Concept
Reliable Principles and Methods
Reliability and relevancy
Reliable Principles and Methods
Rule 702(c) adds a condition that no résumé can satisfy: the testimony must be “the product of reliable principles and methods.”1 Qualification is a separate question. It asks whether the witness has sufficient “knowledge, skill, experience, training, or education,” and a witness can meet that test without being the foremost authority on the subject, without understanding every nuance of it, and without ever having published. Most expert witnesses do not publish.2 A credential describes the person. A method describes the path from the facts to the opinion, and a well-credentialed witness can still travel a bad path. Subsection (c) therefore sends the judge to what the expert did, not to who the expert is.
A judge confronting contested science has more than one way to respond. She can admit the evidence because the experts are qualified and let the jury resolve the dispute; she can decide for herself whether the proof is valid science; or she can defer to the scientific community and ask the proponent to show that scientists generally agree with the proof, which is the approach Frye most clearly resembles.3 The first option treats qualification as enough. Subsection (c) does not allow that, because it makes reliability a condition of its own.
In Daubert, the Supreme Court gave that condition content by reading the phrase “scientific knowledge” in Rule 702. Knowledge, the Court said, does not mean absolute certainty, but an inference or assertion qualifies as scientific knowledge only if it is derived by the scientific method, and the rule is satisfied when the testimony is based on scientifically valid principles.4 The word also excludes guesswork. To constitute knowledge, the methodology must rest on more than subjective belief or unsupported speculation, and to be reliable in a scientific sense it must be empirically validated.5 The Court gave trial judges factors for assessing that validation, among others: whether the methodology is testable and has been tested, whether it has been peer reviewed and published, whether it has a known error rate, whether accepted standards govern its use, and whether it is generally accepted.6 Each factor receives its own section below.
These requirements move the judge’s attention from the expert’s conclusions to the basis for them. As one court applying Daubert put it, the task is “to analyze not what the experts say, but what basis they have for saying it.”7 The expert’s own assurance cannot supply that basis. A bald assurance of validity is not enough: the party presenting the expert must show that the expert’s findings rest on sound science, and that will require some objective, independent validation of the methodology.8
Kumho Tire shows how much can ride on that showing. The right rear tire of a minivan blew out, one passenger died and others were severely injured, and the Carmichaels sued the tire’s maker and distributor, resting their case in significant part on the deposition testimony of Dennis Carlson, Jr., an expert in tire failure analysis.9 In that case and in Joiner, the Court stated that an expert must present more than ipse dixit reasoning, meaning an assertion that rests on the speaker’s say-so alone. An opinion offered merely because the witness has relevant education or experience is exactly that.10
The inquiry is also specific to the method at hand. In Joiner, the respondent never explained how and why the experts could extrapolate their opinions from seemingly far-removed animal studies. Instead, the respondent proceeded as if the only issue were whether animal studies could ever be a proper foundation for an expert opinion. That was not the issue. The issue was whether these experts’ opinions were sufficiently supported by the animal studies on which they purported to rely.11 A proponent who defends the respectability of a field has answered a question the judge never asked.
Consider a witness whose credentials are not in doubt.
Change the facts. Suppose a forensic chemist identifies a seized powder with a named instrumental technique, explains how the technique works, and shows that the laboratory has run it on samples of known composition and that it identified them correctly. The two witnesses may hold identical credentials. What separates them is that only the chemist has handed the judge a method together with the kind of empirical validation the proponent must show.12
Nonscientific testimony does not escape the requirement, though the tools change. One account, taken up in a later section, holds that the proponent must still demonstrate the reliability of the expert’s underlying assumptions, but that the judge need not use the Daubert factors, which derive from a classic scientific model. The judge may instead ask more flexible questions, such as whether members of the public routinely turn to this profession for services other than testimony and whether a feedback loop alerts a member of the profession when she has erred.13
Focusing on method does not cut the method entirely loose from the conclusion. Conclusions and methodology are not entirely distinct, and trained experts commonly extrapolate from existing data. Yet nothing in Daubert or the Federal Rules of Evidence requires a court to admit opinion evidence connected to the data only by the expert’s ipse dixit, and a court may conclude that too great an analytical gap separates the data from the opinion.14 The section on Joiner develops that gap. For now, the working rule is that a qualified expert must still show a method, and the judge examines the basis for the opinion rather than its correctness.