Thursday, August 07, 2008

DNA overturns another bad conviction in Dallas based on faulty eyewitness testimony

Texas 34th DNA exoneree, Stephen Phillips walked out of a Dallas courtroom Monday completely cleared in the ten rapes for which he'd been convicted. His case arguably provides the most compelling example yet regarding flaws with historic eyewitness identification techniques and police tunnel vision. Reported the Dallas News:

Mr. Phillips fought his convictions for years but made little headway until the Innocence Project championed his cause. Attorney Barry Scheck said the Phillips case was “one of the worst cases of tunnel vision we’ve ever seen. Police seized on Steven Phillips as a suspect and refused to see mounting evidence that someone else actually committed these crimes.”

Mr. Phillips was identified by 10 different victims, many of whom had described the attackers striking blue eyes. Mr. Phillips’ eyes are green and his wife and two other relatives testified that he was with them when the crime occurred.

As in every DNA exoneration, not only was an innocent punished but the guilty party was never prosecuted for the crimes. The real offender, a man named Charles Goodyear, died in prison in 1998 serving time for an unrelated offense.

Phillips case compellingly demonstrates the need for requiring use of best practices in police lineups. One of the oft-ignored requirement is that fillers for a lineup should be chosen based on the witness description, not because they look like the chosen suspect. I have to wonder if the lineups shown to witnesses had included five blue-eyed people and Phillips with green eyes, would he still have been chosen? One also wonders if a "blind" administrator might have avoided tainting eyewitness testimony and been less likely to steer witnesses toward the wrong man? The Fort Worth Star Telegram added a little more background on the case:

Phillips was convicted of rape and burglary in two trials in 1982 and 1983 for charges stemming from the same crime. He was eventually linked to nine similar crimes that took place in spring 1982 in the Dallas area, according to the Innocence Project.

The crimes were committed by someone wearing a hood who forced women — sometimes in large groups — to strip and perform sexual acts at gunpoint.

Although Phillips argued that the victims had misidentified him — and his former wife provided an alibi — the juries convicted him anyway, giving him two 30-year sentences. Afraid of getting a life sentence for the other crimes because he was getting "smashed in court," Phillips said he took a plea bargain in the other cases.

"Once they got it started, they could not turn off the machine," Phillips said. "The truth had already slipped away."

Phillips’ request to have the DNA retested was denied by the Dallas DA’s office in 2001. But current Dallas County District Attorney Craig Watkins agreed to re-examine the scientific material in 2007.

5 comments:

Anonymous said...

I'm shocked! Shocked!

A Voice of Sanity said...

If ever I am forced to be in a lineup, I plan to object to the form of it just to get it on the record. Based on what I understand, the only satisfactory way to run one is to have each person alone walk into the room and have each witness say 'yes' or 'no' to them when they walk out. As they are run, everyone together, I believe too often it comes down to "He is the closest".

Soronel Haetir said...

It gets even worse when looking at historicalphoto lineup procedures. Things like having the suspect being the only color photo in the set and plenty of other malfeasence. Couple that with absolutely no recording of what happened other than the witness picking out the suspect and you have a setup that is ripe for abuse.

Anonymous said...

I am surprised that such line up practices are still carried out. I am in the UK and the rules on line ups were one of the items covered under the police and criminal evidence act of 1984 (PACE). The code of practice gives rules for video id parades, line ups, group identification, photo identification and confrontation identification. It also gives rules that the police carrying out the id parade must not be involved with the investigation. If the hyperlink work. you can read the recommendations.

http://www.icva.org.uk/site/downloads/PACE05/PACE_Chapter_D.pdf

Anonymous said...

I do like the bit in the rules where it says where police officers are the subject of an identity parade, identifying badges and numbers must be covered up!