|

The 2023 Amendment to Federal Rule of Evidence 702: The Inside Story

by

Vol. 108 No. 2 (2024) | Judges Under Siege? | Download PDF Version of Article

In the Federal Rules of Evidence, the predominant rule on expert testimony is Rule 702. On December 1, 2023, the rule was amended after six years of work by the Judicial Conference Advisory Committee on Evidence Rules, for which I have served as reporter since 1996.1 Here I provide an account of how the amendment came to be, in the hope that this “insider’s view” will help the reader better understand the amendment’s goals — and how the rulemaking process actually works.

The amendment to Rule 702, in my view, accomplishes three goals: 1) It emphasizes that the reliability requirements of the rule must be established by a preponderance of the evidence, rejecting lower court case law that held that defects in reliability were matters of weight and not admissibility; 2) It requires the court to closely assess whether the expert’s opinion reflects a reliable application of a reliable methodology, and to exclude the opinion if it overstates what the methodology supports; and 3) The Committee Note to the amendment focuses on forensic evidence and suggests that overstatements (such as testimony about a “zero rate of error”) are impermissible.

Genesis of the Change

The road to the amendment began with a phone call that I received in 2016 from Dr. Eric Lander, Co-Chair of a project sponsored by the President’s Council of Advisors on Science and Technology (PCAST). Dr. Lander told me that PCAST was reviewing forensic testimony offered in federal and state courts, and asked me whether the Advisory Committee could write a Committee Note that would set forth standards to assure that forensic testimony would be valid and not overstated. I informed him that, by statute, Advisory Committee Notes can be written only as part of an actual rule amendment.2 He asked whether the Committee could “change a comma” in Rule 702 and use that as an excuse to write a lengthy Committee Note providing reliability requirements for forensic evidence. I told him that this was also not possible because the policy of the Rules Committee is that Committee Notes are to explain and amplify textual changes, and not to set wide-ranging standards beyond the text.

I was nevertheless convinced by the PCAST report that something needed to be done about the sad state of forensic expert testimony,3 and that an amendment could impose useful safeguards on admissibility and overstatement. Because the Advisory Committee on Evidence Rules is not composed of scientists, it was clear that any venture into forensics required background research and input from experts on forensics (and its flaws). So the Committee convened a symposium in 2017.4 Participants included Dr. Lander; a bevy of statisticians; others who worked on the PCAST project; Federal Judges Jed Rakoff, Alex Kozinski, and Patti Saris; legal scholars; and representatives of the Department of Justice (DOJ). I found that the symposium (together with the PCAST report) established the following:

  1. Virtually all of the forensic disciplines are inherently subjective, with fundamental premises that have not been established empirically, and with no indication of a rate of error.
  2. Forensic experts routinely overstate their conclusions. That is, they are allowed by courts to state their conclusions to a degree of certainty that is not supported by the methodology.5

At a meeting after the Symposium, the Advisory Committee resolved to consider the possibility of an amendment to the Evidence Rules that would regulate the admissibility of forensic expert testimony. At that meeting I also informed the Committee of a recent William & Mary Law Review article6 that flagged a different concern about Rule 702: Federal courts were ignoring some of the admissibility requirements that had been added to Rule 702 in 2000 — specifically, that the expert’s opinion must be based on sufficient facts or data (Rule 702(b)) and that the expert must have reliably applied reliable principles and methods to the facts of the case (Rule 702(d)). As the article described, many courts have stated that defects in an expert’s basis, and misapplication of methodology, are questions of weight and not admissibility.7 In fact, because they are admissibility requirements, the proponent must show them by a preponderance of the evidence. Essentially, many courts had treated these requirements under Rule 104(b) of the Federal Rules (a standard of “sufficient to support a finding”) as opposed to Rule 104(a) (a preponderance of the evidence).

After that meeting, the Chair of the Evidence Rules Committee established a Subcommittee to investigate the possibility of amending the Rules to address both the problems of forensics and the problem of courts misapplying the Rule 702 admissibility requirements. What follows is a discussion of the Committee’s work on both fronts.

Addressing Forensics Issues

A New Rule [rejected]. With the help of Dr. Lander, I drafted a rule on forensics that I proposed as a new Rule 707. I did not attempt to add forensic-specific limitations to Rule 702, since it is obviously intended to be a rule of general applicability. Instead, I proposed as follows:

Rule 707. Testimony by Forensic Expert Witnesses. If a witness is testifying on the basis of a forensic examination the proponent must prove the following in addition to satisfying the requirements of Rule 702:

(a) the witness’s method is repeatable, reproducible, and accurate for its intended use — as shown by empirical studies conducted under conditions appropriate to that use;

(b) the witness is capable of applying the method reliably — as shown by adequate empirical demonstration of proficiency — and actually did so; and

(c) the witness accurately states, on the basis of adequate empirical evidence, the meaning of any similarity or match between the evidentiary sample and the source sample.

The Subcommittee, and the Committee thereafter, rejected the proposed amendment. It found the following problems with the proposal:

  1. How does it relate to the Rule 702 requirements, which still apply? There is obvious overlap between Rule 702(d)’s requirement of reliable application and proposed Rule 707(b). Courts would likely have difficulty parsing these differences, and the probable result would be that many courts would just default back to the Rule 702 standards.
  2. How is “forensic” defined? Was it limited to feature-comparison forensics, in which the features of an evidentiary sample are compared to a control (e.g., fingerprints, shoeprints), which was the focus of the PCAST report? Or did it also include other forensic methods, such as arson detection, which are problematic as well? Notably, the PCAST report does not define the term “forensic,” and the word was nowhere in the then-existing Evidence Rules.
  3. Why pick on forensics? It’s fair to say that opinions from a number of questionable disciplines have been admitted under Rule 702. Experts on human factors and linguistics patterns come to mind. Because of the institutional interests supporting forensic disciplines (i.e., the forensic expert community and the DOJ), the optics of singling out forensics as the only expertise with a specific rule addressed to it are not optimal.

Besides rejecting a separate Rule 707, the Subcommittee, and then the Committee, went on to reject a number of other suggestions with regard to forensics.

A Science-Focused Committee Note [rejected]. First, it rejected suggestions that the Advisory Committee should draft a lengthy Committee Note that would set forth detailed scientific standards for forensic disciplines. Such a Note would go well beyond whatever textual change could be made to the rule. The Note would require significant scientific input and could run into the same controversies regarding sources and standards that arose with PCAST.8 And it would run the risk of becoming outmoded by scientific developments and developing forensic disciplines.

A Best Practices Manual [rejected]. Second, it rejected the PCAST report’s proposal that the Committee prepare a best practices manual. Such a manual could not have been published as a work of the Committee, whose authority is limited to rulemaking, and a Committee-sponsored Best Practices Manual could look like an attempt to establish Evidence Rules outside the rulemaking process. And a work by individual authors (e.g., the Reporter and Committee members) would not be as influential as a work issued by the Advisory Committee. Moreover, such a manual would be difficult to write — requiring a dip into the deep end of the science pool, treading in an area of high controversy. Finally, the Committee recognized the unfortunate fact that most courts have generally not budged from precedent in the face of reports from the National Academy of Sciences and PCAST. So why would a best practices manual have any effect?

A Prohibition on Overstatement [rejected]. An opinion that is qualified, even if properly so, is less impactful on the fact finder.9 Thus, among both forensic and nonforensic experts, overstatement is common. The Subcommittee recommended further consideration of an amendment to Rule 702 that would prohibit experts from overstating their opinions. While overstatement is endemic in the forensic field, any rule on overstatement lodged broadly in Rule 702 would necessarily apply to all experts. That said, it seemed obvious to me that while the Committee was not in a position to crunch data like PCAST did, it was in a position to regulate experts who overpromise in their conclusions. Consequently, I drafted the following amendment for the Committee’s consideration (proposed new text in italics):

A witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if the proponent demonstrates by a preponderance of the evidence that:

(a) the expert witness’s scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue;

(b) the testimony is based on sufficient facts or data;

(c) the testimony is the product of reliable principles and methods;

(d) the expert has reliably applied the principles and methods to the facts of the case; and

(e) the witness does not overstate the conclusions that reasonably may be drawn from a reliable application of the principles and methods.

The overstatement amendment was a tough sell — even after I submitted a case law digest indicating that overstatement by a forensic expert could be found in about 25 percent of the reported federal criminal cases. The major opposition was, unsurprisingly, DOJ. It sent a letter to the Chair recommending that the overstatement amendment should be delayed (meaning killed) because the Department was promulgating Uniform Language for Testimony and Reports (ULTRs) that are designed to limit overstatement by forensic experts. As the Department put it: “The issue can be monitored in future meetings, and an amendment re-visited in the future if deemed necessary.” Monitoring and studies are where rule amendments go to die.

The ULTRs published by the Department prohibit an expert subject to them from testifying to a zero rate of error, or to a conclusion to a “reasonable degree of certainty” — a meaningless and confusing term. Despite the DOJ’s apparent good faith, it is pretty apparent that the ULTRs are not the solution to the problem of overstatement. This is so for a number of reasons:

  • Leaving protections up to DOJ means that failure to comply would not be actionable — even though the result might be an unjust conviction or a guilty plea that would not otherwise have been entered.
  • The Department’s reforms, as salutary as they are, would not affect overstatement by experts outside the purview of DOJ. For example, in many federal cases, ballistics testimony is provided by state experts.
  • Administrations and objectives change, and there is no guarantee that the Department’s protocols will remain in place — administrations change, objectives change, and nobody has a right to enforce an existing DOJ protection. With an amendment to Rule 702, there is a pretty strong guarantee that limitations on overstatement will remain in place.
  • Most importantly, the ULTRs allowed the expert to testify to a “source identification.”10 The distinction between a “source identification” and a “match” was explained to me at four separate meetings, and I never got it. If a relatively sane person with 40 years of evidence experience cannot understand the difference between a “source identification” and a “match,” then how could a juror? How could a juror understand the statement: “I am making a source identification, but not to the exclusion of all other sources?” Ultimately, a “source identification” is an overstatement when the forensic opinion is reached by a subjective process prone to error. The DOJ Uniform Language attempts to walk a fine line between allowing the forensic expert to testify to identity of the source of a crime scene sample, and disavowing any certainty that this is in fact the case. But there is no line there to walk. According to the Uniform Language, a “source identification” of, say, a toolmark means only that the examiner has seen sufficient pattern agreement to “provide extremely strong support for the proposition that the two toolmarks came from the same source and extremely weak support for the proposition that the two toolmarks came from different sources.” While this sounds as though the strength of the evidence is based on a statistical assessment, the Uniform Language makes clear that this is merely the examiner’s opinion and has no statistical foundation.

Nevertheless, DOJ’s arguments surely persuaded some Committee members that an amendment on overstatement was unnecessary. Clearly the Department has an outsized effect on the rulemaking process. Since 1996, only one amendment to the Federal Rules of Evidence has been adopted over the Department’s objection: the 2000 amendment to Rule 701, which sought to clarify the line between lay and expert testimony.

There were other objections to an across-the-board limit on overstatement. There did not appear to be disagreement with the premise that experts other than in forensics often overstate their opinion. Indeed experts are encouraged to state their opinion as definitively and superlatively as possible. An opinion that is qualified, even if properly so, is less impactful on the fact finder.11 Rather, the Committee’s concerns were about the need for an amendment barring overstatement, and the problems that amendment might create.

As to need, some members believed that overstatement was already prohibited by the rule’s existing requirement that reliable methodology must be reliably applied. A forensic expert who claims a zero rate of error should be excluded under the existing rule because the subjective methodology cannot support that conclusion — thus the methodology has not been reliably applied. This is a legitimate point, but the rejoinder is that overstatement is happening under the existing rule, so it would make sense to add something more specific in the rule to control the problem.

But the major concern was that a specific limitation on overstatement, applied beyond forensics, could end up in wordsmithing controversies for virtually every opinion. For example, if an accidentologist testifies that the accident was “definitely caused by the defendant veering out of his lane,” is “definitely” an overstatement? Do we need to get a thesaurus out? Comments like this dominated the discussion in several meetings. The Committee also received unsolicited letters from both plaintiffs and defendants groups that expressed concern about an overstatement amendment improperly intruding into an expert’s choice of words.

Still, a minority of Committee members, most notably the public defender, continued to favor a limitation on overstatement in the amendment. And these members pointed out that, with no textual change, any opportunity to express concerns about forensic overstatement in the Committee Notes would be lost.

The Outcome: Revisions to Rule 702 and a Committee Note. Eventually, a compromise was reached. Judge Thomas Schroeder, who chaired the Subcommittee, suggested that the existing Rule 702(d) could be tweaked to indicate that a court should focus on excluding clearly overstated testimony from any expert.11 The Committee ultimately approved the following slight change to Rule 702(d):

(d) the expert has reliably applied expert’s opinion reflects a reliable application of the principles and methods to the facts of the case.

The goal of the change was to encourage courts to focus directly on the opinion, to see if it is sufficiently tied to the methods used; the focus is not on the application per se, but on an assumption that if the opinion itself seems to go further than the methodology allows, then there is a problem in the application that requires exclusion.

This change to the text then justified a passage in the Committee Note to be directed at forensic evidence. It’s more than a comma. And while the amendment applies to all expert testimony, there is nothing wrong, and everything right, in letting the reader know the most important problem to which the text is directed.

The Committee Note language was worked out, word for word, between DOJ, Judge Patrick Schiltz, and myself. Judge Schiltz was especially effective in crafting the Note so that it would regulate overstatement of forensic experts. The paragraphs ultimately adopted are as follows:

Rule 702(d) has also been amended to emphasize that each expert opinion must stay within the bounds of what can be concluded from a reliable application of the expert’s basis and methodology. Judicial gatekeeping is essential because just as jurors may be unable, due to lack of specialized knowledge, to evaluate meaningfully the reliability of scientific and other methods underlying expert opinion, jurors may also lack the specialized knowledge to determine whether the conclusions of an expert go beyond what the expert’s basis and methodology may reliably support.

The amendment is especially pertinent to the testimony of forensic experts in both criminal and civil cases. Forensic experts should avoid assertions of absolute or one hundred percent certainty — or to a reasonable degree of scientific certainty — if the methodology is subjective and thus potentially subject to error. In deciding whether to admit forensic expert testimony, the judge should (where possible) receive an estimate of the known or potential rate of error of the methodology employed, based (where appropriate) on studies that reflect how often the method produces accurate results. Expert opinion testimony regarding the weight of feature comparison evidence (i.e., evidence that a set of features corresponds between two examined items) must be limited to those inferences that can reasonably be drawn from a reliable application of the principles and methods. This amendment does not, however, bar testimony that comports with substantive law requiring opinions to a particular degree of certainty.

In sum, as to forensics, the pertinent textual amendment is subtle. And subtle rulemaking is problematic. But I am convinced that the result reached was the best possible one under the circumstances, and that courts can rely on the textual change and the Committee Note to further the important goal of preventing forensic experts from overstating their opinions.

Addressing Admissibility/Weight Issues

Rule 702 as amended back in 2000 set forth admissibility requirements. The intent, as expressed in the Committee Note, was that the proponent would have to satisfy those requirements by a preponderance of the evidence.12 Yet many courts stated plainly that questions of sufficient facts or data and proper application were matters of weight and not admissibility.

Confusion about the Standard. Why this failure to follow the rule? Even though the 2000 amendments came seven years after Daubert, that case remained good law, and those more inclined to look at Daubert could find the language that they wanted there, because there is language in Daubert for everyone. If you like your expert rule to be liberal, you rely on the Daubert quotes that the remedy for “shaky but admissible”13 evidence is not exclusion but cross-examination, and that the Federal Rules have a “liberal thrust.”14 If you like it strict, then you note that the Court specifically laid the gatekeeper requirement in Rule 104(a) of the Federal Rules — and that rule imposes a preponderance of the evidence requirement.15

The confounding issue for the Committee was what to do when courts fail to apply an amendment. Should there be another amendment to the effect of “we meant what we said”? In this case, that would mean adding a preponderance of the evidence requirement to the text of the rule itself. But the Committee was concerned that including such a requirement in text would raise concerns about virtually every other evidence rule, to which the preponderance standard already applied via Rule 104(a). Why make explicit in one rule what is implicit in virtually every other rule? Would parties try to argue that the preponderance standard did not apply to those other rules because it was not explicit in the text?

Eventually, the Committee decided that adding a preponderance standard to the text of Rule 702 would be valuable, and would not just be “we really mean what we said.” Judge Schiltz pointed out to the Committee that while the preponderance standard already applied, that was not immediately obvious. There is a brief statement to that effect in the very long Committee Note to the 2000 amendment. It is stated in a footnote in Daubert. That footnote refers to a prior case holding that Rule 104(a) requires a preponderance.16 But then if you look to the actual text of Rule 104(a) there is nothing at all about a standard of proof; it simply says that the judge in deciding questions of admissibility is not bound by the rules of admissibility.

The Outcome: Revisions to Rule 702 and a Committee Note. Since the preponderance standard was perhaps not obvious, and courts were getting it wrong, the Committee decided to place it in the text of Rule 702 itself. Accordingly, the rule was amended as follows:

A witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if the proponent demonstrates to the court that it is more likely than not that:

(a) the expert’s scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue;

(b) the testimony is based on sufficient facts or data;

(c) the testimony is the product of reliable principles and methods; and

(d) the expert has reliably applied expert’s opinion reflects a reliable application of the principles and methods to the facts of the case.

The concern about other rules without the standard in text was specifically addressed in the Committee Note, which states:

There is no intent to raise any negative inference regarding the applicability of the Rule 104(a) standard of proof for other rules. The Committee concluded that emphasizing the preponderance standard in Rule 702 specifically was made necessary by the courts that have failed to apply correctly the reliability requirements of that rule. Nor does the amendment require that the court make a finding of reliability in the absence of objection.

Much of the public comment on the proposed amendment was directed to applying the preponderance standard to expert testimony, and the requirement was severely criticized by the plaintiff’s bar as being too high a threshold. Interestingly, though, the negative comments actually supported the need for the rule. For example, many comments claimed the amended rule would change existing law, which presumed expert testimony reliable. This is the very misunderstanding that the amendment is designed to remedy.

The defense bar strongly suggested that the Committee Note include citations to those cases that had misapplied the admissibility requirements of the rule — essentially to call out wrongdoing courts. The Committee rejected that idea because it is not its job to review individual cases for accuracy, and it is unnecessary to attack specific judges. Moreover, the Committee did not have access to all the necessary information. For example, a court might say that a defect in basis is a matter of weight and not admissibility. But in actuality, the court may have found sufficient basis by a preponderance of the evidence. The Committee found it enough to add the following to the Committee Note:

First, the rule has been amended to clarify and emphasize that expert testimony may not be admitted unless the proponent demonstrates to the court that it is more likely than not that the proffered testimony meets the admissibility requirements set forth in the rule. See Rule 104(a). This is the preponderance of the evidence standard that applies to most of the admissibility requirements set forth in the evidence rules. See Bourjaily v. United States, 483 U.S. 171, 175 (1987) (“The preponderance standard ensures that before admitting evidence, the court will have found it more likely than not that the technical issues and policy concerns addressed by the Federal Rules of Evidence have been afforded due consideration.”); Huddleston v. United States, 485 U.S. 681, 687 (1988) (“preliminary factual findings under Rule 104(a) are subject to the preponderance-of-the-evidence standard”). But many courts have held that the critical questions of the sufficiency of an expert’s basis, and the application of the expert’s methodology, are questions of weight and not admissibility. These rulings are an incorrect application of Rules 702 and 104(a).

Time will tell if this is the right choice. If those offending cases continue to be relied upon, then maybe the lesson for next time is to say: “These specific cases are dead wrong and hereby overruled” — with the hope that nobody on the Rules Committee wrote one of those opinions.

Conclusion

The amendment to Rule 702 was the culmination of six years of effort. It provides a mechanism to regulate forensic testimony and clarifies an important question about the applicable standard of proof for the reliability requirements of the rule. The rulemaker’s lament is that you never know whether courts and litigants will follow the changes that are made.17 All you can do is leave them the tools to do so. And that is what the Committee did.


Daniel J. Capra is the Reed Professor of Law at Fordham University School of Law, where he teaches Evidence. He also teaches at NYU Law School. Professor Capra has served for the last 28 years as the Reporter to the Judicial Conference Advisory Committee on Evidence Rules. 


 

  1. I gratefully acknowledge Professor Liesa Richter, academic consultant to the Advisory Committee, for all the stellar work she did on the amendment to Rule 702.
  2. 28 U.S.C. § 2073(d) (stating that a rules committee’s recommendation “shall provide a proposed rule, [and] an explanatory note on the rule . . .”).
  3. See President’s Council of Advisors on Sci. & Tech., Exec. Office of the President, Forensic Science in Criminal Courts: Ensuring Scientific Validity of Feature-Comparison Methods (2016).
  4. See Symposium on Forensic Testimony, Daubert, and Rule 702, 86 Fordham L. Rev. 1463 (2018).
  5. See, e.g., United States v. Casey, 928 F. Supp. 2d 397, 399–400 (D.P.R. 2013) (ballistics expert states that he was 100 percent certain of a match; the court concluded that it would remain “faithful to the long-standing tradition of allowing the unfettered testimony of qualified ballistics experts”); United States v. Taylor, 663 F. Supp. 2d 1170, 1180 (D.N.M. 2009) (holding that expert could testify that the bullet came from suspect rifle to within “reasonable degree of certainty in the firearms examination field”); United States v. Mahone, 453 F.3d 68, 71–72 (1st Cir. 2006) (no error in admitting testimony that footwear-impression identification methodology had a zero error rate).
  6. David E. Bernstein & Eric G. Lasker, Defending Daubert: It’s Time to Amend Federal Rule of Evidence 702, 57 Wm. & Mary L. Rev. 1 (2015).
  7. Id. at 19–30.
  8. See U.S. Dep’t of Just., Statement on the PCAST Report: Forensic Science in Criminal Courts: Ensuring Scientific Validity of Feature-Comparison Methods (2021), https://www.justice.gov/olp/page/file/1352496/dl.
  9. As Judge Lewis Kaplan explains,

    [l]awyers want experts who will express unwavering certainty about their conclusions: Eighty-four percent of lawyers surveyed in a recent study said that the adamancy of an expert’s support for the lawyer’s position was an important consideration in the expert selection process. Experts are well aware of this overwhelming preference. The same study showed that sixty-four percent of experts believe that the willingness to draw firm conclusions was important to being retained. The desire to please lawyers often leads experts to overstate the certainty of their conclusions and to gloss over important nuances in an effort to present the most uncompromising support for the lawyers’ position.

    Hon. Lewis A. Kaplan, Experts in the Courthouse: Problems and Opportunities, 2006 Colum. Bus. L. Rev. 247, 250 (2006).

  10. See Uniform Language for Testimony and Reports, U.S. Dep’t of Just. (Mar. 6, 2024), https://www.justice.gov/olp/uniform-language-testimony-and-reports.
  11. A similar proposal was made by Judge Caroline Kuhl, a member of the Judicial Conference Rules Committee, who provided outstanding service as liaison from that Committee to the Advisory Committee on Evidence Rules.
  12. See FED. R. EVID. 702 Advisory Committee Note to 2000 Amendment (“Consequently, the admissibility of all expert testimony is governed by the principles of Rule 104(a). Under that Rule, the proponent has the burden of establishing that the pertinent admissibility requirements are met by a preponderance of the evidence.”).
  13. Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 596 (1993).
  14. Id. at 588.
  15. Id. at 592–93, 592 n.10 (“These matters should be established by a preponderance of proof.” (citing Bourjaily v. United States, 483 U.S. 171, 175–76 (1987))).
  16. See Bourjaily, 483 U.S. at 175–76.
  17. See generally Richard Marcus, The Rulemakers’ Laments, 81 Fordham L. Rev. 1639 (2013).