Erosion of the presumption of innocence

Erosion of the presumption of innocence is the process by which people—especially poor people and people of color—are treated as guilty before their cases have been fairly investigated or proved. The book presents it not as a single courtroom error but as a chain: suspicion shapes police encounters, racial assumptions influence whom authorities pursue, and institutional judgments determine whose testimony is credible. The result can be the imprisonment of innocent people and the silencing of those most able to challenge an official account.

From presumption to consequence

The presumption of guilt has consequences far beyond an arrest. The book links it to poverty, racial bias, and structural and political forces that have produced wrongful convictions: scores of people sentenced to death have been exonerated, while hundreds more have been released after DNA testing proved them innocent of noncapital crimes. Thousands of innocent people, the narrative warns, remain in prison. These errors are not presented as isolated accidents; they emerge from a system in which suspicion and unequal power can displace the demand for reliable proof.

The book also traces how racialized suspicion has been reinforced by laws and customs. After Reconstruction, the return of white supremacy in the South brought voting restrictions and laws designed to preserve racial hierarchy. Fear of interracial relationships, including Walter McMillian’s relationship with Karen Kelly, was part of this history. For generations, Southern law enforcement treated the policing and punishment of Black men’s intimacy with white women as a duty. In this account, the presumption of guilt is sustained not only by individual prejudice but by a social order that has long marked some people as threatening before their conduct is established.

Police encounters and unequal credibility

Bryan Stevenson’s encounter with Atlanta police made the danger immediate. After an officer drew a weapon, Stevenson reflected that, at twenty-eight, his legal experience had helped him remain calm; at sixteen, nineteen, or even twenty-four, he might have run. He worried about young Black men who would not know how to respond to a police stop. He found statistics indicating that Black men were eight times more likely than white men to be killed by police. Although the disparity later improved to four times as likely, he also noted the growth of Stand Your Ground laws, which empowered armed citizens to use lethal force.

Stevenson’s written complaint ran to nearly nine pages. It detailed an illegal vehicle search and the absence of probable cause, citing roughly half a dozen cases, though he had tried not to identify himself as a lawyer. Police repeatedly dismissed his complaint, and appeals up the chain of command failed. A deputy chief eventually apologized and promised extra community-relations work for the officers, but Stevenson did not feel vindicated. The episode left him questioning his own response and, more importantly, what might happen to someone without his legal knowledge, employment, or standing. If such a person were arrested after an illegal search, Stevenson wondered, would a lawyer take the claim seriously, and would a judge believe the person was innocent?

He could not set those questions aside even while his death-penalty caseload crowded out other police-misconduct work. He began speaking to youth groups, churches, and community organizations about the presumption of guilt assigned to poor people and people of color, and about the need to hold law enforcement accountable. The point was not that public safety and accountability were incompatible: he argued that police could protect the public without abusing people. His talks carried the issue from legal argument into community life, where the consequences of a stop could be understood as a shared risk rather than an individual misfortune.

Walter McMillian: suspicion in place of investigation

The prosecution of Walter McMillian shows how racialized suspicion could harden into a capital case. Tom Tate had not investigated McMillian’s life, background, or whereabouts on the day Ronda Morrison was murdered. He knew about McMillian’s affair with Karen Kelly and had heard rumors that McMillian’s independence meant he was dealing drugs. Eager to make an arrest, Tate accepted Ralph Myers’s accusations without first testing them against a thorough investigation.

There was evidence pointing the other way. On the day of the murder, McMillian was at home for a family fish fry, with relatives and church parishioners present. He had no tree job that day; he was replacing the transmission in his truck, and his mechanic friend helped dismantle it that morning. Black residents told Tate that he had arrested the wrong man. Their account and the details of McMillian’s whereabouts did not prevent the state from proceeding against him. The case illustrates a reversal at the heart of the concept: rather than requiring the accusation to survive scrutiny, the process treated the accusation as the starting truth and left McMillian and his community to overcome it.

The imbalance continued when Tom Chapman defended the conviction. Stevenson raised the community’s belief in McMillian’s innocence, including the claims of people who said McMillian had been miles away when the crime occurred. Chapman dismissed those views as uninformed and insisted that evidence implicated McMillian. The dispute was also about who could define evidence: community knowledge was discounted even as the prosecution’s case remained under challenge. Later, at an evidentiary hearing, Ralph Myers repudiated his trial testimony. That reversal put the reliability of a central prosecution witness directly before the court, but it did not by itself ensure that the system would reconsider the full case.

Who gets to be heard

The jury system supplied another structural barrier. In 1987, all forty elected district attorneys in Alabama were white, despite the state’s sixteen majority-Black counties. Legislative arrangements had also preserved white majorities in judicial circuits that included majority-Black counties. These facts do not establish the motivation behind every individual decision, but they show the racial imbalance within institutions that controlled prosecution and jury selection.

The right to serve on a jury had long been undermined by exclusionary practices. Despite earlier Supreme Court rulings against racial discrimination in jury selection, voting rolls and vague standards such as being “intelligent and upright” were used to keep Black people and women off jury lists. Later decisions challenged underrepresentation, but prosecutors retained peremptory strikes that could remove potential jurors without stating a cause. The evidentiary standard for proving discriminatory use of those strikes had been so demanding that no challenge had succeeded in twenty years. In 1986, Batson v. Kentucky made challenges more direct, offering Black defendants hope while prompting prosecutors to seek other ways to exclude Black jurors. In counties that were 40 or 50 percent Black, defendants such as McMillian could still face all-white juries, especially in capital cases.

Access to the courtroom could also be uneven. During McMillian’s hearing, Stevenson found the courtroom transformed: a metal detector stood at the entrance, with a police officer holding back a German shepherd, and the benches were occupied largely by older white supporters of the Morrison family and the prosecution. McMillian’s family and supporters had been told they could not enter, although the state’s supporters were inside. The scene made access itself part of the contest over whose presence and account the court would recognize.

Vulnerability, judgment, and the record

The book extends the concern beyond McMillian’s case. Trina Garnett grew up amid severe poverty, abuse, and mental-health difficulties. After a fire in which two boys died, she was traumatized and barely able to speak when arrested. Her appointed lawyer believed she might be incompetent to stand trial but failed to file the motions or present the evidence needed to establish that. The lawyer also did not challenge the decision to try her as an adult. Trina was convicted of second-degree murder after her friend testified in exchange for the charges against the friend being dropped. Although the judge found that Trina had no intent to kill, Pennsylvania’s mandatory sentencing law left no room to consider her age, illness, poverty, abuse, or the circumstances of the fire. Her case shows how a child’s vulnerability can be lost inside procedures that treat prosecution and punishment as inevitable.

Even after evidence of innocence is presented, the official record may resist correction. Thomas B. Norton Jr. denied relief in McMillian’s post-conviction case in a terse order of only a few pages. It addressed Myers’s testimony but did not engage with the legal claims or the testimony of more than a dozen other witnesses, and it cited no case law. The order’s narrowness became another form of institutional judgment: a court could receive extensive evidence without meaningfully answering it.

Taken together, these episodes make the presumption of innocence a practical question of power, not merely a statement of legal principle. Who is investigated, who is believed, who can enter the courtroom, and which evidence receives a considered response all shape whether an accusation remains an allegation or becomes an apparently settled truth. The book’s insistence on community testimony, careful investigation, and meaningful review challenges that conversion—and exposes its human cost when it goes unchecked.