Appendix C: Landmark Cases and Wrongful Convictions

Forensic science is taught best through the cases that made and broke it. This appendix gathers, in one place, the real, publicly documented cases featured across the book — the triumphs that show what validated science can do, and the failures and exonerations that show what happens when junk science, bias, or fraud is allowed into a courtroom. For each case: what happened, the forensic lesson it carries, and the chapter that develops it in full.

Everything here is Tier-1 — drawn from court records, official investigations, and the public documented record. We confine ourselves to documented public facts; where the record is contested or a detail is uncertain, we say so and keep the claim qualitative. Several of these cases involve real, living people who were wrongly accused and later cleared; they are catalogued precisely because the system failed them, and nothing here asserts wrongdoing the record does not establish. The book's one fictional case — the Mill Creek "Cold Case File" that runs through every chapter — is not in this catalog; it is clearly labeled as a teaching construction wherever it appears.

The catalog is organized in two halves. Section C.1 — Triumphs and foundations collects the cases where forensic science worked, established a method, or exposed an error. Section C.2 — Failures, exonerations, and scandals collects the cases where it failed, and what the failure taught. We begin, in both senses, with the three anchors.

C.1 Triumphs and foundations

The anchor: the Golden State Killer / Joseph James DeAngelo (Chapters 7, 8, 29, 39)

For decades, a series of burglaries, rapes, and murders across California — attributed to an offender known by various names, including the "Golden State Killer" — went unsolved despite a recovered crime-scene DNA profile that matched no one in law-enforcement databases. In 2018, investigators uploaded the crime-scene profile to GEDmatch, a public genealogy database, and identified not the offender directly but distant relatives who shared segments of DNA with him. By building out family trees from those genetic cousins and narrowing by age, geography, and other records, they reconstructed a candidate — Joseph James DeAngelo — then confirmed the identification by covertly collecting and testing his discarded DNA against the crime-scene profile. He was arrested and ultimately admitted responsibility in a 2020 plea.

The forensic lesson. This is the book's emblem of real, validated forensic progress — investigative genetic genealogy (IGG): crime-scene DNA → a public genealogy database → family-tree reconstruction → a candidate → confirmation by direct comparison. It is powerful and it rests on solid ground, but it also reintroduces interpretation and raises genuine privacy and consent questions about the genetic databases it depends on (Chapter 8, §8.5; Chapter 29; the ethics in Chapter 38). It is the proof that the field's rigorous end is genuinely transformative — closing cases cold for forty years — and the standing answer to anyone who concludes from the failures in Section C.2 that forensic science is worthless. The whole craft is telling the two ends apart.

Colin Pitchfork — the first conviction by DNA (Chapter 7)

In England in the 1980s, the murders of two girls were linked, and an innocent man had given a false confession to one of them. Using the newly developed technique of DNA profiling, scientists showed that the same person had committed both crimes and that the man who had confessed was not that person — excluding him. A mass screening of local men followed; the true offender, Colin Pitchfork, was identified after he attempted to evade the screening, and his DNA matched the crime-scene samples.

The forensic lesson. The case that opened the DNA era carried, from the very first, the book's central asymmetry: the same test that convicted Pitchfork first exonerated an innocent man who had falsely confessed. DNA's power to exclude arrived in the same case as its power to associate — and the false confession that nearly closed the case wrongly previews the science of Chapter 33.

Kirk Bloodsworth — the first U.S. death-row DNA exoneration (Chapter 6)

In 1984, Kirk Bloodsworth, a former Marine, was convicted of the rape and murder of a nine-year-old girl in Maryland and sentenced to death, the case resting heavily on eyewitness accounts. He maintained his innocence. Years later, biological evidence was subjected to DNA testing that had not existed at trial — and his profile did not match. In 1993 he became the first American sentenced to death and later exonerated by DNA.

The forensic lesson. Bloodsworth is the asymmetry of forensic science made flesh: a mismatch did not merely weaken the case, it excluded him, cleanly and outright, where confident eyewitness testimony had convicted him. Exclusion is forensic science's surest and most morally important power — the power to say not this person, and to keep saying it until someone listens (Chapter 1, §1.6; Chapter 6).

Lindbergh kidnapping / State v. Hauptmann (Chapter 18)

In the 1930s, the kidnapping and death of the infant son of aviator Charles Lindbergh produced one of the most famous trials of the century. Among the evidence against Bruno Richard Hauptmann were the ransom notes, examined by questioned-document examiners who compared their handwriting to his, and wood-identification analysis tracing the kidnapper's ladder.

The forensic lesson. The Lindbergh case is the historical showcase of questioned-document examination — and, read with modern eyes, a useful illustration of both the discipline's reach and the ongoing debate over the validity of handwriting identification (Chapter 18, §18.3, §18.6). It shows a comparison discipline at the height of its courtroom authority, in an era before anyone asked the validation questions this book insists on.

Georgi Markov — the ricin assassination (Chapter 20)

In 1978 in London, the Bulgarian dissident Georgi Markov died after being jabbed in the leg, reportedly by a modified umbrella, with a tiny pellet. Forensic examination recovered the pellet and the investigation concluded it had carried the lethal toxin ricin.

The forensic lesson. A landmark of forensic toxicology and the detection of an exotic poison in trace quantity — and a reminder that the toxicologist's question is not only "what is present?" but "at what dose, and with what effect?" The dose makes the poison (Chapter 20).

George Trepal — thallium poisoning (Chapter 23)

George Trepal was convicted in a Florida poisoning case in which a family was sickened and one member died after ingesting the heavy metal thallium, introduced into bottled drinks. Instrumental analysis identified the toxin.

The forensic lesson. A showcase of instrumental analysis turning suspicion into identification — the confirmatory power of the modern laboratory bench to detect and confirm a specific poison (Chapter 23).

The Beltway snipers — Muhammad & Malvo (Chapter 15)

The 2002 sniper attacks around Washington, D.C., that killed ten people were ultimately linked to John Allen Muhammad and Lee Boyd Malvo, in part through firearms evidence connecting recovered ballistic material to the rifle in their possession, alongside other investigative evidence.

The forensic lesson. A high-profile application of firearms examination — and a hook for the book's harder question: firearms identification can often establish that a particular gun fired a particular bullet, but "matching" striations to one weapon is less certain than juries believe, and PCAST found its foundational validity not yet established (Chapter 15, §15.6). Even a famous "match" must be stated at its true strength.

R v Huntley — the Soham murders and forensic geology (Chapter 24)

In the English Soham murders case, Ian Huntley was convicted in connection with the deaths of two girls, with forensic evidence including geological and trace material linking locations and items.

The forensic lesson. An illustration of soil and geological evidence and the broader family of microscopic class evidence — valuable when the material is distinctive and the reasoning honest, dangerous when its probabilistic, class-level nature is overstated into individualization (Chapter 24).

BTK / Dennis Rader — caught by metadata (Chapter 25)

After decades evading identification, the offender who called himself "BTK," Dennis Rader, was identified after he sent police a floppy disk; forensic examination of the disk's metadata pointed investigators toward him.

The forensic lesson. A clean demonstration that digital evidence is everywhere and hard to fully erase — and that metadata, the data about data, can place authorship where the author believed himself anonymous (Chapter 25, §25.3).

Bernard Madoff — following the money (Chapter 27)

The collapse of Bernard Madoff's decades-long Ponzi scheme — the largest in history — is the showcase of forensic accounting: the reconstruction of where money actually went, and the recognition, in advance by analyst Harry Markopolos, of mathematical red flags that the claimed returns could not be real.

The forensic lesson. Follow the money. Financial records often outlive every other kind of evidence, and anomaly detection (Chapter 27, §27.5) can reveal fraud that confident testimony and glossy statements conceal — motive and method written in the ledgers.

Grim Sleeper / Lonnie Franklin Jr. — familial DNA (Chapter 29)

The "Grim Sleeper" serial-murder case in Los Angeles was advanced when a familial DNA search — a database search for a partial match indicating a close relative rather than the offender himself — pointed to a relative of Lonnie Franklin Jr., leading to his identification and conviction.

The forensic lesson. Familial searching is a real and powerful extension of DNA databasing, distinct from the genealogy used in the Golden State Killer case, and it carries its own privacy and fairness questions (Chapter 29, §29.5). A genuine emerging tool, held to the same yardstick as everything else.

Melendez-Diaz v. Massachusetts — the right to confront the analyst (Chapter 30)

In Melendez-Diaz v. Massachusetts, the U.S. Supreme Court held that forensic laboratory reports are testimonial, so that under the Confrontation Clause a defendant has the right to confront — to cross-examine — the analyst who produced a report, rather than have the report admitted as a bare certificate.

The forensic lesson. A report is only as trustworthy as the human who signed it. The scandals in Section C.2 are the strongest argument ever written for this right: the analyst must be available to be questioned, not hidden behind a certificate (Chapter 30; Chapter 4, §4.6).

Disaster victim identification — the World Trade Center and the 2004 tsunami (Chapter 35)

The identification effort after the September 11, 2001 attacks (led by the New York City Office of Chief Medical Examiner) and the international operation after the 2004 Indian Ocean tsunami are landmark disaster victim identification (DVI) efforts, applying the primary identifiers — fingerprints, dental records, and DNA — to commingled, degraded remains at enormous scale.

The forensic lesson. When the dead number in the dozens or thousands, identification becomes a logistical and humanitarian science of its own, reconciling antemortem and postmortem records to return names to the dead — forensics as an act of care for the living families (Chapter 35).

Wildlife and engineering forensics — ivory DNA and the Hyatt Regency collapse (Chapter 36)

Researchers (notably Samuel Wasser, working with U.S. wildlife authorities) used DNA to determine the geographic origin of seized elephant ivory, mapping poaching hotspots; and the 1981 Hyatt Regency walkway collapse in Kansas City became a defining case of forensic engineering and failure analysis.

The forensic lesson. Not every victim is a person. The same validated logic — comparison, sequence, failure analysis — extends to protecting ecosystems and to reconstructing why a structure failed (Chapter 36).

The reckoning itself — the NAS 2009 and PCAST 2016 reports (Chapters 5, 6, 38)

Not cases, but the documents against which every case in this book is measured. The 2009 National Academy of Sciences report, Strengthening Forensic Science in the United States, surveyed the whole field and found most of it scientifically underbuilt; the 2016 PCAST report focused on feature-comparison methods and asked which had demonstrated foundational validity. The Houston Police Department crime-lab scandal, which led to the creation of the independent Houston Forensic Science Center, stands as one model of structural reform (Chapter 38).

The forensic lesson. These reports are the book's organizing yardstick — the validity spectrum made official. They are catalogued among the triumphs because naming the problem honestly is itself the field's most important step toward fixing it (Chapter 6, §6.4–§6.6; Chapter 38).

C.2 Failures, exonerations, and scandals

The anchor: Cameron Todd Willingham (Chapters 5, 21, 22, 34, 38)

On December 23, 1991, a fire in Corsicana, Texas, killed the three young children of Cameron Todd Willingham, who escaped. Investigators concluded the fire had been deliberately set — and set by him. The arson finding rested on the fire-investigation folklore of its era: "crazed glass," low burns read as "pour patterns," char-depth readings, and reasoning by elimination ("negative corpus"). He was convicted of capital murder in 1992 and executed on February 17, 2004, maintaining his innocence throughout.

The forensic lesson. This is the book's emblem of junk science with a body count. Every fire indicator relied upon has since been discredited: crazed glass is an artifact of rapid cooling from fire-hose water; low burns and "pour patterns" are the normal product of a post-flashover fire; char readings do not reliably indicate accelerant; and negative corpus is a logical fallacy that modern fire-science standards (NFPA 921) reject. There was no instrumentally confirmed ignitable-liquid residue. A noted fire scientist, Dr. Gerald Hurst, reviewed the case before the execution and concluded the science did not support arson; the execution proceeded anyway. After his death, the Texas Forensic Science Commission's independent reviewer, Dr. Craig Beyler, concluded the original findings could not be scientifically supported. The case is not a formal posthumous exoneration and remains legally contested — but in the forensic-science community the verdict on the science is not seriously in dispute. It teaches the book's deepest themes made lethal: an unvalidated method (Theme 2) applied in good faith by sincere experts (junk science needs no malice) and heard by a jury as proof (the CSI effect's over-trust), with the law too slow to correct itself before an irreversible punishment (Chapter 22).

The anchor: Brandon Mayfield (Chapters 14, 30, 31, 34)

After the 2004 Madrid train bombings, which killed 191 people, Spanish police developed latent fingerprints on a bag of detonators and shared images internationally. The FBI Laboratory — among the most experienced fingerprint operations in the world — ran the latent through AFIS, and Brandon Mayfield, an Oregon attorney whose prints were on file from military service, appeared on the candidate list. An FBI examiner declared an identification; additional FBI examiners and even an examiner appointed for Mayfield's own defense concurred; the match was called "100 percent" and "absolutely incontrovertible." The Spanish National Police never agreed, and ultimately matched the latent to an Algerian national, Ouhnane Daoud. The FBI conceded the error, withdrew the identification, and apologized. Mayfield, held for roughly two weeks as a material witness and never charged, later received a public apology and a civil settlement.

The forensic lesson. Mayfield is the cautionary middle: even the "gold standard" is human judgment, and it can be confidently, unanimously, catastrophically wrong. The official DOJ Inspector General review found the failure was systemic and cumulative — a poor-quality latent that invited interpretation; circular reasoning in which the reference print shaped the reading of the latent; priming by the AFIS hit; a "100 percent" culture; verification that was not blind (so four examiners agreeing was one judgment wearing four coats); and case context (Mayfield's religion and associations) that made a match feel confirmed. The fix is not better cameras but context management and blind verification (Chapter 31). Mayfield proves a method can be foundationally valid (Theme 2) and still not infallible — both truths in one case (Chapter 14).

JonBenét Ramsey and Jeffrey MacDonald — scene-processing failures (Chapter 2)

The 1996 death of six-year-old JonBenét Ramsey in Colorado became a textbook example of a crime scene compromised early — the scene was not tightly controlled, and the contamination and disturbance complicated everything that followed. The Jeffrey MacDonald case likewise turned, in part, on disputes over how the scene was handled.

The forensic lesson. The case is often won or lost at the scene, and scene errors are irreversible. A scene not secured, logged, and documented before disturbance hands the defense (and history) a permanent set of doubts no later analysis can resolve (Chapter 2).

The lab-fraud scandals — Annie Dookhan, Sonja Farak, Fred Zain (Chapters 4, 38)

Annie Dookhan, a Massachusetts state drug chemist, "dry-labbed" — reported results for tests she never performed — and tampered with samples; Sonja Farak, a chemist at a different Massachusetts lab, consumed the drug standards and case samples she was supposed to analyze, working while impaired; and Fred Zain, a serologist in West Virginia and then Texas, systematically fabricated and overstated serological results across two states. Each scandal cast doubt on every case the analyst touched; the two Massachusetts cases together led to the dismissal of tens of thousands of drug convictions, among the largest such dismissals in American history.

The forensic lesson. A single unchecked analyst can poison an entire jurisdiction's casework — and accreditation alone will not stop them. The common structure: a single point of failure with too much trust and too little verification, no effective blind check on whether reported work was actually done, production pressure over verification, and errors that ran in the prosecution's favor. This last pattern points to the structural independence problem (Chapter 4, §4.6; Chapter 38). Keep this failure distinct from junk science: drug chemistry is sound, and an honest chemist using it gets the right answer — this is fraud, fixed by oversight, not by abandoning the method.

O.J. Simpson and David Camm — collection, contamination, and contested BPA (Chapter 10)

In the O.J. Simpson trial, the defense attacked the prosecution's DNA evidence not on the biology but on the handling — collection and contamination at and after the scene. The David Camm case turned in part on disputed bloodstain pattern analysis (BPA).

The forensic lesson. Two halves of the blood-evidence chapter. Simpson shows that the most rigorous method in the book can be neutralized by sloppy scene work — a contaminated sample with a flawless laboratory behind it is still contaminated (Chapter 10, §10.3; Chapter 2). Camm shows BPA for what it is: a method that sits low on the validity spectrum, highly examiner-dependent, where a vivid reconstruction can outrun the evidence (Chapter 10, §10.6).

George Floyd / State v. Chauvin — cause and manner of death (Chapter 11)

The 2020 death of George Floyd and the prosecution of Derek Chauvin turned in significant part on forensic-pathology testimony about cause and manner of death.

The forensic lesson. A high-profile illustration of the medicolegal determination of cause and manner of death, and of how that determination is examined and contested in court — the autopsy's central questions under adversarial scrutiny (Chapter 11). (Treated soberly, as a real death and a matter of public record.)

Casey Anthony and David Westerfield — entomology and the CSI effect (Chapter 13)

In the Casey Anthony trial, forensic entomology (and other decompositional evidence) featured in the dispute over time and circumstances of death; the case is also a touchstone for the CSI effect, the jury acquitting amid expectations of forensic certainty. The David Westerfield case also featured entomological time-of-death evidence — on which experts disagreed.

The forensic lesson. Insects are clocks, but the math of decomposition (accumulated degree days) is complicated by fire, season, and other variables, and experts can reach different estimates from the same remains (Chapter 13, §13.3–§13.4). And the CSI effect cuts both ways: juries primed by television may demand a certainty the science cannot provide.

Stephan Cowans and Shirley McKie — fingerprint errors beyond Mayfield (Chapter 14)

Stephan Cowans was convicted in part on a fingerprint identification later shown to be erroneous and was exonerated by DNA. Shirley McKie, a Scottish police officer, was wrongly implicated by a fingerprint identification that was later widely disputed.

The forensic lesson. Mayfield was not a singular freak. Latent-print identification carries a real, non-zero error rate, and these cases — an exoneration and a contested identification of a police officer — show the same human interpretive failure modes recurring (Chapter 14, §14.4, §14.6).

United States v. Glynn — a court limits firearms testimony (Chapter 15)

In United States v. Glynn and similar rulings, courts began restricting how firearms examiners may state their conclusions — for example, barring claims of certainty and requiring more qualified language.

The forensic lesson. The judicial response to the validity questions about firearms identification: when a method's foundational validity is contested, the honest courtroom remedy is to constrain the verb — "consistent with," not "to the exclusion of all other firearms" (Chapter 15, §15.6; Chapter 30).

Ray Krone and Roy Brown — the bite-mark exonerations (Chapter 16)

Ray Krone — the "Snaggletooth Killer," convicted partly on bite-mark testimony — and Roy Brown were both convicted on bite-mark "matches" and later exonerated by DNA. Krone became, in the public record, an emblem of bite-mark wrongful conviction.

The forensic lesson. Bite-mark analysis is the field's most thoroughly discredited discipline — visual "matching" of teeth to a mark with no validated basis — and these exonerations are why. The science put innocent people in prison and the DNA took them out (Chapter 16, §16.5–§16.6).

Florida v. Bundy / Chi Omega — the bite-mark contrast (Chapter 17)

Bite-mark evidence featured in the prosecution of Ted Bundy in the Chi Omega case — a high-profile use of the method.

The forensic lesson. Catalogued deliberately alongside Krone and Brown as the contrast that makes the point: that a discredited method occasionally appears in a case where the defendant was, by other evidence, guilty does not validate the method. A broken clock and a guilty defendant do not make bite-mark matching science (Chapter 16; Chapter 17, §17.5).

Mark Hofmann — forgery and the document examiner (Chapter 18)

Mark Hofmann was a master forger of historical documents whose frauds — and ultimately murders committed to cover them — were exposed in part by document and ink examination.

The forensic lesson. The sounder half of questioned-document work: detecting forgery and alteration through the chemistry of ink and paper and the physical evidence of fabrication, distinct from the more contested business of handwriting identification (Chapter 18, §18.4).

Santae Tribble, Kirk Odom, and the microscopic-hair scandal (Chapter 19)

Santae Tribble and Kirk Odom were convicted in part on FBI microscopic hair comparison testimony and later exonerated; their cases were among those that prompted a sweeping FBI/Department of Justice review, in which the FBI acknowledged that examiners had, over many years, overstated the significance of microscopic hair matches in testimony. The Wayne Williams Atlanta case, by contrast, is a noted fiber case.

The forensic lesson. Microscopic hair comparison is weak — it can rarely individualize, and for decades its conclusions were systematically overstated until the FBI itself admitted it. mtDNA now adds the rigor microscopy lacked. The Williams fiber evidence shows the same family of trace evidence used more defensibly when the materials are distinctive (Chapter 19, §19.2–§19.4). (Some exact totals and percentages from the FBI review are stated qualitatively here, as the precise figures have been reported variously.)

Han Tak Lee — arson science matured in time (Chapter 22)

Han Tak Lee was convicted of arson-murder in Pennsylvania on fire-science reasoning of the same discredited vintage as Willingham's, and — unlike Willingham — lived to be released after the science was re-examined.

The forensic lesson. The companion to Willingham, and the cruel difference a reversible punishment makes: the same maturing fire science that came too late for Willingham freed Lee. Science is self-correcting; an execution is not (Chapter 22, §22.6; Chapter 5, §5.1).

The FBI Lab and Frederic Whitehurst — the DOJ OIG 1997 report (Chapters 21, 23)

Whistleblower Frederic Whitehurst's allegations led to a 1997 Department of Justice Office of the Inspector General review of the FBI Laboratory that found significant problems in certain units and practices; the Oklahoma City bombing investigation figured among the matters examined.

The forensic lesson. Even the most prestigious laboratory is not above scrutiny, and internal whistleblowing plus external review is part of how the field corrects itself. The report is a Tier-1 source used from different angles for chemistry and instrumental analysis (Chapter 21; Chapter 23).

Houston-area roadside field tests — the presumptive-test exonerations (Chapter 21)

In Harris County, Texas, and elsewhere, numerous people pleaded guilty on the basis of roadside presumptive color tests for drugs that later laboratory confirmation showed were false positives, leading to a wave of exonerations.

The forensic lesson. Presumptive tests suggest; only instrumental confirmation establishes. The "field test" problem is the gap between the two — and a warning that a presumptive result acted on as if it were confirmatory can convict the innocent (Chapter 21, §21.6; Chapter 23).

Boston Marathon bombing — Tsarnaev and the Sunil Tripathi misidentification (Chapter 26)

After the 2013 Boston Marathon bombing, official image and video forensics helped identify the Tsarnaev brothers — while online crowdsourced "investigation" wrongly named an innocent missing student, Sunil Tripathi, in a damaging public misidentification. Separately, manipulated news photographs (the Reuters/Adnan Hajj episode) illustrate image-authentication failures.

The forensic lesson. Two faces of image forensics: the legitimate, disciplined authentication and reading of footage, versus the catastrophe of untrained crowdsourced misidentification — and the broader point that images can be manipulated and must be authenticated, not trusted on sight (Chapter 26).

Richard Jewell and the limits of profiling (Chapter 28)

Richard Jewell, the security guard who discovered a bomb at the 1996 Atlanta Olympics and helped save lives, was wrongly suspected — fitting a "lone bomber" profile — before being cleared. (The chapter also treats competency through Dusky v. United States and the Jared Lee Loughner matter.)

The forensic lesson. Criminal profiling is the most over-sold forensic product on television, and Jewell is its cautionary tale: a profile pointed at an innocent and very nearly ruined him. The honest science of forensic psychology — competency, sanity, risk assessment — is narrower and far more careful than the TV profiler (Chapter 28, §28.4–§28.5).

Joyce Gilchrist — the forensic analyst who overstated (Chapter 30)

Joyce Gilchrist, an Oklahoma forensic analyst, was found to have given testimony that overstated or misrepresented forensic findings, contributing to wrongful convictions later identified (including cases such as Jeffrey Pierce and Curtis McCarty).

The forensic lesson. The expert witness who climbs past what the science supports. Gilchrist is the courtroom counterpart to the lab scandals — not (only) faked bench work but testimony that outran the evidence — and the reason the duty to communicate uncertainty honestly is a forensic skill, not an afterthought (Chapter 30, §30.4).

Ronald Cotton / Jennifer Thompson and Marvin Anderson — eyewitness error (Chapter 32)

Jennifer Thompson confidently identified Ronald Cotton as her rapist; he was convicted, served years, and was exonerated by DNA — the real assailant was another man. Thompson and Cotton later became advocates for eyewitness reform together. Marvin Anderson was similarly convicted on a cross-racial eyewitness identification and later exonerated.

The forensic lesson. The most persuasive evidence to a jury — a confident witness pointing — is among the least reliable. Memory is reconstruction, not recording; the confidence-accuracy relationship is weak under poor witnessing conditions and suggestive procedures; and cross-racial identification is especially error-prone. These cases drove the reforms in lineup procedure (Chapter 32).

The Central Park Five / Exonerated Five and Brendan Dassey — false confessions (Chapter 33)

Five teenagers (the Exonerated Five) falsely confessed under interrogation to a 1989 Central Park assault and were convicted; they were later exonerated when the actual perpetrator, Matias Reyes, was identified and confirmed by DNA. Brendan Dassey, a juvenile, gave a confession whose voluntariness and reliability have been widely disputed.

The forensic lesson. Innocent people — especially the young and vulnerable, under long interrogation — confess to crimes they did not commit, often enough that false confession is a leading cause of wrongful conviction. The DNA that identified Reyes is what finally overrode the confessions (Chapter 33). The fix is recording, time limits, and reformed interview methods.

Earl Washington Jr. — a near-execution corrected by DNA (Chapter 34)

Earl Washington Jr., a man with intellectual disability, falsely confessed to a Virginia murder, was sentenced to death, and came within days of execution before DNA testing led to his exoneration.

The forensic lesson. The wrongful-conviction chapter's primary case, in which the contributing factors converge — a false confession from a vulnerable defendant, and DNA as the corrective that arrived just in time. It anchors the book's survey of how bad forensics, bias, eyewitness error, and false confessions combine, and why conviction-integrity review matters (Chapter 34).

The untested rape-kit backlog — Detroit and Cleveland (Chapter 37)

In Detroit (Wayne County), Cleveland (Cuyahoga County), and other jurisdictions, tens of thousands of sexual-assault evidence kits were found to have gone untested for years; testing backlogs, when finally cleared, identified serial offenders and produced convictions long delayed.

The forensic lesson. Evidence collected from living victims is only useful if it is actually tested. The backlog is not a neutral logistical fact — it leaves offenders free and victims without answers — and clearing it is both a forensic and a justice imperative, handled in a trauma-informed frame (Chapter 37, §37.5).

The Beatrice Six — false convergence (Chapter 39)

In the Beatrice Six case (the murder of Helen Wilson in Beatrice, Nebraska), six people were wrongly convicted — several having given false confessions that appeared to corroborate each other — before DNA exonerated all of them and identified the true perpetrator.

The forensic lesson. The capstone's cautionary mirror. Evidence that all seems to converge can converge on the wrong answer when the "independent" pieces were not independent at all — confessions contaminated by the same pressure and feeding each other. Genuine convergence requires genuinely independent lines of evidence; false convergence is how a tidy, agreeing case can still be entirely wrong (Chapter 39, §39.3–§39.5). It is the warning that governs the honest assembly of any case, including the book's own cold case.

C.3 Reading the catalog

Set the two halves side by side and the book's argument appears in miniature. The triumphs — DNA from Pitchfork to the Golden State Killer, the confirmatory bench from Markov to Trepal, the money trail in Madoff, the names returned to the dead after 9/11 — are real, and they are why forensic science at its best is one of the great instruments of justice ever built. The failures — Willingham executed on fire folklore, Mayfield's "100 percent" error, the bite-mark and hair exonerations, the false confessions of the Exonerated Five and the Beatrice Six, the lab fraud that voided tens of thousands of cases — are equally real, and they are why no method may be trusted on its reputation alone.

The throughline is the book's first theme: in case after case, what freed the innocent was DNA's power to exclude, and what convicted them was a weak inclusion — a hair "consistent with," a bite mark "matching," a confident "that's him," a confession extracted in the interrogation room — presented to a jury as if it were proof. Knowing where a method sits on the validity spectrum, and refusing to climb past the honest verb, is not an academic exercise. In the cases in this appendix, it was the difference between justice and its opposite — and, more than once, between a life and a death.