The “Gold Standard” of Evidence is Unreliable

The right to a trial by jury in a criminal case is perhaps the most revered element of the American criminal justice system. Putting one’s fate in the hands of fellow citizens, rather than in the hands of the government, provides us with an important protection against government overreach or the politicization of criminal prosecutions. But can one really expect justice if the jury isn’t made up of one’s peers?

Another, perhaps even more important problem, is how a jury relies so heavily on eyewitness testimony to determine what in fact happened. Putting an eyewitness on the stand to either support or undercut the prosecution’s argument that the defendant did what the complaint or indictment said he did has long been considered the “gold standard” of evidence. This is the kind of evidence the jury will give more weight than any other. But the sad fact is that eyewitness testimony is notoriously unreliable.

There are two intertwined problems with the reliability of eyewitness testimony.

The first is the inherent flaws in human memory—our memories are not what we think they are, and they fail us far more often than we know, particularly when we attempt to recall traumatic events like violent crimes.

The second is how police and prosecutors, either inadvertently or intentionally, steer witnesses to believe, and testify, the way they the police and prosecutors want. And the witnesses have no idea they are being steered.

The elusive nature of witnessing and recalling

Your memory is in very few ways analogous to the memory in a computer. Noted psychologist David Kahneman has spent years studying memory, and he has concluded that far from a recording device, memory is driven largely by a process of sorting, weighing, valuing, and, often, construing, with preconceived notions and biases having a big effect on what we remember, and how much. Anticipation is also important. If you are convinced that you live in a neighborhood where violent Hispanic gangs are active, and you get mugged in a dark parking lot, it’s much more likely that you’ll recall that you were mugged by a Hispanic, even if you never got a good look at the perpetrator.

Finally, time is the enemy of reliable witnessing. The longer between an event and trial, the more unreliable the testimony becomes, as the brain’s need to make sense of history and rationalize past behavior can put certainty around a recollection that at the time was uncertain.

The power of suggestion

Police are people too, of course, and subject to the same biases and rationalizations described above. Ladle on top of this the pressure they are under to find and prosecute someone for a particular violent crime, especially one that has attracted media attention. It’s perfectly understandable that police are likely to zero in on a particular “person of interest” and want that person to be the perpetrator.

This group mind-set is compounded by police investigatory methods. When a police officer zeroes in on a suspect in a case, then brings a victim or other witness in for a “line-up,” the power of suggestion can take over. There is often no standard “script” that the witness is read before viewing several people at one time. This provides the chance for the officer to suggest, perhaps subtly, which person is the suspect. And when a witness looks at several people at one time, knowing one of them is the suspect, there’s real pressure to select one of them, even if the witness does not really recall seeing any of these people at the scene. When the witness selects someone, the witness’ mind goes to work rationalizing that selection as correct in order to prevent the witness from suffering a torturous future of responsibility (based in uncertain recollection) for what ultimately happens to the suspect.

Studies have proven that having officers who are not working a particular case speak to witnesses in that case can prevent “suggestion.” Having a witness see a group of suspects one at a time instead of all together also improves results as there is far less pressure to pick one from the group.

Some police agencies have embraced the conclusions of some of these studies. Most haven’t. And why would they? They’d be making their jobs (arresting and successfully prosecuting someone) much more difficult. Police psychologically need to believe that since the defendant gets a jury trial, and the state has to prove guilt beyond a reasonable doubt, convictions of innocent people are rare.

But are they? Referring to this study, Steven B. Duke of Yale Law School wrote:

Even more disturbing are the results of the FBI’s DNA analysis of biological specimens in 10,000 cases from 1989 to 1996. These were all cases in which eyewitnesses had identified a suspect who had been arrested for the crime (usually sexual assault) and biological material from the perpetrator was available for comparison with the suspect’s. In 20 percent of the cases, no conclusive results could be obtained. In the remaining 8,000 cases, however, the suspect was cleared in 2,000, or 25 percent. Assuming that without DNA evidence half of these defendants would have been convicted, then as many as 12 percent of those convicted in disputed eyewitness cases may be innocent.

What is to be done?

The U.S. Supreme Court had a chance to improve things last December. They declined, in an 8-1 decision, that Scotusblog summarized this way:

Holding: The Due Process Clause does not require an inquiry into the reliability of an eyewitness identification when the identification was not procured under unnecessarily suggestive circumstances by law enforcement.

The dissent came from Justice Sotomayor, who spent her first four years as a lawyer working as an assistant New York City District Attorney. It’s easy to wonder how much that criminal litigation experience affected her perspective.

(image: face of scared woman from Shutterstock)


  1. Avatar Noah Kovacs says:

    Fantastic article and a lot valid points. In a lot of ways I think we don’t want to admit that people aren’t always the most reliable of sources. I wonder how many times memories are affected by the “power of persuasion” when giving statements.

  2. Avatar Matt Haiduk says:

    Every time I hear somebody proclaim the guilt of another because “witnesses saw him do it” or “he was identified by the victim” I politely ask them to rent Murder on a Sunday Morning. There they had an identification and a confession. Neither could have been more wrong.

  3. Avatar Matt Haiduk says:

    I believe somebody has uploaded it to YouTube. If you do a lot of criminal, though, it’s worth owning. Not so much for you to watch, but for you to loan out to other people who have no idea what we are dealing with.

  4. Avatar Lukasz Gos says:

    I could go on about that for hours. In fact, it’s basically the thing I’m writing a Ph.D. paper about right now. Nowadays, we’re hyped on “free evaluation of evidence” to the detriment of good old rules, as abused as they were, demanding certain strength of proof. Now what is needed is a strength of convinction, and the bar for that is very easy to lower, if only because nobody can externally measure the strength of conviction in a judge. He claims he’s sure, and that’s enough. The rule of two witnesses especially is one that has gone out the window. We’re experiencing overreaction in the form of courts affirming their right to put people in jail on the sole account of an interested witness, i.e. the injured party being the accuser. It’s politically correct to emphasise that party’s credibility, while undermining the defendant’s credibility on account of his interest in defending himself and avoiding liability is somehow acceptable.

Leave a Reply