As detailed in Part One of this series, Gonzales was on the hook for a serious offense, facing upward of 30 years. The two men who held up the bakery had pointed a gun at the head of the bakery owner’s one-year-old son as they demanded cash from the register. Gonzales was looking at five charges, including aggravated robbery, criminal extortion and child abuse. The eyewitness identification weighed heavily against him, but the facts of the case didn’t quite add up. So when memory expert Dr. Edie Greene testified about what she saw as the problems with the way the witness identification process had been conducted — mainly that his photo stood out from the others for its unique tint — the witness, in a made-for-television-style courtroom turn, hesitated.
As Gonzales remembers: “They put [the bakery owner] back up on the stand, pointed to me, and asked her, ‘Is that the man that robbed you?’ She looked right at me, paused, and said, ‘No, that’s not him.’ [My lawyer] leaned over to me with her jaw dropped.”
When the witness, visibly shaken, fled the courtroom, Gonzales’s sense of doom began to lift.
“Then I knew maybe I had a chance,” he said.
After Dr. Edie Greene’s testimony and the eyewitness’ sudden but decisive change of heart, the jury was convinced of Gonzales’s innocence. And so, the verdict came down: Not guilty. Gonzales described feeling an immeasurable amount of relief, which quickly turned into anger.[pullquote]“I just lost a whole year of my life,” he remembers thinking, “It was just a big, long sham and then the real robbers are still out there.”[/pullquote]
“I just lost a whole year of my life,” he remembers thinking, “It was just a big, long sham and then the real robbers are still out there.” Still, despite everything, he says that to this day, “I find myself very lucky, especially when there are so many people who aren’t so lucky.”
He’s right that a lot of suspects aren’t so lucky. Disturbing numbers of innocent people serve time for crimes they didn’t commit — a fact that came to light when the rise of DNA testing in the late 90s started shoring in exonerations. The Innocence Project, a national litigation and public policy organization dedicated to understanding and preventing wrongful convictions, started examining all the data newly available from the DNA-based exonerations. It found that eyewitness misidentification played a role nearly 75 percent of the time in wrongful convictions. So when an innocent person finds him or herself behind bars in America, more often than not, eyewitness misidentification is to blame.
For the Project’s policy advocate focused on Colorado, Amshula Jayaram, the injustice is two-fold: “for every conviction of an innocent person, the true perpetrator is still out there.” That makes wrongful convictions not just a human rights issue, but also a public safety problem.
The two guys who robbed the bakery in Lakewood all those years ago were never brought to justice. And the victim still lives with the trauma of having seen a gun pointed at her child’s head.
Criminals, especially violent ones, who get away with it aren’t likely to stop there. Looking at the more than 320 known wrongful convictions related to an eyewitness misidentification, the Innocence Project found that the real perpetrator was brought to justice in just under 40 percent of cases. But of those perpetrators who did eventually end up incarcerated, most had committed a whole slew of other violent crimes while an innocent person served time in their place. “The only person who actually benefits from a wrongful conviction is the true perpetrator,” is Jayaram’s frank assessment.
But even though the string of DNA-based exonerations over the last two decades has somewhat raised awareness about the prevalence and danger of eyewitness misidentifications, McGuire notes that awareness alone didn’t protect her client from the fate he came so close to getting deal. “I think about [Gonzales] all the time. His case could’ve turned out completely different,” she says.
Was it a near miss? “It was just a great win,” McGuire reflects, “I mean, thank God…”
Amshula Jayaram points out that “many of these wrongful convictions stemmed from preventable problems, particularly with advances in scientific research.” Indeed, the body of scientific research in question is Dr. Greene’s focused area of academic enquiry — namely, the intersection of cognitive psychology and criminal justice.
Dr. Elizabeth Loftus was a professor at the University of Washington in the 1970s when she started researching episodic memory — our ability to recall explicit and specific information about a particular event. Studies in this vein follow a basic methodology: subjects are shown a staged event (usually in the form of a slideshow or video), then researchers describe the event back to them. The test group receives a description containing misinformation; the control, an accurate one. Rather unsurprisingly, the subjects who were fed misinformation were far more likely to remember the event incorrectly. As it turns out, our memories are quite suggestible.
Over the past four decades, further research by Loftus and others on eyewitness memory has revealed some other surprising — and some unsurprising — truths.
There are all sorts of factors that affect the accuracy of memory, some of which just boil down to the contingencies of a given experience. Things like duration, physical proximity, field of vision, and level of cognizance matter when recalling the various minutiae of an event.
But researchers also began to find that the human factor matters, too. Receiving feedback — whether it be verbal, gestural or even a slight twitch of a lip — from the questioner sways our answers. We’re especially bad at identifying someone of another race (because, you know, “all [ insert ethnic group here ] look alike.”) And though some may claim that an adrenaline rush throws even the most subtle particularities into vivid clarity, research shows that our perception of minor but crucial facts actually clouds during such moments of trauma.
On one level, it all sounds a lot like common sense. People are known to sometimes remember things incorrectly. But taken outside the lab and into the realm of criminal justice, a shaky memory poses serious risk.
Eyewitnesses are often crucial in the prosecution of violent crimes like murder, rape and robbery — the type of cases that carry the heaviest sentences. Twenty people in the U.S. have sat on death row before getting exonerated, so it’s not a stretch to say that our legal system relies on the faulty faculty of memory in matters of life and death.
And that, many would say, is a gaping hole in the pursuit of justice.
Patching the hole
Loftus herself began to focus on criminal prosecutions in 1974 when an article she wrote about a murder trial ushered in a flood of calls from lawyers and judges asking for legal advice. In the state of Washington, she set a nationwide precedent by appearing in court as the first expert to testify on the science behind eyewitness memory. Now, in many states, defense attorneys can request that a judge allow a cognitive psychologist to be named as an expert witness in cases that rest on an identification procured using questionable lineup practices. In some states, the judge will read instructions to the jury explaining certain things to consider when weighing an eyewitness identification. The hope is that being made privy to some of the complications in the evidence, a jury can make a more informed decision on the defendant’s guilt or innocence.
As it is, a sketchy eyewitness identification can only be challenged long after the procedure is over and done with. Once it’s procured in the investigation, that’s the evidence that sticks. So the defense can try to get it suppressed before it gets admitted in trial or, if that fails like it did in Gonzales’s case, by bringing in an expert witness to testify about its unreliability. But, not every defense attorney is successful in getting an expert witness admitted, let alone knows or cares to do so. “If it’s not on someone’s radar, it has to be,” McGuire says about the available strategies to litigate against a misidentification.
Knowing what we do about flaws with eyewitness memory, we are in a position to deal with the reliability issue preemptively in the investigation rather than retroactively in court. Stakeholders in the criminal justice system have long had the opportunity to improve the methods used by law enforcement to retrieve and record memory. And now, the pieces are starting to fall into place to capitalize on that opportunity.
As things stand, it’s up to individual agencies to set their own lineup policies, so the exact methods used are highly variable across and even within jurisdictions. Some local police departments are starting to review and reform lineup policy on their own volition. Greenwood Village Police Chief John Jackson says his department is in the midst of reform now. “You can have an improper identification, you can have some kind of bias in the procedure,” he has noticed during his time in law enforcement. Now, as head not only of his own department, but also of the Colorado Association of Chiefs of Police, he thinks “there are some steps we can take to make these things better.”
But even in a department that’s made reforms, there’s no compliance mechanism in place that guarantees every lineup is air-tight. And that, some in law enforcement argue, is nobody’s fault: “They’re only human,” says Annmarie Jensen who lobbies on behalf of police in Colorado, “Things are going to happen.”
Swift and/or fair
Once a misidentification makes its way to court, the defendant is far-and-away the most at risk. Still, prosecutors too are on their heels when dealing with an unreliable piece of evidence at the center of a case.
Tom Raynes, executive director of the Colorado District Attorneys’ Council, says that “there’s some hesitation” on the prosecutorial side about bringing the complications of eyewitness identification to the fore, provoking questions like “How will it impact the trial? Will it damn my case?” But, at the end of the day, Raynes insists that everyone involved in the legal process has an interest in getting it right: the fact that wrongful convictions happen is “gut-wrenching, it’s horrible, there’s nothing worse you can imagine,” he says. “If there’s an innocent person sitting in prison, he adds, “I want to know. I want to find them and I want to release them.”
But even though everyone might agree they should avoid putting innocent people behind bars, there’s still an expectation for justice to be served in a timely and conclusive manner. So prosecutors are under pressure to win cases — lots of them. For DAs who are politically elected and judged largely on their conviction rates, that pressure can be especially acute.
In the bakery case, defense attorney Kathy McGuire thought her motion to suppress the eyewitness identification “should have caused some pause in the prosecution,” she recounts with some residual frustration. “But it didn’t. I remember it didn’t matter.”
“I think there is a dynamic of ‘we want to win our cases’” says Maureen Cain, policy director for Colorado Criminal Defense Institute, speaking from a career’s worth of experience in the multi-faceted legal system. “However,” she continues pointedly, “the extent to which [landing convictions] becomes super important over seeking justice, I think that depends on the individual DA.”
Raynes, for his part, acknowledges that prosecutors do sometimes take a full-steam-ahead approach. “It’s competitive, people want to fight hard, move up the ladder,” he says. But he rejects the notion that justice might get steamrolled along the way: “It’s a popular thing on TV shows, but it’s not about that.” Of all of the prosecutors he has worked with, he says he hasn’t known any to count wins and losses. If anything, he says the numbers game is in the sheer volume of cases piled on a prosecutor’s desk. “Quite frankly, when you get a stack of 500 cases, the pressure is how to get out of cases,” he says, “So when you see something shaky, it’s a blessing.”
Whether Raynes’ attitude is the exception or the rule is hard to say. As Cain put it: “It depends.” There are 22 DAs in the state, and, of course, defendants don’t get to choose which one prosecutes their cases. So whether their DA is of the righteous ilk also depends.
Nip it in the bud
Rather than bank on individual players in the criminal justice system getting wise to the existing science behind eyewitness identification on their own, reformers like Jayaram and Cain advocate another approach: codifying policy that would make the most up-to-date best practices in eyewitness identification mandatory and universal in the state of Colorado. Since 2001, eleven states including Oregon, Texas and Ohio have standardized the lineup procedures used by police in criminal investigations, either through legislative reform or court action.
Colorado could be next.
Practically speaking, the proposed reforms aren’t a radical departure from how things generally go — they’re a set of minor tweaks that, together, do a lot to safeguard against common pitfalls. The Innocence Project calls them the “Core Four.”
The first of the Four is blind administration, or having the officer who conducts the lineup be one who’s unaware of who the suspect is. This practice prevents the lineup administrator from consciously or unconsciously giving the witness any kind of feedback or cue. Whether it be verbal or in the form of body language or subtle physical reactions, administrator feedback has been shown to have a significant effect on the witness’s decision-making process. A blind administrator by nature can’t affect the outcome.
One concern raised over blind administration is that smaller police departments simply don’t have the personnel to be sure there’s an officer uninvolved in the investigation on-hand. Especially in rural departments, staffing a blindly administered lineup may not be feasible. But, there’s an alternative method that Jayaram calls the “folder shuffle method” which basically involves placing photos in different folders, shuffling them and handing them to the eyewitness or victim, one by one, so that the officer is unaware of the photo being viewed. (No peeking.) It’s simpler than the most simple card tricks and can be done by the lead investigator on a case.
The second reform is the one that would’ve done the most to prevent Gonzales from getting charged in the first place: Proper filler selection. Every lineup, whether it’s in-person or in photos, consists of one or more suspects and a handful of fillers. Though it seems almost too obvious to state, it’s crucial that the fillers consistently match the original description of the perpetrator, ie. same race, gender, physical specifications, etc. And the suspect should not stand out in anyway — like, say, being photographed against a different colored background than all the other fillers.
The third of the Core Four reforms entails the lineup administrator instructing the witness that the real perpetrator may or may not be in the lineup and that the investigation will continue regardless of whether an identification is made. This practice lifts the pressure a witness might feel to definitively identify the perpetrator — a pressure, real or perceived, that could compel the witness to go out on a limb for the sake of expediting the investigation.
The last of the proposed reforms the immediate eliciting of a confidence statement from the witness. That is, once an identification is made, the witness is asked to write down, in his or her own words, the level of certainty he or she feels in the selection. Having a signed record of confidence becomes very important later in trial when the jury needs to understand exactly how to weigh the reliability of the eyewitness’ identification.
Though the Innocence Project has certain other recommendations to safeguard the protocol even further, the Core Four are the staples of a good eyewitness identification.
There’s a lot of weight behind the push for reform: the U.S. Department of Justice, the American Bar Association have endorsed the Innocence Project’s recommended best practices for eyewitness identification and the International Association of Chiefs of Police have issued their own recommendations that are consistent with the Project’s.
The most recent and high-profile endorsement of these reforms came this past October from the National Academy of Sciences, when it released a landmark report “Identifying the Culprit.” Through a thorough review of the existing body of research on memory, the report states that “the fidelity of our memories to actual events may be compromised by many factors at all stages of processing, from encoding to storage and retrieval,” leading to a forceful call for “new law enforcement training protocols, standardized procedures for administering line-ups, and improvements in the handling of eyewitness identification in court.” Such a definitive report from the nation’s preeminent scientific body is seen as a message that the science is settled on the eyewitness identification, and it’s time for the criminal justice system to act.
“The NAS report was a huge milestone,” Jayaram notes. “Sometimes policy change is a little like two steps forward, one step back. But In Colorado, we really are moving forward. I think there’s a lot of momentum behind it.”
That momentum here in Colorado began last year, with the formation of a working group that brought together all the relevant, albeit typically adversarial, sides of the issue.
An unlikely alliance
Last Spring, the Attorney General’s office and various stakeholders formed a roundtable committee to start examining and devising best practices in different areas in law enforcement. For Raynes, who has sat on this Best Practices Committee since its inception, it was an opportunity to bring people together to “look around the country, see what’s going on and figure out what the most pressing issues are.” Shortly after the committee set out on this course, representatives from the Innocence Project came to Denver to meet with various players in Colorado’s criminal justice community. It was during this visit that Raynes was introduced to Amshula Jayaram. They got to talking, and Jayaram suggested Raynes and several other prosecutors attend an upcoming symposium on eyewitness identification best practices in San Francisco.
“They don’t reach out to us [prosecutors] much,” Raynes notes with a laugh. But he describes his experience at last summer’s symposium as enlightening: “From our world, the presentation did feel a bit slanted, but we walked away saying, ‘there’s some really good stuff here.’” After the symposium, “we went back to our committee and said, ‘Hey, we need to bring this to a front burner.’”
A bill in the works
Since then, members of the Best Practices Committee have put in painstaking work to issue an eyewitness identification model policy. That model policy serves as the basis for legislation which would require police to implement written best practices. As a default, departments can just adopt the Best Practices Committee’s model policy if they do not choose to write their own. The bill, SB15-58, is sponsored by chair of the House Judiciary Committee Rep. Daniel Kagan and chair of the Senate Judiciary Committee Sen. Lucía Guzmán, both Denver Democrats. Titled “Eyewitness Identifications Policies And Procedures,” the measure is set to be heard at the Capitol this afternoon.
Rep. Kagan says that when he himself was a practicing criminal defense attorney, he “found that time and again the IDs were wobbly.” Now that he’s a legislator, Kagan is confident, especially in the light of growing body of scientific research, that police lineups “are a notorious source of inaccuracy and miscarriage of justice,” but that “we also know how to remedy it.” That’s why he has used one of his five available bill sponsorships to push for eyewitness identification reform. “We’re constantly striving to perfect and improve the criminal justice system,” he says, “because that’s our charge [as lawmakers.]” Though at this point it’s yet unclear how much support the bill will garner, Kagan is “hopeful because both sides of the aisle are in favor of better, more accurate outcomes in the justice system.”
Dennis Maher, who was exonerated in 2003 after having served 19 years in Massachusetts for two rape and a sexual assault he didn’t commit, says that he can’t turn back the clock on what happened to him, but he can do everything in his power to spread the word about the danger of eyewitness misidentification and the importance of DNA testing. “I could be an angry and bitter person but then I wouldn’t have what I have now,” he says, that being his wife, two kids, house and nice job. “I made a choice to let go of the anger.” Now, Maher does a fair amount of public speaking at universities and in legislatures. “I testified about the DNA laws in Massachusetts and some of the senators were in tears,” he remembers. Maher hopes that nobody ever has to go through what he went through.
Lukewarm on reform
Despite the hopefulness of some, not not everybody thinks legislation is the best approach. Though Greenwood Village Police Chief Jackson agrees that “there are things we can do to make these things better,” he thinks that “the Legislature does not need to fix the problem.” The Colorado Association of Police Chiefs is officially “neutral” on the upcoming eyewitness identification reform bill, not wanting to go so far as to lobby against it, but still not necessarily enthused about getting handed a mandate from politicians.
Raynes of the DAs’ council acknowledges that Jackson’s attitude toward the mandate is shared by many in the law enforcement community. “There’s always a danger when you start micro-legislating,” he says. “The initial reaction is, ‘Why are you jumping on me? You don’t even know what I do, so how can you tell me what to do?’”
The police chief’s lobbyist at the Capitol, Annmarie Jensen, explains that “police just don’t think a bill mandating best practices is needed because they’re already doing them.” Chief Jackson’s department in Greenwood Village is already doing them and Jensen says every member of their legislative committee is doing them, but that leaves a good number of departments unaccounted for.
Codifying every last policy and procedure can also sometimes create unforeseen problems or or restrict adaptation down the line, and though that was initially a concern among law enforcement, the eyewitness identification reform bill has gained enough speed at this point that outright opposing it would be more trouble than it’s worth. That said, Jensen says that especially with limited resources and a lot going on this session, “I don’t think we’ll go so far as to support it.”
No time like the present
All that told, reform advocates feel that the time is ripe to put best practices into law. “A lot of people have been working on this for a long time,” notes one of those people, Maureen Cain. “But I think we’re seeing recently a heightened interest in issues related to police/community interactions.”
The undercurrent running through any debate over police practices right now is hard to ignore. The killing of Mike Brown in Ferguson, Eric Garner in Staten Island and Marvin Booker here in Denver struck a chord in public opinion — a chord that doesn’t sound like it’ll dissipate any time soon. Every non-indictment just stokes the flames more, as distrust and resentment of law enforcement quickly becomes ingrained in communities.
In July, 20-year-old Ryan Ronquillo was shot 12 times by Denver police and just last week, 17-year-old Jessie Hernandez suffered a similar fate. In both cases, cops say the teens were behind the wheels of stolen cars. This being the fourth incident in seven months in which police shot someone supposedly using a car as a weapon, the Denver Police Department is now reviewing its policy around firing weapons into vehicles. Hernandez’s angry and grief-stricken cousin got face-to-face with prosecutors at Denver DA Mitch Morrissey’s office last Tuesday, asking “how would you feel if your kid got killed?” Morrissey will decide whether to charge the officers, one of whom is recovering from a leg injury after Hernandez allegedly struck him in a car. Hernandez’s family is demanding a federal investigation.
So tension with police is high right now — in Denver and across the country. To quell what’s quickly turning into outright animosity, law enforcement is doing as much as possible to engage with the public in meeting rooms rather than in riot gear in the streets. “I think what we’re missing in the narrative is that the average Joe/Jane becomes a cop because they want to help people,” Jensen, the lobbyist, says. “Police work brings them into contact with the darkest stuff out there every day. We forget how hard that is on the human psyche,” she added. Mostly, cops get invited to forums by community members, not the other way around. There’s one scheduled nearly every day in Colorado, Jensen remarks.
Any way you shake it, the zeitgeist is such that a lot more people are talking about the way our criminal justice system works (or doesn’t work) than before, making it an opportune time to make real assessment and improvement. This session, we can expect to see upwards of ten bills addressing various policing and criminal justice issues, including one addressing police body cameras and one amending the restraint practices used on juvenile offenders. Cumulatively, regardless of whether they’re actually related, these specific issues are all part of “a growing interest in what law enforcement is doing,” Cain explains.
As for Mario Gonzales who narrowly escaped a hefty prison sentence for charges based on a sketchy eyewitness identification, all that’s left for him to do is be grateful for his freedom every day. He now has two kids of his own and works at a hotel in Michigan, where he moved after he couldn’t find work in Colorado. Having been incarcerated for a full year made it impossible to get hired here. Even his old stint as a day-laborer was tough to get back into.But all that is in his past now. Gonzales chuckles ironically when asked to recount his experience of getting falsely accused all those years ago. The details are hazy, he says. Memory is a slippery thing.[Featured graphic, license-free composite by Nat Stein.]