Chapter 28 — Key Takeaways

A scannable one-page card. For the full argument and the worked examples, see index.md.

The core claims

• Forensic psychology is two different things wearing one name. The working discipline — competency to stand trial, criminal responsibility (sanity), violence risk assessment — is genuine, court-admitted, standards-governed forensic science. Criminal profiling — a psychological portrait of an unknown offender — is the over-sold exception, low on the validity spectrum.
• Competency ≠ sanity. Competency is present-tense (can the defendant understand the proceedings and assist counsel now?); the insanity defense is past-tense (their mental state at the time of the offense). Different questions, different evidence, decided separately. Incompetent ≠ acquitted; it pauses the trial for restoration.
• Structured beats intuition. Actuarial / structured risk assessment measurably outperforms unstructured clinical judgment — the same lesson the whole book teaches: a validated, error-rate- bearing procedure beats expert gut. Even the best risk tools give a group probability, never an individual certainty (the prosecutor's-fallacy discipline, Chapter 9).
• Profiling's foundation is shaky. The FBI organized/disorganized typology does not hold up (scenes mix; "types" blur). The core leap — scene behavior → an unknown offender's biography — is unvalidated, built on an overstated "behavior reflects personality" premise, from a small non-representative sample of caught offenders, with no measured error rate.
• Profiling accuracy is unimpressive. The studies that exist find profilers do not substantially outperform comparison groups, and profiles are full of vague statements. "It helped catch him" ≠ "it works" — the survivorship trick (remember the hits, drop the misses).
• The Barnum effect explains why profiles feel accurate. People accept vague, broadly applicable descriptions as specific to themselves (or an offender). A statement that cannot be wrong cannot be informative; real profiles live in that unfalsifiable register.
• A wrong profile is directionally dangerous. It doesn't just fail — it misdirects, pointing toward innocents (Richard Jewell) and away from the actual offender. Treat a profile as one tentative, disconfirmable hypothesis, never a template the evidence is bent to fit.
• MO vs. signature. Modus operandi = learned, functional behavior to commit the crime (evolves). Signature = non-functional behavior meeting a psychological need (more stable; better for linkage). Linkage ("same offender?") is more defensible than prediction ("who is he?") — but still "consistent with," never "proves."

The method-validity verdict (NAS 2009 / PCAST 2016)

Where they sit: the assessments (competency, sanity, risk) are genuine forensic science, admitted in court daily. Predictive profiling sits low — above the discredited pattern methods only because its cousins (linkage, geographic profiling) have traction and it is rarely offered as courtroom proof, but well below DNA, toxicology, and fingerprints, and below even contested firearms/BPA, because those at least attempt a physical comparison that could be validated.

What you can honestly say on the stand

• Competency: "In my opinion, to a reasonable degree of professional certainty, the defendant [does / does not] presently have a rational and factual understanding of the proceedings and the ability to assist counsel, as the Dusky standard requires."
• Risk: "Based on a validated structured instrument, this individual falls in a category whose members reoffend at approximately [rate] over [period]. I cannot say whether this person will be among them; this is a group probability, not an individual prediction."
• Profiling / behavioral analysis: at most, "The behavior across these offenses is consistent with a common offender" (linkage). NOT "the defendant fits the profile, therefore he committed the crime."
• What you must NOT say: that a profile identifies or individualizes an offender; that "fitting the profile" is evidence of guilt; or any conversion of a behavioral narrative into proof of who did it.

The cold-case line

Early on, a profiler's "lone drifter / stranger" narrative pointed investigators away from Roy Keller toward a phantom outsider. Honest status: profiling misled; excluded as proof. It is not evidence, includes and excludes no one, and teaches method, not fact — a profile is a hypothesis to be tested and disconfirmed, and here it pointed the wrong way.

Key terms (one line each)

• Forensic psychology — applying psychological science to legal questions (competency, sanity, risk) in examinable people.
• Criminal profiling — inferring an unknown offender's traits from the crime; weak, unproven, not court proof of identity.
• Modus operandi (MO) — learned, functional crime-commission behavior; evolves with experience.
• Signature — non-functional behavior meeting a psychological need; more stable; used for linkage.
• Barnum effect — accepting vague, universally applicable descriptions as specifically accurate; why profiles feel uncanny.

The themes this chapter advanced

• The CSI effect (Theme 4) — the purest case in the book: no forensic product is more glamorized on TV and less validated in fact than the profiler. The Barnum effect is the mechanism by which the over-sold product sells itself.
• The validity spectrum (Theme 2) — extends the yardstick into the behavioral sciences: the methods that examine testable subjects (competency, risk) sit far above the one that reads an absent mind (profiling), which lands low.
• (Also touched: exclusion over proof — risk and linkage constrain and give probabilities, never identify; cognitive bias — confirmation/hindsight bias + tunnel vision turn a useless profile into a felt "hit," §28.5 and Chapter 31.)