Why a Mixed Bag in Michigan on False Confessions?

With thanks to the Wrongful Convictions Blog, I learned that Michigan’s Supreme Court has  issued a decision on the issue of false confessions.  It’s one step forward and two steps back.

This type of testimony can be vital, given the power of confessions to convince juries.  According to Richard Leo, a psychologist/law professor and author of the multiple-award-winning book Police Interrogation and American Justice (Harvard, 2008), most people cannot accept that an innocent person would ever confess to a serious crime.  (Full disclosure: Richard Leo provided an endorsement for my book Failed Evidence.)  Despite the fact that 25 percent of the 300 cases of defendants exonerated based on DNA featured a false confession or false plea of guilty, people cling to the myth that no one would confess falsely, absent torture, mental infirmity, or lack of sobriety.

Thus it was heartening to see People v. Kowalski, decided July 30, in which the Michigan Supreme Court declared that a defendant could present expert testimony on why a person might confess falsely.   While an expert could not comment on truthfulness of the defendant in the case, expert testimony on false confessions and how they happen is admissible because what lies behind false confessions is “beyond the common knowledge of the ordinary person….”

So far, so good.  But the court then went on to exclude the expert testimony in the case, based on Michigan’s version of Evidence Rule 702 and the Daubert case.  The court said that the methodology used to generate its conclusions was unreliable.

That testimony was given by Professor Richard Leo  — the same person who wrote Police Interrogation and American Justice, and so many of the other seminal works that help us understand today that the idea that an innocent person would never confess is a myth.  Professor Leo’s body of work has been widely published, and much of it has been peer reviewed — two important factors for reliability under the Daubert rule.  And the research and conclusions he presented in Kowalski were generated independent of any work he did as an expert in the case.

What gives?  It’s an example of a phenomenon I discuss in Failed Evidence: the Daubert rule is applied much more stringently to evidence advanced by the defense than to evidence that comes from the prosecution.  More on that when the book is published, in September.

DOJ New Orleans Consent Decree: Eyewitness Identification

In yesterday’s post on the DOJ/New Orleans PD consent decree, I pointed out that early news reports said that the decree required a science-based best practice: audio and video recording of suspect interrogations.  Now that we can read the full consent decree itself, we can see that it requires science-based methods of conducting photo lineups with eyewitnesses, too.

Paragraphs 171 through 173 of the decree require three of the most important changes that scientists have long recommended: blind line ups, in which the administrator of the lineup does not know which of the pictures shown is the actual suspect; telling the witness that the perpetrator may or may not be present; and the proper selection of “filler” photographs.  All of these procedures are fundamental to avoiding cognitive biases, and with them erroneous identifications.

We also see in paragraph 164 of the decree that the requirement to record interrogations applies only to suspects in homicides and sexual assaults, but that the recording must include the entire questioning session between suspect and police.  Police will not be allowed to do an unrecorded “pre-interview,” which is actually an interrogation that gets the suspect to admit guilt, and then a recorded “confession” that does not show how the confession was obtained.

The consent decree ranges across a host of other subjects.  But it is truly good to see the Department of Justice put its considerable weight into moving police departments toward the best that science has to offer.  This won’t cut through law enforcement’s resistance to science all by itself, but it will help.

 

 

 

Why “Failed Evidence?”

I titled this book Failed Evidence for a simple reason.  Evidence is the proof of facts lawyers use to prove matters in court.  Evidence can take many forms: the statements of the defendant to the police, eyewitness identification testimony, and the results of forensic testing., to name just three important categories.

Twenty-five years ago, before I became a law professor, I was trying cases as a criminal defense attorney and (more briefly) as a prosecutor.  Confessions, eyewitness identifications, and forensic tests were, for the most part, unassailable.  Then came the DNA revolution, beginning in 1989.  With DNA , we could identify the real perpetrator from trace amounts of bodily fluids, with a degree of certainty formerly unimaginable.  But we also learned that, in many cases, routine types of evidence had convicted the innocent.  DNA proved it, just as it had proved the guilt of others.

We now have decades of scientific work and evaluation on confessions, eyewitness identification, and forensics, and we have learned something disturbing: many of our traditional methods of conducting investigations create risks of miscarriages of justice.  Thus, for some of our most common ways of gathering evidence, we fail in an unacceptable number of cases.

But it is the book’s subtitle — Why Law Enforcement Resists Science — that brings the the real failure into focus.  For example, almost four decades of solid science tells us that showing an eyewitness a lineup of similar-looking people — think of any lineup in the movies or on TV that you’ve ever seen — produces a risk of mis-identification.  The science also tells us that we can reduce this risk considerably by showing the people in the lineup to the witness not simultaneously, but sequentially: one at at time.   But most of law enforcement — both police agencies and prosecutors’ offices — have refused to embrace the science and the improved practices this would create.  A few jurisdictions — New Jersey, for example — have changed their procedures based on the science, but they remain a distinct minority.  Most agencies continue to actively resisting any change.

Failed Evidence explains why law enforcement has resisted the scientific work that looks into its most basic tasks, even in the face of DNA-based proof that improvement is badly needed.  Failed Evidence then lays out concrete strategies for overcoming the resistance, with examples of success.

So, for me,  Failed Evidence seemed like the only logical choice for a title.