The problem of false confessions

The case seemed clear. Colleen Blue had confessed to killing her baby.

The story was that Blue, a 41-year-old homeless woman, had wrapped her newborn in clothing, slipped a vinyl bag over his head and left him behind a grocery store in far north suburban Round Lake Beach.

Before proceeding with the murder prosecution, though, prosecutors took an extra precaution to strengthen their case. They ran DNA tests on the suffocated baby.

The tests had an astonishing result. Blue was not the mother of the child. Police still suspect that Blue may have a connection to the baby's death, but the DNA evidence contradicting her "confession" was a stunning development.

Unfortunately, it's not an uncommon development. People have spent years in prison for crimes to which they supposedly confessed, but didn't commit.

They may spend years trying to get someone to believe their confession was false. If they're fortunate, some advocate will decide to mount the long legal battle required to prove innocence. If they're not fortunate, they're stuck in prison.

False confessions occur in particular among suspects who have psychiatric problems, as appears to be the case with Blue. Other times, it involves those who are young, those with minimal language skills or low IQ, those who are coerced or tortured, or who are eager to please or particularly susceptible to outside influence.

Two little boys, ages 7 and 8, falsely told police in 1998 that they killed Ryan Harris, the 11-year-old who was riding her bike through the neighborhood when she was knocked off, sexually molested and murdered.

This month, evidence mounted in the case surrounding Lori Roscetti, the medical student who was raped and murdered in 1986, that the four men convicted of the crime didn't do it. Two of those men, Calvin Ollins and Marcellus Bradford, confessed to police. But new DNA test results of two pubic hairs found in Roscetti's car, along with earlier tests on roughly two dozen semen samples taken from the woman and her clothes, fail to link any of the suspects.

False confessions may close cases, but they don't solve crimes. When the truth comes out years later, police are forced to start their investigation over from the beginning, except all their leads have gotten cold and witnesses' memories have faded. That is why it makes sense to videotape all custodial interrogations and confessions. Only two states do this as a matter of law or policy--Minnesota and Alaska--but it's far past time when law enforcement agencies nationwide should follow course.

It is the most sensible, efficient way to protect suspects from police brutality and law enforcement from false charges of police brutality. And a videotaped confession, coupled with documentation of the interrogation leading up to it, can be a prosecutor's best friend in winning a case.

In Michigan, two teenagers last year spent six months in jail for a pizza house murder they did not commit. They won release from jail last spring only after someone else confessed to the killing and passed a lie detector test about his involvement. The damage done to the credibility of police officers who took the undocumented confessions was such that legislators and even prosecutors in Michigan are starting to call for mandatory videotaping of all custodial police interrogations.

How many more false confessions will it take for legislators, cops and prosecutors elsewhere to realize the same thing?