Chapter 1 — Key Takeaways

A one-page reference card. If you remember nothing else from this chapter, remember these.

The core ideas

  • Forensic science = science for the forum. It is the application of established scientific disciplines to questions a court must answer — performed under adversarial, high-stakes, often one-shot conditions that ordinary science never faces.
  • Criminalistics is the lab-bench subset (prints, fibers, firearms, drugs, DNA). Forensic science is broader: pathology, anthropology, accounting, psychology, digital, and more.
  • Television is ~90% fiction. The CSI effect distorts justice in two opposite directions: jurors demand impossible certainty (and acquit without it) and over-trust weak evidence (and convict on it). The cure is calibration, not cynicism.

The logic of comparison

  • Class characteristic = shared by a group (a sole pattern, a blood type). It narrows.
  • Individual characteristic = effectively unique to one source (accidental wear, a fracture match). It can strongly associate.
  • Individualization — "this source and no other, in all the world" — is demonstrable for quantified DNA but an overstated assumption for most pattern methods (NAS 2009).

The honest verbs (memorize the ladder)

exclude → consistent with / cannot exclude → strongly supports → ~~proves~~

Reserve "proves" for the rare, quantified claim. Most wrongful convictions in this book came from presenting "cannot exclude" as if it were "proves."

The yardstick

  • The validity spectrum ranks methods by the strength of the science behind their core claim, measured against the 2009 NAS report and the 2016 PCAST report (foundational validity + known error rate).
  • Top of the spectrum: single-source DNA, instrumental chemistry. Bottom: bite-mark analysis.
  • A published error rate is a good sign, not an admission. "No one can tell you the error rate" is the finding.

The book's spine

  • Theme 1: forensic science excludes more reliably than it proves.
  • Theme 2: not all methods are equally valid (DNA ≠ bite marks).
  • Theme 3: cognitive bias is the biggest threat to accuracy.
  • Theme 4: the CSI effect cuts both ways.

What you could honestly say on the stand

"The evidence is consistent with the suspect and does not exclude him; how strongly it supports him as the source depends on how rare these features are and on ruling out alternatives like relatives, transfer, and laboratory error." — Not "it's a match, so it's him."

The Cold Case

The file opens with one question — Was Marcus Diallo's death an accident, or a homicide staged to look like one? — and one rule: the evidence's job is to exclude what it can, not to confirm the first story. Don't theorize before you have data.