Skip to Navigation
University of Pittsburgh
Print This Page Print this pages

April 4, 2013

Books, Journals & More

A closer look: David Harris

harris“If you spend any time in courts, it’s not hard to get cynical,” says David A. Harris, professor of law and associate dean for research. But, he adds, “we have to be trying to improve things, because we know it’s not perfect.”

And it’s appropriate, Harris believes, “to ask hard questions of those who are running the system … about why the system isn’t working the way it should be.”

Harris has been asking the hard questions for years now, in his previous books “Profiles in Injustice: Why Racial Profiling Cannot Work” and “Good Cops: The Case for Preventive Policing.”

To solve the puzzle of why police and prosecutors are fighting the use of scientific tools to solve crimes and instead clinging to older, less reliable investigative methods, Harris undertook his latest work, “Failed Evidence: Why Law Enforcement Resists Science.”

Television’s image of police employing the latest in detection, imaging and analytic equipment, promulgated by multiple “CSI” programs, is a myth, Harris writes in his book. “With the exception of DNA (and then, only sometimes), most of our police and prosecutorial agencies do not welcome the findings of science; they do not rush to incorporate the latest scientific advances into their work.” This is puzzling and irresponsible, he says, leading to innocents punished and criminals going free, which is “a catastrophe for our system of justice.”

The introduction of DNA evidence, with its ability to free people previously convicted — even those on death row — long ago called into question age-old methods of police work. Moreover, exonerations from DNA evidence proved what previously had been “unimaginable” — that people make false confessions.

Then a 2009 National Research Council report showed only toxicology, DNA and a few other forensic methods could be performed without contamination by human influence. Traditional identification lineups, for instance, played on the very human need to pick one among those offered, rather than trying to match each to memories of the crime. Interrogations, when police employ threats and tell arrestees that incriminating evidence exists when it does not, create false confessions in turn. Fingerprinting experts claiming “zero errors,” Harris writes, create an “untenable” claim to perfection.

All of this, he concludes, “represents not just a missed opportunity to do better, but a likely source of future cases in which the train of justice derails and the wrong people pay the price for crimes they did not commit while the real predators and perpetrators remain free to strike again.”


failed evidenceFor “Failed Evidence,” says Harris, speaking in his law school office, “I wanted to focus on why there had been so little progress toward better practices that were well established. Most of the science is not controversial.

“I was all about practical solutions to real problems. I’m not anti-police. I want us to be better at what police and prosecutors do in their very difficult jobs.”

Perhaps most surprising is that Harris’s answers turn heavily on psychological research and principles of organizational management.

“I’m prepared when I do any type of scholarship to go where it takes me,” he says. Looking at law enforcement’s resistance to change “led me to the literature on cognitive psychology and especially to behavioral economics.” The results were a book “just as much about human thinking and human response to new knowledge and the capacity to change as it was about law enforcement.”

Why do those in law enforcement doubt science and trust their intuition and experience more? It turns out that, like all humans, they experience cognitive obstacles to change, Harris writes. Police, for instance, view non-police as outsiders who do not and cannot understand how to do their jobs. They see anything that potentially helps suspects as hurting police. And they see themselves as good guys putting away criminals, not bad guys convicting the innocent.

It doesn’t help that police job performance reviews are predicated partly on making arrests and closing cases.

Thus, science that erodes the public’s confidence in their long-practiced methods, and frees arrestees, threatens the status of police in society.

“It’s fascinating to think that the people with very grave responsibilities are tempted to say, ‘I know best and I don’t want to change,’” Harris says.

“My experience with the police is that they see themselves as the ones with the job of guaranteeing peace and order on the streets and it’s not always a clear and neat process, so they go out and do what they need to do at the moment it needs to be done.” Assistant district attorneys prosecuting cases later may consult with police, “but once the arrest is being processed through the system, [police officers] really feel like they’ve done their jobs” and it’s up to the ADAs to get a conviction.

Getting convictions is the job performance standard for prosecutors, of course, so the ability of scientific evidence to overturn convictions makes it just as unwelcome at trials and hearings.

Judges’ actions — or inactions — have contributed to the situation as well, Harris writes. In a 1993 case, Daubert v. Merrell Dow Pharmaceuticals, Inc., the Supreme Court overrode 70 years of precedent by saying scientific evidence no longer would be admitted to court just because it was generally accepted in the scientific community. Besides acceptance, federal courts now were charged with assessing whether scientific evidence was reliable, had been peer-reviewed, what rate of error had been found and whether it had been replicated. State courts began to follow these new rules as well.

“Judges have responded by being pretty strict gatekeepers on the civil side” of the law, Harris says, applying the new standards to admit or eliminate the science used by plaintiffs suing large companies or the government over pollution, disaster relief, birth defects or breast implants, for instance. “But on the criminal side it’s been very different. Judges have been very reluctant to apply that same Daubert rule to challenges to science” that defense attorneys might use to save their clients.

As Harris points out in his book, today prosecutors successfully challenge defense evidence 66 percent of the time, while defense attorneys only win such challenges 10 percent of the time.


Effecting change is perhaps the most difficult step, Harris says.

“I always want to talk to people about what will work for them and their institutions,” he says — “what will best influence police and prosecutors to go along with best practices.”

Much as peace treaties in wartime come most easily from conservative politicians who can fight accusations of being soft on the enemy, reform in law enforcement is going to come best and most easily from inside police bureaus and prosecutors’ offices; only they can point to their histories of being “tough on crime.” New government-imposed rules also work to shift the practices of police and prosecutors, Harris allows — just not as well.

What shape should these reforms take? Forensic techniques that do not have the backing of good science, such as bite-mark, tire-mark and hair analysis, Harris says, are “rescue-able” with the proper research, but for now “they should be used in ways that are much more limited.” Hairs can be assessed for certain qualities — color and texture — “but we’re a long way from being able to use any of that to identify sources to the exclusion of other sources.”

Fingerprints, ballistics and tool-mark assessment “certainly can be made more accurate and more data-based than they are now,” he says. “It will take work; it will take time.

“What’ll change in the next 10 years,” Harris believes, “is that DNA analysis will become even more powerful than it is. It’ll become faster, much more able to pull apart DNA mixtures … and they’ll do this with a level of precision that is unimaginable now.” He points to a Pittsburgh-based company, Cybergenetics, as leading this push.

“We think in the next five to 10 years we will have most jurisdictions recording interrogations; we will have most but not all police encounters between citizens … on camera, and I think that’ll do a lot to bring the law enforcement practice in line with best practices.”

Lineups for eyewitness identification of suspects also will use the newer technique of serial, rather than simultaneous, display, he predicts, “and forensics will be forced to improve.” In “Failed Evidence,” Harris also points to state innocence commissions and district attorney-run conviction integrity units as solutions helping law enforcement today.

“It’s not going to be instant,” he concludes. “If we have the guts to say, yes, we are human but we can do better, the opportunities for positive change are there. If we know what to do and just turn a blind eye because change is difficult, well, shame on us.”

—Marty Levine

Leave a Reply