This installment of commentary on the case briefly reviews the earlier opinions. Later, I shall explain the state of the science that troubled the court and consider its advice to the scientific community.
I. The First Court of Appeal Opinion: An Old Debate
In 1990, Pizzaro was sentenced to spend the remainder of his life in prison for raping and suffocating his thirteen-year-old half-sister, Amber Dawn Barfield, in 1989. Vaginal swabs revealed semen that matched Pizarro’s VNTRs at three loci. An FBI analyst testified that the probability that a randomly selected “unrelated Hispanic individual [would have] a similar profile as Mr. Pizarro is one in approximately 250,000” and that the probability for Caucasians was one in 10,000,000.
Pizarro appealed, contending that the DNA evidence was inadmissible because the prosecution had failed to demonstrate that the DNA test procedure and the simple multiplications of allele frequencies used to arrive at the 1/250,000 and 1/10,000,000 figures were generally accepted in the scientific community. The court of appeal remanded for a hearing on these matters. People v. Pizarro, 12 Cal.Rptr.2d 436 (Ct. App. 1992). By the time of the rehearing in 1998, the California Supreme Court had resolved these questions in favor of admissibility. Deprived of his original arguments, Pizarro raised “several new issues concerning the reliability and relevance of the DNA evidence presented to the jury” eight years earlier. “The trial court again ruled that the evidence was admissible and reentered the judgment.” Pizarro, 3 Cal.Rptr.3d 21 (Ct. App. 2003).
II. The Second Court of Appeal Opinion: A Legal and Logical Misunderstanding
Pizarro appealed once more. In 2002, the appellate court reversed the conviction due to a variety of perceived defects in the analysis and presentation of the random-match probability. After publishing this opinion, the Court of Appeal decided to rehear the second appeal a second time “to ensure that the complex issues in this case were thoroughly examined and briefed by both parties.” In an opinion in 2003, the appellate court determined that it was improper for the prosecution to have offered an estimate of the frequency of a genotype in the Hispanic and Caucasian population groups. These were chosen because the defendant was said to be half Caucasian and half Hispanic.
The Court of Appeal had a point. The state’s trial presentation was problematic. The frequencies of DNA alleles vary across self-identified racial or ethnic groups. If the perpetrator could have come from any group, looking to only one or two groups for a random-match probability could lead the jury to jump to the conclusion that the perpetrator must have come from those particular groups. The manner of presenting the DNA statistics therefore could be regarded as unfairly prejudicial.
But the Court of Appeal did not stop here. It announced in dictum that the standard practice of giving a range of frequencies for all the major racial or ethnic groups in the United States also would have been unacceptable. According to the court:
[I]n the absence of sufficient evidence of the perpetrator's ethnicity, any particular ethnic frequency is irrelevant. The problem is . . . one of preliminary fact . . . . It does not matter how many Hispanics, Caucasians, Blacks, or Native Americans resemble the perpetrator if the perpetrator is actually Asian. If various ethnic frequencies are presented to the jury, each will have been admitted without adequate foundation.The Pizarro court insisted that the race of the possible perpetrator was one of those “preliminary facts” that had to rest on “independent proof.”
This might seem strange to scientists who understand the need to evaluate all the plausible alternatives before concluding that a given theory is true. One might think that this also is how jurors should reason before they can find a defendant guilty. But the incongruity between ordinary reasoning and the court's understanding of the rules of evidence did not trouble the court. If the court’s interpretation of the law in Pizarro seemed weird to the DNA experts, it must be because those scientists did not appreciate “the subtle, even unexpected, differences between the scientific and legal approaches to the same problem.”
The odd requirement that the racial identity of the perpetrator be known seems to arise from the court’s recognition that if this fact could be known, some of the statistics would be irrelevant. According to the court, “the jury is not assisted by knowing how many Hispanics possess the perpetrator's traits if the perpetrator is actually Asian.” (Emphasis added).
This last statement is true but unhelpful. To be sure, if the perpetrator is known to have come from a small village in Gansu populated by the Dongxiang, no major U.S. population group’s profile frequency is directly relevant. But what does this fact have to do with a case in which the race of the perpetrator is unknown? In this situation, it remains entirely sensible to consider profile frequencies under all the plausible hypotheses involving major U.S. populations. Contrary to the Pizarro court's assertions, in a general-population case—one in which the investigation cannot be limited to a particular racial group—the statistics for a range of groups are relevant. They assist the jury in understanding the implications of the DNA match. Thus, in an incisive opinion in People v. Wilson, 136 P.3d 864 (Cal. 2006), by Justice Ming Chin, the California Supreme Court recognized the logical difficulties with the dictum in Pizarro.
II. On Retrial: More Powerful DNA Evidence and Juror Misconduct
To prepare for the retrial, the state conducted new DNA tests. One STR locus after another in the DNA from a vaginal swab showed alleles that were not the victim’s. All of these alleles were part of Pizarro’s genotype. “The frequency of the profile was calculated to be approximately one in 3.9 quintillion African–Americans, one in 350 quadrillion Caucasians, and one in 4.2 quadrillion Hispanics.” People v. Pizarro, 216 Cal.App.4th 658 n.28 (Ct. App. 2013).
Pizarro filed a 105-page motion to exclude this damning evidence. He maintained that the analysis of samples containing alleles from more than one individual lacked general acceptance, that STR typing lacked general acceptance, and that the typing and interpretation was done incorrectly. The trial court saw no reason for a hearing on the general-acceptance claims, but it conducted an evidentiary hearing on the last claim. Discerning no major problems with the STR analysis, the court denied Pizarro’s motion. With the DNA (and other) evidence before it, the jury in Michael Pizarro’s third trial returned another verdict finding him guilty of the rape-murder of his half-sister on a June night 19 years earlier.
But this is not the end of the legal story. After the jury was discharged, the parties learned that during the trial a juror had read the second appellate opinion in the case, which revealed that defendant had previously been convicted on all counts by another jury, that he been sentenced to life in prison without the possibility of parole, that VNTR analysis also had incriminated him, and that other case-specific facts not presented on retrial indicated his guilt. The judge found that the juror engaged in “gross misconduct” that was “absolutely outrageous.” Nonetheless, he denied a motion for a new trial. Once more Pizarro appealed.
III. The Third Court of Appeal Opinion
The Court of Appeal reversed because of the juror’s misconduct. It practically ordered the state to prosecute the juror, writing that “a juror who so brazenly disregards and willfully violates the court's admonition not to investigate or consider matters outside the evidence received in the courtroom should be criminally prosecuted. Jurors take an oath to follow the court's instructions.” Id. at n.4.
In addition, the appellate court was unhappy with the introduction of the DNA evidence. Its opinion regarding this evidence is hard to follow. On the one hand, the court wrote that “we uphold the admission of the DNA evidence.” On the other hand, and in the same sentence, it complained that “the widely held belief that allelic dropout cannot cause false results in a criminal case as long as the defendant's and the perpetrator's DNA samples are tested in a consistent manner is based on the improper assumption that the defendant is the perpetrator—in other words, that the defendant is guilty.” In the next sentence, it “determine[d] in this case, however, that any error was harmless.” By the end of the prolix opinion, it “recommend[ed] that the scientific community reexamine the possible effects and management of allelic dropout in criminal cases” and avoid “a very serious falsehood based on the improper assumption that the defendant is guilty.”
Plainly, this recommendation and the underlying understanding of the science of DNA identification bear inspection. Stay tuned for that.