One of the most enduring scenes of the 1995 Oklahoma City bombing is the image of suspect Timothy McVeigh being led in shackles and leg irons from a county jail to a waiting car.
Now, almost two years later, attorneys for Mr. McVeigh are hoping memories of that same image will force a federal judge to toss out a portion of the government's case against the accused terrorist.
At issue is whether the massive media coverage of McVeigh's jailhouse walk influenced the government's eyewitnesses to positively identify McVeigh as being connected with the bombing.
The Oklahoma City case, scheduled to go to trial in March, will provide a new test of the reliability of eyewitness accounts - one of the most heavily used and effective tools in a prosecutor's arsenal.
In jury trials, eyewitness testimony can be more incriminating than fingerprints or DNA evidence. But experts say that mistaken identification by eyewitnesses is also a major cause of wrongful convictions. Thus the reliability of their testimony is emerging as a central issue in the pretrial maneuvering leading up to the March 31 case.
"Eyewitness identification evidence in a classic sense is direct evidence of guilt. In some ways it is better than fingerprints," says Gary Wells, a professor of psychology at Iowa State University, and a nationally recognized expert on the reliability of eyewitness testimony.
In a motion filed in federal court in Denver, Stephen Jones, defense attorney for McVeigh, argues that seven government witnesses should be excluded from the trial to prevent their tainted testimony from influencing the case.
Mr. Jones says that by the time investigators presented photo spreads to the government's prospective eyewitnesses, McVeigh's face had already "became so well known that monks living on the mountainside in Tibet could have made the same identification."
Justice Department spokeswoman Lisa Brown says prosecutors will file a reply brief in two weeks. She declines any further comment. An FBI spokesman in Washington says his agency has no comment.
Although the FBI had interviewed some of the eyewitnesses prior to McVeigh's jailhouse walk, none of them were asked to identify McVeigh in a controlled photo spread or a police lineup until after pictures of him were widely distributed following his jailhouse walk.
Experts who study eyewitness testimony say that such a procedure undermines the reliability of anything an eyewitness may later say on the witness stand at McVeigh's trial. They say it makes it impossible to determine whether the eyewitnesses will testify about their recollection of the man they believe they saw prior to the bombing, or their recollection of televised images of McVeigh being led in shackles to a waiting car, or a combination of both.
The purpose of introducing eyewitness testimony at a trial is to offer the jury independent verification that the defendant is the same person who was seen engaging in aspects of the alleged crime. Such independent verification comes when the eyewitness positively identifies the same person in a police lineup or photo spread, without any outside help from investigators as to who their suspect is.
Such identifications can be difficult because most people don't pay close attention to the physical characteristics of strangers, experts say. Nonetheless, during trials juries pay close attention to the testimony of eyewitnesses.
McVeigh's arrest came as a result of what has been described as the largest manhunt in the nation's history. He was apprehended after FBI agents obtained a composite drawing from workers at the body shop where McVeigh allegedly rented the Ryder Truck used to carry the bomb.
A county deputy in Oklahoma noticed that McVeigh, who had been arrested earlier on a traffic violation, looked like the bombing suspect in the composite sketch.
It remains unclear why the FBI did not attempt to present photo spreads to the body-shop workers prior to McVeigh's much publicized jailhouse walk.
Among other prospective government witnesses is a gas station worker in Billings, Okla., who told the FBI he believes he sold gas to a man driving a Ryder truck in the early morning before the bomb blast, according to court documents. He contacted the FBI after he saw McVeigh on television and said he was now "sure" McVeigh was the same man he'd sold gas to.
Another possible government witness is a man who drove near the federal building in downtown Oklahoma City shortly before the blast. He told federal agents that he saw a man in his mid- to late 20s with clean-cut hair and wearing blue jeans get out of a Ryder truck parked in front of the building. The witness told the FBI that the man looked similar to McVeigh but that he could not tell if it was McVeigh, according to court documents.
Experts say investigators must use extreme caution when administering photo spreads and lineups because mistaken identification by eyewitnesses is the principle cause of wrongful imprisonment in the US justice system.
If law-enforcement authorities in any way influence the prospective witness toward a particular suspect, the witness is likely to follow the investigators' lead. And that can lead to false testimony against an innocent man.
A June 1996 study showed that in 24 of 28 cases involving US prisoners who were freed in recent years as a result of new DNA evidence that proved their innocence, eyewitness testimony had presented the most compelling evidence leading to the false convictions. The study was funded by the Department of Justice.
In his own research, Professor Wells has found that the testimony of a prospective witness can be significantly distorted if police notify the witness that he or she has identified the person they had identified as a suspect. The witness accepts that information and uses it to bolster his or her own fading memory.
At trial, that same witness is prone to testify with much greater confidence than other eyewitnesses who received no feedback from police. It is often an honest mistake by witnesses, he says, but a mistake that can lead to injustice. "When people made errors and falsely identified someone from a lineup or a photo spread often they were certain they were right," Wells says.
Wells says investigators could eliminate the problem of distorted eyewitness testimony by conducting photo spreads and lineups as if they were scientific experiment. He says the agent administering the photo spread should himself be unaware of the identity of the suspect to prevent influencing the prospective witnesses.
"My approach is let's find ways to make eyewitnesses as reliable as jurors expect them to be," Wells says. "Let's put in guarantees that the system itself is not contributing in any way to false identifications."