People v. Kowalski

• Decision Date: 30 July 2012
• Docket Number: Calendar No. 7.,Docket No. 141932.
• Citation: 492 Mich. 106,821 N.W.2d 14
• Parties: PEOPLE v. KOWALSKI.
• Court: Michigan Supreme Court

492 Mich. 106

821 N.W.2d 14

PEOPLE

v.KOWALSKI.

Docket No. 141932.

Calendar No. 7.

Supreme Court of Michigan.

Argued Nov. 9, 2011.

Decided July 30, 2012.

Bill Schuette, Attorney General, John J. Bursch, Solicitor General, David L. Morse, Prosecuting Attorney, and William J. Vailliencourt, Jr., Assistant Prosecuting Attorney, for the people.

Hertz Schram PC, Bloomfield Hills, (by Walter J. Piszczatowski and Michael J. Rex) for defendant.

Cramer & Minock, PLC, Ann Arbor (by John R. Minock), and Marla R. McCowan, Detroit, for the Criminal Defense Attorneys of Michigan.

Dykema Gossett PLLC, Ann Arbor (by Jill M. Wheaton), for the Innocence Network.

David W. Ogden, for the American Psychological Association.

Bill Schuette, Attorney General, John J. Bursch, Solicitor General, and Mark G. Sands, Assistant Attorney General, for the Attorney General.

MARY BETH KELLY, J.

This case requires that we determine whether expert witness testimony regarding interrogation techniques and psychological factors claimed to generate false confessions is admissible under MRE 702 and MRE 403 and whether exclusion of this testimony violates the Sixth Amendment right to present a defense. The circuit court excluded the testimony of two experts regarding the occurrence of false confessions and the police interrogation techniques likely to generate them as well as the psychological characteristics of defendant that allegedly made him more susceptible to these techniques.

We hold that the circuit court did not abuse its discretion by excluding the expert testimony regarding the published literature on false confessions and police interrogations on the basis of its determination that the testimony was not reliable, even though the subject of the proposed testimony is beyond the common knowledge of the average juror. We also hold, however, that the circuit court abused its discretion by excluding the proffered testimony regarding defendant's psychological characteristics because it failed to consider this evidence separately from the properly excluded general expert testimony and therefore failed to properly apply both MRE 702 and MRE 403 to that evidence. Accordingly, we remand this case to the circuit court for it to determine whether evidence of defendant's psychological characteristics is sufficiently reliable for admissibility under MRE 702. We further hold that the circuit court's application of MRE 702 did not violate defendant's constitutional right to present a defense.

I. FACTS AND PROCEDURAL HISTORY

In May 2008, the brother and sister-in-law of defendant, Jerome Walter Kowalski, were found dead in their home. Defendant was charged with both murders. Testimony elicited at defendant's preliminary examination and Walker¹ hearing indicates that police questioned defendant about the killings four times over the course of several days: first at defendant's home, next at the Brighton Police Station, then at the Ann Arbor Police Department, and finally at a Michigan State Police post.

During the third interview session, defendant acquiesced to the interviewer's statement that there was a “fifty percent chance [he killed his brother], but a fifty percent chance [he] didn't.” Defendant discussed having a “blackout” and “blurred” memory and stated, “I thought I had a dream Thursday, but it was the actual shooting.”

Defendant confessed to the murders during the last interview session, which followed a night in jail. Defendant stated that he went to his brother's home, walked into the kitchen, and murdered his brother and sister-in-law after a brief verbal exchange. The record suggests that defendant initially described shooting his brother in the chest from a distance of several feet, although he eventually changed his account after a detective illustrated through role-playing that defendant's first version of events did not corroborate the evidence recovered from the victims' house. At this point in the pretrial proceedings, defendant's confession is the primary evidence implicating him in the murders.

Before trial, defendant filed a motion to suppress his statements to the police, which the circuit court denied after conducting a Walker hearing. Defendant then filed a notice of intent to call two expert witnesses. Dr. Richard Leo, a social psychologist, would testify regardingpolice interrogation techniques and the existence of false confessions. Dr. Jeffrey Wendt, a clinical and forensic psychologist, proposed to testify about the psychological testing he performed on defendant and offer his opinion about defendant's mental state during police questioning. Wendt would also offer testimony that the “circumstances of Mr. Kowalski's confession were consistent with the literature on false confessions” and that the interaction between defendant and police “was consistent with a coerced internalized confession.”

The prosecutor moved to exclude this proposed expert testimony, arguing that it was inadmissible under MRE 702. Both experts testified at a Daubert² hearing. Leo explained that his research classified each confession he believed to be false as either a “proven false” confession, a “highly probable false” confession, or a “probable false” confession.³ This categorization involved comparing the narrative of a defendant's confession with other evidence, checking whether the confession led to independent evidence, and looking for other indicia of reliability, with a researcher determining whether the confession fell into one of the three categories of false confessions.⁴ While some of the facts involved in this analysis came “directly from case files,” many were gleaned from secondary sources, including popular media accounts.⁵ In addition to classifying confessions by his confidence in their falsity, Leo also classified confessions as “voluntary false confessions,” “stress-compliant false confessions,” “coerced-compliant false confessions,” “coerced-persuaded false confessions,” or “non-coerced-persuaded false confessions.” ⁶ Leo categorized each confessionin this manner by comparing the circumstances of each confession with those of other confessions he had already determined to be false.

On the basis of this research, Leo proposed to testify that “false confessions are associated with certain police interrogation techniques,” that “some of those interrogation techniques were used in this case,” and that “risk factors associated with false and unreliable confessions, especially persuaded false confessions, were [also] present in this case.” In support of Leo's opinions, defendant offered research conducted by Leo and by others. Some of this research appeared in peer-reviewed scientific journals, while some appeared in law reviews, which are not peer-reviewed.

Next, Wendt testified that he had administered a battery of standard psychological tests on defendant, performed an extensive clinical interview of defendant, and reviewed both the police reports recounting the circumstances of defendant's police interrogation as well as the transcripts of those interrogations.⁷ Wendt testified that these types of data are routinely used at the Center for Forensic Psychiatry.⁸ Wendt then combined all these “data sources” to form a psychological profile, which allowed him to discuss how defendant's traits affected his ability to interact with other people.⁹ Lastly, Wendt proposed to testify that “[t]he circumstancesof [defendant's] confession were consistent with the literature on false confessions” and that the interaction between defendant and the police “was consistent with a coerced internalized confession.” ¹⁰

At the conclusion of the hearing, the circuit court excluded both experts' proposed testimony. The circuit court ruled that Leo was qualified in terms of knowledge but that his testimony was unreliable and would not assist the trier of fact. The circuit court stated that “the lack of precise information” precluded Leo from measuring the accuracy of his studies and also critiqued the sources underlying Leo's classifications of particular confessions as false:

[Leo] doesn't ... have the ability in his studies to review video tapes, which would be reliable. He relies on newspaper accounts, magazine articles. He relies on information provided from sources that are prejudice[d], for example a defense attorney that represented a defendant, [and] advocates against the death penalty.... I don't understand why he couldn't have done a [Freedom of Information Act request], police agency files where confessions were given and the suspect was tried or not tried, pull court files, order transcripts, review police records.... [This] would be a ... more reliable methodology.

The circuit court examined the manner in which Leo analyzed the confessions that he determined to be false:

[Leo] starts with the conclusion that the confession is false and then he works backwards.... He doesn't take into consideration why someone might falsely confess, other than because of a police interrogation technique.... [A]nd there are reasons why people would falsely confess, they might be trying to protect someone.... He hasn't determined a reliable means to have a study group consist of innocent people who wrongfully confess that weren't mentally ill or youth.

The circuit court criticized this methodology for failing to compare true and false confessions and identify factors that contribute to false confessions but not true confessions. As the circuit court stated, “[I]f true and false confessions can be derived from the same police interrogation techniques, [how] is it possible to blame police interrogation techniques with any degree of reliability?” Given what the circuit court considered to be inadequacies of Leo's data and methodology, the circuit court concluded that Leo's testimony was unreliable.

The circuit court further determined that Leo's testimony would not assist the trier of fact because the jury could evaluate the credibility and reliability of de...