Admissibility: Frye, Daubert and Cross-Jurisdiction Rules

In 1923 the US Court of Appeals for the DC Circuit was asked whether the results of an early systolic blood pressure deception test, a forerunner of the polygraph, could be introduced at trial. The court declined, holding that a novel scientific technique must have crossed the line from experimental to demonstrable by gaining 'general acceptance in the particular field in which it belongs'. The case was Frye v United States and it produced a standard that American courts used for the next seventy years and that a significant number still apply today.

The logic of the Frye test is conservative: it defers to the scientific community rather than expecting a judge to evaluate scientific methodology independently. If the community of practitioners in a field broadly uses and accepts a technique, courts applying Frye will admit evidence based on it. The community acts as an informal quality filter. This has the advantage of keeping genuinely fringe science out of court without requiring the judge to become a scientist.

In Daubert v Merrell Dow Pharmaceuticals (1993), the US Supreme Court addressed whether statistical epidemiological evidence about whether a drug caused birth defects was admissible under the Federal Rules of Evidence. The Court held that the Frye general-acceptance test had not survived the enactment of the Federal Rules, and that trial judges had an independent gatekeeping role under Rule 702. The Court identified non-exhaustive factors the judge should consider:

1. Whether the theory or technique can be, and has been, tested.
2. Whether it has been subjected to peer review and publication.
3. The known or potential error rate and the existence and maintenance of standards controlling the technique's operation.
4. General acceptance in the relevant scientific community (retained as one factor, no longer the sole criterion).

Kumho Tire Co v Carmichael (1999) extended Daubert to all expert testimony, not just science. A tyre failure analyst's opinion about what caused a blowout is subject to the same gatekeeping as a DNA analyst's opinion about a profile match. The trial judge has discretion in how to apply the factors: not every factor is relevant to every technique, and the inquiry is flexible rather than formulaic.

Federal Rule of Evidence 702 was amended in 2023 to clarify that the proponent of expert testimony must demonstrate by a preponderance of the evidence that the requirements are met, and that the expert's opinion must reflect 'a reliable application' of reliable principles to the facts. The amendment responded to concerns that some courts had been too deferential to experts, treating Daubert as a low bar rather than a genuine reliability screen.

The US is unusual in having a clear split between state court systems on which standard applies. Federal courts and the majority of states follow Daubert (or their own Daubert-equivalent). A smaller but significant group, including California (under Kelly/Frye), Illinois, and New York (in many contexts), retain the Frye general-acceptance standard. This means that identical forensic evidence can be subject to different admissibility tests depending purely on which court is hearing the case.

England and Wales does not have a single codified admissibility standard comparable to FRE 702. The framework is built from case law, the Civil Procedure Rules, and the Criminal Practice Directions. The foundational appellate test is usually traced to R v Bonython (1984, South Australia Supreme Court), widely adopted by English courts, which requires: (a) that the subject of the evidence forms part of a body of knowledge or experience sufficiently organised or recognised to be accepted as a reliable body of knowledge; and (b) that the witness has sufficient familiarity with and knowledge of the field.

More recent English case law has sharpened this. In R v Reed and Reed [2009] EWCA Crim 2698 the Court of Appeal considered the admissibility of DNA low-copy number profiling and set out a reliability inquiry that echoes Daubert's themes without using its terminology: has the underlying science been peer-reviewed, is there a body of expertise in the field, and can the specific method be tested for reliability? In civil proceedings, CPR Part 35 and its accompanying Practice Direction impose duties on experts and give courts wide powers to control expert evidence, including limiting each party to a single expert and ordering experts to confer before trial.

The UK Forensic Science Regulator's Codes of Practice and Conduct, which became legally enforceable under the Forensic Science Regulator Act 2021, add a statutory layer: accredited providers must meet defined quality standards, and the Regulator can issue quality standards that courts take into account when evaluating whether evidence is reliable. This is not a formal admissibility test but it creates a quality floor below which courts are increasingly reluctant to admit evidence.

India's approach to expert evidence is rooted in a different legal tradition. The Bharatiya Sakshya Adhiniyam 2023 (BSA), which replaced the Indian Evidence Act 1872, addresses expert opinion in Section 39. The section provides that when the court has to form an opinion on a point of foreign law, science, art, or the identity of handwriting or fingerprint impressions, the opinions of persons especially skilled in those areas are relevant facts. This is a relevance rule, not an admissibility gate of the Daubert type: if the witness is qualified as an expert, their opinion is relevant and may be placed before the court.

The reliability of the method underlying the opinion is addressed in cross-examination and in the court's assessment of weight rather than at a pre-admission hearing. Courts do consider whether a method is widely recognised and whether the expert followed proper procedures, but there is no equivalent to the formal Daubert hearing at which the judge rules on admissibility before the jury hears the evidence. The court is the fact-finder in most Indian trials (jury trials were formally abolished by the Code of Criminal Procedure 1973, which came into force in 1974, though they had declined sharply after the Nanavati case in 1959-60) and it exercises more integrated judgement across admissibility and weight.

Reformers in India have argued for a stronger reliability screen, citing cases where poorly validated forensic techniques reached courts unchallenged. The Law Commission's 185th Report (2003) and subsequent academic work highlighted the absence of a formal admissibility gateway. The BSA 2023 consolidates the prior provisions but does not adopt a Daubert-style screen. This remains an area of active discussion in Indian legal scholarship.

In predominantly inquisitorial systems, the question of whether to admit expert evidence takes a different form. The court or examining magistrate typically selects an expert from a list of accredited specialists, instructs them to examine specific questions, and receives a report that enters the case dossier. The reliability concern is addressed upstream, through the accreditation and qualification requirements for the court expert list, rather than at a hearing.

France's system of experts judiciaires (judicial experts), Germany's Sachverstaendiger (expert witnesses), and the Netherlands' registerforensisch expert all require formal qualification and listing before a person can be court-appointed. This does not eliminate the possibility of poor-quality evidence, but it shifts the accountability mechanism from adversarial challenge at trial to professional qualification before appointment. Party-instructed experts exist but typically have lower formal status than the court expert.