Unfair: The New Science of Criminal Injustice: Summary & Key Insights

One of the first myths I had to dismantle is the idea that our legal actors—judges, juries, police officers—operate as neutral machines of reason. The system assumes that facts speak for themselves and that every decision rests on pure logic. Cognitive science has proven otherwise. When an officer evaluates a suspect, subtle cues such as race, clothing, or tone of voice activate patterns of unconscious judgment. When a juror listens to testimony, the story’s emotional rhythm can outweigh the factual strength of evidence. Even judges, trained for decades to distance themselves from emotion, can be swayed by fatigue, hunger, or whether their favorite team won the night before.

The illusion of objectivity comforts us; it lets us believe that justice is statistical, not psychological. But as experiments show, perception is always filtered through bias. When we recognize this, we realize that objectivity isn’t something we have—it’s something we must build into procedures. Structural changes like blind administration of lineups, random assignment of cases, or computer‑assisted sentencing can help shield decisions from unconscious influence. Yet these reforms only succeed when we accept that human fallibility is universal and predictable. The first step toward fairness is humility: admitting that the mind is not objective at all.

Few elements of the courtroom carry more persuasive power than an eyewitness, yet the science of memory tells us that this confidence is misplaced. Memory is not a video recording—it is a reconstruction shaped by expectation, stress, and external suggestion. I recount numerous cases where victims and witnesses, certain of their recollections, identified innocent people as perpetrators. Even small details—lighting conditions, the presence of a weapon, or subtle hints from police—can completely alter what is remembered.

Through laboratory studies, psychologists have shown how malleable memory becomes under pressure. Each retelling changes the past a little more, until imagination and recollection blend together. In law, however, we have tended to treat memory as sacred testimony. When we place such faith in a witness’s confidence, we create a dangerous feedback loop: jurors interpret confidence as accuracy, strengthening belief in false memories. The way out requires reengineering procedure—standardizing lineups, instructing jurors about memory distortion, recording all identification processes. Science doesn’t strip away human testimony; it refines it, ensuring that what counts as evidence aligns with how the mind truly remembers.