The 2016 review

A presidential science council found that DNA profiling stood on solid ground — and that most other forensic disciplines did not.

What the council found

In September 2016, the President's Council of Advisors on Science and Technology released a report titled Forensic Science in Criminal Courts: Ensuring Scientific Validity of Feature-Comparison Methods. The council — scientists, not practitioners — examined whether the forensic disciplines routinely admitted as evidence had been properly validated. The answer, for most of them, was no.

Bite-mark analysis, footwear comparison, and hair microscopy all failed the council's test for foundational validity. The report did not find that these methods were always wrong; it found that the research establishing their error rates, reproducibility, and reliability under realistic conditions had never been done ↗. That absence of evidence, the council argued, should matter to courts deciding what to admit.

CROSS-REF

DNA profiling occupied a different category. Single-source profiles and simple two-person mixtures, the council concluded, met the standard for foundational validity: the underlying biology is understood, the chemistry has been independently replicated, error rates can be measured, and NIST ↗ maintains the reference data behind the core loci. That endorsement was not unconditional. The report drew a firm line at complex mixtures — three or more contributors, or low-template samples — where it found that probabilistic genotyping software had not yet been sufficiently validated for the full range of conditions in which it was being used.

The distinction matters because courts had been admitting complex-mixture interpretations on the strength of DNA's broader reputation, rather than on evidence that the specific computational methods applied to those samples had been tested. The council's position was that the reputation of the technique and the validity of a particular application of it are two separate questions.

SWGDAM and the broader forensic science community pushed back on parts of the report, arguing that the council had set a validation standard higher than was workable in practice. That disagreement has not been fully resolved. What the 2016 review did establish, beyond argument, is a framework: foundational validity first, applied validity second, and the burden of demonstrating both sits with the proponent of the evidence, not with the defence attempting to challenge it.