There are few experiences more devastating than being accused of a crime of which one is not guilty. It is even worse when the accusation could lead to the death penalty. A society likes to think of its judicial system as fair and just, yet evidence has been mounting in recent years that innocent men and women are tried, sent to prison and even executed for crimes they did not commit. Various databases spell out the evidence. A database maintained by the Justice Institute, a small nonprofit organization, currently lists 4,401 cases in 109 countries where 584 people have been sentenced to death and 789 have been sentenced to life in prison for crimes they did not commit.
One of the oldest cases is that of John Gordon, wrongly convicted of murder and executed in Rhode Island in 1845 after the judge instructed the jury to give more credibility to Yankee depositions than to those of the newly arrived and suspect Irish. The most recent case is that of Tony Yarbough. Condemned for stabbing and strangling his mother, half-sister and her friend in Brooklyn in June 1992, his murder conviction was vacated on DNA evidence after he had spent 22 years in prison. On Aug. 7 he told The Daily News: “If there had been a death penalty in New York, I would be dead by now.” The common reaction of the general public to convictions is to presume they are just; but these cases are “life issues” that demand an institutional, moral and legislative response.
The Death Penalty Information Center, another national nonprofit organization, documented 143 defendants sentenced to death and later exonerated since 1973. Of those released, 14 had been in prison for over 20 years. A record 87 people were exonerated in 2013 from past convictions in various types of criminal cases, according to the National Registry of Exonerations. In the case of Nicole Harris, who in 2005 was sentenced in Chicago to 30 years for killing her son, the police had coerced a false confession. Most appalling, in some cases the person had already been executed before innocence was established. Among the best known cases is that of Cameron Todd Willingham, accused in Texas in 1991 of burning down his own house to kill his three young children. After his conviction, a nationally known fire investigator examined the scene and found no evidence of arson.
Until recently these studies have been the work of university researchers, civil rights activists and journalists, exemplified by The Innocence Project at the Benjamin N. Cardozo School of Law of Yeshiva University in New York, which published Actual Innocence in 2000. The book describes how false convictions happen and celebrates the introduction of DNA analysis as an instrument for justice. More recently, a report of the U.S. Department of Justice and the International Association of Chiefs of Police, titled “National Summit on Wrongful Convictions: Building a Systematic Approach to Prevent Wrongful Convictions” (August 2013), includes 30 action recommendations. This is a milestone in engaging police and prosecutors in correcting the procedural shortcomings exposed by DNA profiling.
A catalogue of factors that produce bad outcomes, familiar to viewers of television police procedurals, includes the following: eyewitnesses are the least reliable source of information; tortured suspects will lie to their torturers and implicate others to get reduced sentences for themselves; suspects, sleepless and depressed, will confess to almost anything; people of color from crime-filled neighborhoods already have two strikes against them; lab tests are rigged; defense lawyers with overwhelming caseloads cannot fight for their clients. Ambitious prosecutors and politicians trumpet “law and order” as a campaign theme, as if a murder trial were a spectacle in ancient Rome.
Most of the report’s recommendations concern day-to-day practices: record all interviews involving major crimes; review cases to avoid investigative bias; do not accept jailhouse informers without corroborating evidence; create risk-assessment tools to help identify wrongful arrests; identify the five most frequent errors that lead to wrongful convictions. But the emphasis is on the most problematic investigative tool, eyewitness identification. The current practice of the lineup ought to be replaced by double-blind/sequential protocols, says the report. The investigator, for example, should not conduct the witness’s viewing of suspects in a lineup, and a “blind” administrator, who does not know the suspect, should be the one who presents photos to a witness for identification.
Bureaucracies resist change, and some law enforcement officials may drag their feet at taping all interrogations or limiting rough treatment. The hope, of course, is that the new level of care applied to this stage of an investigation will set the tone for every step. This may lead to fewer convictions, but it will at least help restore confidence in a system badly in need of it.