Black Box Thinking

The book opens with the story of Elaine Bromiley, a 37-year-old mother of two who entered a hospital near her home in Buckinghamshire, England, on March 29, 2005, for a routine sinus operation. During anesthesia, the anesthetist could not insert a breathing device because Elaine's jaw muscles had tightened, and her oxygen levels plummeted. Two additional doctors joined the effort, but all three fixated on accessing the airway through the mouth. Jane, the head nurse, fetched a tracheotomy kit, the indicated emergency procedure, and informed the doctors, but they did not respond. Inhibited by the steep hierarchy of the operating room, she did not press further. Elaine was deprived of oxygen for 20 minutes, suffered catastrophic brain damage, and died 13 days later. The lead surgeon told her husband, Martin Bromiley, only that there were "some problems during the anesthesia" and that it was "a one-off" (7), making no mention of the failed attempts or the nurse's intervention.

Matthew Syed uses Elaine's death to frame the book's central argument: How individuals and organizations respond to failure determines whether they achieve progress. He contrasts aviation, where nearly indestructible black boxes record flight data and independent investigators analyze every accident to mandate industry-wide reforms, with health care, where a 2013 study in the Journal of Patient Safety estimates more than 400,000 Americans die annually from preventable medical error. Syed argues that the problem is not incompetence but culture: Errors are concealed through euphemisms, creating what he calls "closed loops," systems where failure never leads to progress because information about errors is ignored or suppressed. He contrasts these with "open loops," where feedback from failure drives systematic improvement.

To illustrate, Syed examines the 1978 crash of United Airlines Flight 173 in Portland, Oregon. Captain Malburn McBroom, a veteran pilot, entered a holding pattern to troubleshoot a suspected landing-gear malfunction and became so fixated on the problem that he lost track of fuel levels. Flight Engineer Forrest Mendenhall warned repeatedly about dwindling reserves but could not bring himself to directly challenge his captain, an illustration of how hierarchies inhibit assertiveness through deferential, indirect phrasing. The plane ran out of fuel and crashed, killing 10 people including Mendenhall. The parallel to the Bromiley case is direct: In both incidents, authority figures fixated on one problem while the real danger went unaddressed, and subordinates who saw the solution were silenced by hierarchy. The critical difference was the institutional response. Aviation launched a full investigation, leading to the creation of Crew Resource Management (CRM) training, which teaches assertiveness procedures for junior crew and listening skills for captains. In health care, no investigation occurred until Martin Bromiley, a commercial pilot with over 20 years of experience, refused to accept the hospital's dismissal and pressed for answers.

Syed argues that success is paradoxically built upon failure. Captain Chesley Sullenberger's 2009 landing of US Airways Flight 1549 on the Hudson River relied on CRM protocols, fly-by-wire technology, and checklists that all emerged from prior crashes. Sullenberger himself stated that everything known in aviation was learned because "someone somewhere died" (40). Syed extends this logic through philosopher Karl Popper's framework: Science progresses by making testable predictions vulnerable to falsification, while pseudosciences like astrology immunize themselves from failure and therefore can never learn. This leads to a practical insight: Expertise improves only in fields where practitioners receive clear, timely feedback on their errors. Dr. Gary Kaplan, CEO of the Virginia Mason Health System in Seattle, demonstrates this principle by adapting the Toyota Production System, where any worker can halt production to flag an error, into Patient Safety Alerts. After a cultural shift toward open reporting, Virginia Mason becomes one of the safest hospitals in the world, with a 74 percent reduction in liability insurance premiums.

Syed then asks why people so often refuse to learn from failure, turning to Leon Festinger's theory of cognitive dissonance: When evidence challenges deeply held beliefs, people reframe the evidence rather than change their minds. Festinger demonstrated this by infiltrating a doomsday cult in 1954; when the prophesied apocalypse did not arrive, core members became more committed rather than less, reinterpreting the failure as proof their faith had saved the world. This mechanism explains why prosecutors resisted DNA exonerations, why politicians offered escalating justifications for the Iraq War as each prior rationale collapsed, and why physicians redefine errors through euphemisms until they vanish from conscious thought. Syed traces the case of Juan Rivera, a 19-year-old who had a history of psychological conditions and was convicted of a 1992 murder based on a confession signed during a psychotic episode. When DNA evidence excluded Rivera in 2005, prosecutors claimed the 11-year-old victim had had consensual sex with a lover, a theory contradicted by evidence of a brutal assault. Rivera spent 19 years in prison before his release in 2012.

Shifting to systems that learn, Syed recounts how Unilever's expert mathematicians failed to design an efficient detergent nozzle despite deep knowledge of fluid dynamics. A team of biologists with no understanding of the physics succeeded by testing variations iteratively: After 45 generations and 449 failures, they produced a far superior nozzle whose shape no mathematician could have predicted. This bottom-up, evolutionary process mirrors the lean start-up methodology, where entrepreneurs test a minimum viable product before committing resources. Syed also champions randomized control trials (RCTs) as tools for cutting through ambiguity. The Scared Straight program, which took delinquent teenagers to Rahway State Prison in New Jersey for intimidation sessions, was celebrated after an Oscar-winning documentary reported that most participants reformed. But when criminologist James Finckenauer of Rutgers conducted the first RCT, he found that the program actually increased delinquency by as much as 28 percent.

Syed examines the concept of marginal gains, breaking large problems into small, rigorously testable parts, through British cycling coach Sir David Brailsford, who led British track cycling from one Olympic gold medal in 2000 to eight in both 2008 and 2012. He connects this approach to creative innovation, arguing that failure fuels creativity. James Dyson's dual-cyclone vacuum cleaner began with frustration at clogging filters and required 5,127 prototypes after the initial eureka moment. At Pixar, every film undergoes rigorous iteration, with president Ed Catmull acknowledging that all their movies start out poorly and are reworked until they find their core.

Syed devotes attention to the psychology of blame, arguing that instant blame collapses complex events into simple stories and prevents systemic reform. Amy Edmondson's study of hospital nursing units found that blame-heavy cultures reported fewer errors but actually produced more, because staff were afraid to speak up. The consequences are dramatized through the 2007 death of 17-month-old Peter Connelly ("Baby P") in London: Media outrage targeted social workers rather than systemic failures, causing an exodus from the profession and a rise in child deaths of more than 25 percent the following year. In aviation, Captain William Glen Stewart was convicted after a near-miss at Heathrow despite a chain of compounding circumstances largely beyond his control. Three years later, Stewart took his own life.

In the final chapters, Syed explores the psychological foundations of resilience. Carol Dweck's research distinguishes between a Fixed Mindset, which views talent as innate, and a Growth Mindset, which views abilities as developable through effort. Jason Moser's EEG studies show that Growth Mindset individuals pay three times more neural attention to their errors, and this attention correlates directly with improvement. Angela Duckworth's research finds that "grit," or perseverance through setbacks, predicts success at West Point's grueling initiation better than the military's comprehensive evaluation metrics. Syed argues that redefining failure as educative rather than shameful is the most important step toward building high-performance cultures.

The book's coda traces the battle between closed and open attitudes to error from ancient Greece, where criticism of ideas was first tolerated, through the medieval suppression of dissent, to Francis Bacon's challenge to dogmatic authority in 1620, which helped launch the modern scientific revolution. Syed argues that politics, business, and public policy still protect ideas from failure through narrative and cognitive dissonance rather than testing them through experimentation. He returns to Martin Bromiley, who continues to lead the Clinical Human Factors Group as an unpaid volunteer. Clinicians worldwide credit Elaine's case with changing training, challenging hierarchies, and saving lives. Bromiley's daughter Victoria concludes: "Our mother's death was very hard for all of us and we know that nothing can bring her back. But I hope Dad continues with his work, and helps to spare other families from what we have had to go through . . . I think Mum would have liked that" (294).