after determining that the trial court erred by admitting expert testimony without performing its gatekeeping role. In Estate of Barabin v. AstenJohnson, Inc., the court found the admission of the testimony to have been harmful error, reversed the trial court's judgment, and remanded for a new trial. (4) The Barabin court recognized the necessity of finding harmful error before a mistake in ruling on admissibility is deemed to require reversal, (5) and it implicitly acknowledged that reversal would be unnecessary if the record demonstrated that the expert testimony was in fact admissible. (6) But the court nonetheless concluded that the error was harmful because the appellees' claim depended wholly on the erroneously admitted evidence, and held that a new trial was required because the record was too sparse to determine whether the expert testimony satisfied the Daubert requirements. (7)
The appellees sought review in the Supreme Court, asserting that remand for new trial was improper because without a ruling on whether the evidence was indeed admissible under Daubert, the court of appeals could not have properly found harmful error. (8) They also argued that, in contrast to other circuits, both the Ninth and Tenth Circuits were applying an automatic-new-trial rule in cases in which a district court failed to perform its Daubert gatekeeping role. (9) They urged the Supreme Court to create a limited-remand rule for this situation.
The rule that they proposed would require remand for the limited purpose of the trial court's undertaking the Daubert analysis that it neglected to perform in the first instance, rather than remanding for a new trial. If, and only if, the district court concludes that it erred in admitting (or excluding) the expert testimony would a new trial be required.
This limited-remand rule would come into play when a district court erred in failing to perform its Daubert gatekeeping duty, and the court of appeals could not deem the error harmless. Thus, the rule would apply only if it was not readily apparent from the appellate record that (i) the expert testimony was either admissible or inadmissible, and (ii) other competent evidence was not sufficiently strong to permit the conclusion that the improper admission or exclusion of the evidence had no effect on the decision.
III. REQUIRING REMAND FOR GATEKEEPING ERRORS WOULD FRUSTRATE CONGRESSIONAL INTENT
A. An Analysis of 28 U.S.C. [section] 2106
A limited-remand rule would run afoul of Congress's expressed intent in 28 U.S.C. [section] 2106. When a district court fails to comply with its Daubert gatekeeping duty, it commits error. What to do about that error is a matter committed to the appellate court's sound discretion under [section] 2106. (10) Exercise of that discretion by the court of appeals, and its choice of remedies, is informed by the character and degree of harm resulting from the district court's error.
Under [section] 2106, it is left to the courts of appeals to determine, after concluding that error has been committed, what further proceedings are "just under the circumstances." (11) The Fifth Circuit has, for example, explained that "[o]nce jurisdiction attaches, Courts of Appeals have broad authority to dispose of district court judgments as they see fit." (12) This includes the authority to, among other things, grant a new trial in the interest of justice. (13)
The First Circuit has explained the difference between the [section] 2111 issue (whether there is harmful error) and the [section] 2106 issue (the remedy for trial-court error) in a case where the district court excluded an expert's testimony under Daubert:
[W]hen a trial court erroneously excludes evidence, and the exclusion meets the standard criteria of harmfulness, the harm is not cured by a mere possibility that other appropriate grounds for exclusion of the same evidence may later be found to exist. The question is one of degree and the choice of remedies (including whether to require a new trial or merely remand for further findings) is ours. (14)
This observation applies equally when the district court's admission of evidence ran afoul of Daubert. The resulting harm is not cured by a mere possibility that appropriate grounds for admitting the evidence may later be found to exist. That said, the character and degree of harm informs the court's choice of remedies. And in that situation, whether to remand for a new trial or to remand solely for further findings is the court of appeals' choice to make under [section] 2106. (15)
The federal courts of appeals do not adhere, and should not be required to adhere, to a rigid limited-remand rule when the district court has erred in making or failing to make a threshold admissibility determination. In some cases involving this type of error, a federal court of appeals will, in the exercise of its broad discretion, properly conclude that although the district court's error was not harmless, a limited remand is appropriate. (16) But in other cases, a federal court of appeals may properly exercise its discretion under [section] 2106 by concluding that a remand for a new trial is the appropriate remedy. (17)
In one case, for example, the district court excluded expert testimony without first conducting a Daubert analysis. (18) The Sixth Circuit held that it was not harmless error because "the complexion of the proceedings would likely have changed had the district court conducted a Daubert hearing and determined that [the expert's] testimony was admissible." (19) Rather than ordering a limited remand for the district court to conduct the omitted Daubert analysis, the court of appeals reversed for a new trial. (20)
Similarly, the Seventh Circuit remanded for a new trial when the district court failed to conduct a proper Daubert analysis in a case involving a defendant who sought to present expert testimony on false confessions and his susceptibility to coercion. (21) The district court excluded the testimony without any indication that it applied the Daubert framework. (22) The court of appeals neither decided that the excluded testimony was admissible under Daubert nor ordered a limited remand for the district court to make that determination, instead ordering a new trial because "[t]he district court's failure to test the [proffered expert testimony] under [the Daubert] framework may have led to the exclusion of critical testimony." (23)
Likewise, the Fifth Circuit has exercised its discretion to order a new trial under [section] 2106 when the district court failed to decide a threshold question of admissibility. (24) On the other hand, however, the Fifth Circuit has also held a limited remand appropriate so that the district court could determine the admissibility issue in light of a new federal agency report issued after the district court made its initial admissibility determination. (25) And the D.C. Circuit has also considered a limited remand for an admissibility hearing, but decided against ordering one. (26)
In short, under [section] 2106, it is up to the federal courts of appeals to decide, on a case-by-case basis, and in the exercise of their broad discretion, the proper disposition of each case on appeal. (27) Thus far, the federal courts of appeals have taken various approaches to analyzing the necessity for remand when confronted with errors involving threshold admissibility rulings. Some have ordered limited remands, and some have remanded for new trials. The cases demonstrate, however, that the outcome always depends on case-specific factors. Imposing a rigid limited-remand rule would thwart the discretion that [section] 2106 affords the federal courts of appeals in deciding how to dispose of the cases before them.
B. An Analysis of 28 U.S.C. [section] 2111
A rule requiring a limited remand to complete the harmful-error analysis would frustrate the purpose of 28 U.S.C. [section] 2111. In that statute, Congress expressed its preference for determining harm by "case-specific application of judgment, based upon examination of the record." (28) And the Supreme Court has explained that the harmful-error analysis is intended to be flexible and without "rigid rules." (29) Thus, "[t]he factors that inform a reviewing court's 'harmless-error' determination are various," (30) and include case-specific considerations like an "estimation of the likelihood that the result would have been different" and a "consideration of the error's likely effects on the perceived fairness, integrity, or public reputation of judicial proceedings." (31) Consequently, the federal courts of appeals must refrain from generalizing too broadly about particular kinds of errors, and must remember that "the specific factual circumstances in which the error arises may well make all the difference." (32)
Under this flexible analysis, the federal courts of appeals may properly find harmful error when a district court does not perform its gatekeeping obligation, and they may do so without ordering a limited remand solely for the purpose of determining admissibility. After all, establishing harmful error is not "particularly onerous," (33) and allowing a verdict to be based on expert testimony that has not first been vetted under Daubert plainly implicates an appellant's substantial rights. (34)
A district court's failure to abide by its Daubert gatekeeping duty always affects "the perceived fairness, integrity, or public reputation of judicial proceedings." This effect alone may not be sufficient to require reversal, but under the elastic harmful-error analysis, it is a factor to be considered. As the Supreme Court has said, "[o]ften the circumstances of the case will make clear to the appellate judge that the ruling, if erroneous, was harmful." (36) Ultimately, then, it is up to the federal courts of appeals to make the harmful-error decision on a case-by-case basis. (37) And they are competent to do so.
For example, the Sixth Circuit held that a district court's error in excluding expert testimony without first conducting a Daubert analysis was not harmless because "the complexion of the proceedings would likely have changed had the district court conducted a Daubert hearing." (38) The Seventh Circuit, too, has held that when the district court excluded evidence without conducting a full Daubert analysis, the error was harmful because it "may have led to the exclusion of critical testimony." (39) And the First Circuit has also found a district court's error in failing to conduct a Daubert analysis to be harmful because there was no basis in the record other than the expert's testimony for the jury's damages award. (40)
But the federal courts of appeals also routinely find this sort of error to be harmless. The Tenth Circuit has concluded, for example, that a district court's failure to perform its gatekeeping role was harmless, and affirmed the district court's judgment. (41) Similarly, the Sixth Circuit has held that a district court's failure to conduct a formal Daubert hearing amounted to harmless error because the expert's testimony was cumulative. (42) And the Ninth Circuit has held, on several occasions, that a district court's Daubert gatekeeping error was harmless. (43)
In short, the courts of appeals are doing precisely what the Supreme Court said they should be doing: deciding the harmless error issue on a case-by-case basis by considering case-specific factors and by not generalizing too broadly about particular types of errors. (44)
IV. A LIMITED-REMAND RULE WOULD NOT SERVE JUDICIAL ECONOMY
In this era of vacant judgeships, confirmation gridlock, and heavy judicial workloads, (45) the federal courts of appeals cannot blind themselves to reality. Whatever the theoretical attractions of a limited-remand rule, it seems likely to result in the hearing of three appeals in many cases. If a federal court of appeals were to order a limited remand for a Daubert gatekeeping analysis in the first appeal brought in a case like Barabin, for example, the district court could conclude that the evidence was in fact admissible. In that event, the losing party might challenge that finding on appeal, resulting in the case's coming before the court of appeals yet again. (46) And if the court of appeals concludes on that second hearing that the district court erred in finding admissibility, the court of appeals could then remand for a new trial. The results of that trial would themselves be appealable, so the case might go up to the court of appeals for a third time. Remanding for a new trial in the first instance--instead of ordering a limited remand--seems far more likely to generate only a single second appeal instead of leaving the door open to the third appeal that might be required to resolve a case in which the court chooses a limited remand.
Shortly after the en banc decision in Barabin, a panel of the Ninth Circuit reversed United States v. Christian, remanding for a new trial because the district court had not performed its Daubert gatekeeping obligation. But the court was careful to "emphasize that neither Barabin nor this decision requires a new trial whenever a district court errs in analyzing the admissibility of expert testimony." (48) The court observed that "under different circumstances ... a limited remand remains available." (49) Read together, Christian, StorageCraft, and other similar decisions show that there is no need for imposing a rigid limited-remand rule when a district court fails to make the appropriate threshold determination of admissibility. The current analytical framework is functioning as Congress intended, allowing the federal courts of appeals to make case-specific decisions.
(1.) 28 U.S.C. [section] 2111 (2015) (providing that "the court shall give judgment after an examination of the record without regard to errors or defects which do not affect the substantial rights of the parties"), available at http://uscode.house.gov.
(2.) 28 U.S.C. [section] 2106 (2015) (providing that the court "may affirm, modify, vacate, set aside or reverse any judgment, decree, or order ... and ... may remand the cause and direct the entry of such appropriate judgment, decree, or order, or require such further proceedings to be had as may be just under the circumstances"), available at http://us code.house.gov.
(3.) See Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 594-97 (1993) (discussing admission of scientific and technical evidence; recognizing gatekeeper role of trial judge, who is charged with determining whether expert testimony rests on necessary foundation and is relevant to matters at issue; and noting that standard of admissibility must be flexible).
(4.) 740 F.3d 457, 460 (9th Cir.) (en banc), cert, denied, _U.S. _, 135 S. Ct. 55 (2014).
(5.) Id. at 465 (discussing court's procedure in two earlier cases, noting specifically that "we engaged in harmless error review, found that the error was not harmless, and remanded for a new trial" in one case and "went straight to harmless error review, found the evidence to be prejudicial, and remanded for a new trial" in the other) (internal citations omitted).
(6.) Id. at 467.
(7.) Id. at 465-67.
(8.) See Pet. for Writ of Certiorari at 10-18, Barabin v. Asten Johnson, Inc., No. 13-1252, 2014 WL 1494074 (Apr. 15, 2014).
(9.) Id. at 10, 18-21.
(10.) See United States v. Edwards, 728 F.3d 1286, 1296-97 (11th Cir. 2013) (observing that under [section] 2106, courts of appeals have broad discretion to grant relief as may be just under the circumstances).
(11.) 28U.S.C. [section]2106.
(12.) GuideOne Specialty Mut. Ins. Co. v. Missionary Church of Disciples of Jesus Christ, 687 F.3d 676, 682 n.3 (5th Cir. 2012) (citing [section] 2106); see also U.S. Aviation Underwriters, Inc. v. Pilatus Bus. Aircraft, Ltd., 582 F.3d 1131, 1145 (10th Cir. 2009) ("When a district court enters a judgment with a legally insufficient basis, this court has three options: (1) order the entry of judgment as a matter of law, (2) order a new trial, or (3) remand to the district court to decide the appropriate option."); Smith v. Washington Sheraton Corp., 135 F.3d 779, 783 (D.C. Cir. 1998) ("Having concluded that the district court erred in denying [defendant's] motion for judgment as a matter of law, we have three choices. We may enter judgment for that party, or we may order a new trial, or we may remand the case to the district court to determine whether a new trial is appropriate.").
(13.) Samuels v. Health & Hosp. Corp. of N.Y., 591 F.2d 195, 199 (2d Cir. 1979) (noting that "an appellate court has the authority to grant a new trial in the interest of justice").
(14.) Ruiz-Troche v. Pepsi Cola of Puerto Rico Bottling Co., 161 F.3d 77, 88 (1st Cir. 1998) (citing [section] 2106).
(15.) See id.
(16.) See, e.g., United States v. Johnson, 388 F.3d 96, 102-03 (3d Cir. 2004) (remanding with instructions to district court to undertake weighing analysis); United States v. Downing, 753 F.2d 1224, 1244 (3d Cir. 1985) (remanding for evidentiary hearing).
(17.) See, e.g., United States v. Sriyuth, 98 F.3d 739, 744 n.8 (3d Cir. 1996) (pointing out that a district court's failure to make proper threshold admissibility determination "may require remand to the court for such proceedings or even for a new trial").
(18.) United States v. Smithers, 212 F.3d 306, 315 (6th Cir. 2000).
(19.) Id. at 317.
(20.) Id. at 318 & n.6.
(21.) United States v. Hall, 93 F.3d 1337, 1341 (7th Cir. 1996).
(22.) Id. at 1342 (noting that "we cannot be confident that the district court applied the Daubert framework," because "[t]he judge never mentioned Daubert specifically, and thus he never focussed [sic] on the individual questions that must be answered").
(23.) Id. at 1346.
(24.) See United States v. Lang, 8 F.3d 268, 271 (5th Cir. 1993) (pointing out that "the record indicates that the district court, at most, made only an initial determination ... and left the ultimate determination ... to the jury").
(25.) LeBlanc v. Chevron USA, Inc., 275 F. App'x 319, 321 (5th Cir. 2008).
(26.) Coleman v. United States, 397 F.2d 621, 621 (D.C. Cir. 1966) (exercising discretion under [section] 2106 to decide against remand and explaining that "a remand for hearing on the issue of admissibility alone [was] inappropriate" because "the trial judge's reasons for refusing to resolve the admissibility issue were highly prejudicial").
(27.) See, e.g., Barabin, 730 F.3d at 456 (indicating that remand for new trial can be appropriate "[w]hen the district court has erroneously admitted or excluded prejudicial evidence," and also when "the district court erred by failing to answer a threshold question of admissibility"); Kosty v. Lewis, 319 F.2d 744, 749 (D.C. Cir. 1963) (noting that [section] 2106 "grants this Court broad discretion in the disposition of a case on appeal"); Morgan Guaranty Trust Co. of N.Y. v. Martin, 466 F.2d 593, 600 (7th Cir. 1972) (quoting Kosty).
(28.) Shinseki v. Sanders, 556 U.S. 396, 407 (2009); see also Kotteakos v. United States, 328 U.S. 750, 761 (1946) (observing that proper application of the harmless error rule requires "judgment transcending confinement by formula or precise rule").
(29.) Shinseki, 556 U.S. at 407 (2009).
(30.) Id. at 411.
(31.) Id. at 411-12.
(32.) Id. at 412.
(33.) Id. at 410.
(34.) Cf. Downing, 753 F.2d at 1226, 1243 (refusing to hold that district court's error in failing to make threshold admissibility determination was harmless). It bears noting in this connection that an error is prejudicial--and so not harmless--when there is a reasonable probability that it affected the outcome of the trial. United States v. Marcus, 560 U.S. 258, 262 (2010). And if the testimony is critical to the plaintiffs case, the error plainly affects the outcome of the proceeding. Messner v. Northshore Univ. Health Sys., 669 F.3d 802, 812-14 (7th Cir. 2012) (holding that district court's failure to perform Daubert gatekeeping analysis was harmful error because expert testimony played substantial role in district court's class-certification ruling).
(35.) Shinseki, 556 U.S. at 411-12.
(36.) Id. at 410.
(37.) See id. at 407-08.
(38.) Smithers, 212 F.3d at 317
(39.) Hall, 93 F.3d at 1346.
(40.) Smith v. Jenkins, 732 F.3d 51, 65, 68-69 (1st Cir. 2013).
(41.) StorageCraft Tech. Corp. v. Kirby, 744 F.3d 1183, 1990-92 (10th Cir. 2014) (analyzing district court's analysis and concluding that "more words from the district court would not have altered the admissibility of the expert's evidence"). Even before StorageCraft, the Tenth Circuit had held that a district court's error in failing to make Daubert findings before admitting expert testimony was harmless because the jury could properly have found the defendant guilty even without the expert's testimony. United States v. Roach, 582 F.3d 1192, 1208 (10th Cir. 2009). Thus, a new trial is unnecessary when either (1) it is readily apparent from the record that the expert testimony was admissible, or--when the district court excluded evidence without making the necessary threshold determination--inadmissible, or (2) other competent evidence is sufficiently strong to permit the conclusion that the improper evidence had no effect on the decision. StorageCraft, 744 F.3d at 1191.
(42.) United States v. Smith, 27 F. App'x 577, 582 (6th Cir. 2001) (indicating that there was "overwhelming and diverse evidence" supporting the appellants' convictions).
(43.) United States v. Jawara, 474 F.3d 565, 583 (9th Cir. 2006) (opining that "[t]he lack of an explicit finding of reliability was harmless," and citing earlier Ninth Circuit cases to the same effect); see also United States v. Blaine County, 363 F.3d 897, 915 (9th Cir. 2004) (holding that "the district court made the necessary reliability determination with respect to [one expert's] testimony and report," and also holding that district court's failure to make determination for other expert testimony was harmless because outcome was supported by evidence found to be reliable); United States v. Williams, 29 F. App'x 486, 487 (9th Cir. 2002) (holding that district court's failure to perform Daubert gatekeeping duty was harmless error).
(44.) See Shinseki, 556 U.S. at 407 (cautioning against use of "rigid rules").
(45.) An analysis of the effects of these factors on the federal courts of appeals is beyond the scope of this article, but other commentators continue to consider and write about them. See, e.g., Andrew Adler, Extended Vacancies, Crushing Caseloads, and Emergency Panels in the Federal Courts of Appeals, 15 J. APP. PRAC. & PROCESS 163 (2014).
(46.) Cf United States v. Holloway, 740 F.2d 1373 (6th Cir. 1984) (entertaining appeal after remand for district court's threshold admissibility determination involving application of co-conspirator exception to hearsay rule).
(47.) 749 F.3d 806 (9th Cir. 2014).
(48.) Id. at 813 n.3 (emphasis in original).
(49.) Id. at 814 n.3 (noting that a limited remand on the question of prejudice might on occasion be proper).
Robert B. Gilbreath, Mr. Gilbreath is a partner at Hawkins Parnell Thackston & Young LLP in Dallas, where he serves as Appellate and Legal Issues Practice Group Leader.
|Printer friendly Cite/link Email Feedback|
|Author:||Gilbreath, Robert B.|
|Publication:||Journal of Appellate Practice and Process|
|Date:||Mar 22, 2015|
|Previous Article:||A practical guide to appellate judging.|
|Next Article:||A past and future of judicial elections: the case of Montana.|