Science in the courtroom must be reliable. To ensure that, we want and expect judges to serve as gatekeepers, examining both reliability and relevance; but we also want and need to protect the right to present a defense. Two cases illustrate the apparent collision of these demands and the result—two new trials.
The first decision arose from a capital murder case—a woman sentenced to death for the murder of her child, a crime that at trial she denied committing. (Lucio v. Davis, No. 16-70027, 2019 U.S. App. LEXIS 22447, at *1–2 (5th Cir. July 29, 2019).) Here, in the succinct summary provided by the Fifth Circuit, is what transpired:
Her daughter’s body had been badly bruised, but the State’s examiner concluded that she died from a final blow to the head. . . . After about five hours of interrogation, Lucio admitted to, in her words, spanking her daughter. The State used those statements to argue to the jury that she confessed to abusing Mariah and that, by inference, she must have killed her. Lucio tried to rebut the impact of the interrogation by putting on an expert witness [Pinkerman] to explain why she would admit to facts that were not true. But the state trial court would not allow the expert to testify because it concluded such testimony was “irrelevant” . . . .
Against Lucio’s claim of having been denied due process of law by being prevented from presenting a “complete defense” lay the trial judge’s conclusion that this proof was irrelevant. How?
The state trial court concluded that Lucio “admitted actions that she took that could have resulted in the death. But she denied ever having anything to do with the killing of the child.” It thus had “a hard time figuring out how it goes to the guilt or innocence” and denied Lucio the opportunity to present Pinkerman’s testimony. (Id. at *22.) This was quickly and forcefully repudiated. There was no physical evidence or eyewitness testimony of abuse by Lucio; the statement was core to the prosecution; and “Pinkerman’s opinion was that Lucio was susceptible to taking blame for something that was not her fault and that this behavior was manifested in the interrogation video. It thus cast doubt on the State’s key evidence and was paramount to Lucio’s defense.”
(Id. at *25.)
Here, neither party questioned the expert’s qualifications or the sufficiency of the information supporting the opinion. What happened here was a fundamental miscarriage—a failure to understand principles of relevance and how a defense story must be heard.
The second case was also one involving the murder of a child. (Allison v. State, No. 2651, 2019 Alas. App. LEXIS 94 (July 26, 2019).) The defense—with significant medical and scientific support—was that the child’s death was consistent with having resulted from a fall. At issue was Ehlers-Danlos Syndrome, “a group of inherited disorders that affect a person’s connective tissues—primarily their skin, joints, and blood vessel walls. People who have Ehlers-Danlos Syndrome usually have overly flexible joints and stretchy, fragile skin. They can also be more susceptible to bruising and excessive bleeding.” (Id. at *6–7.)
After her child’s death, the mother was diagnosed as having a form of this disease. In her defense, she sought to prove that there is a 50 percent chance of her child having inherited this condition, and that before any determination of what caused death in a case involving significant bleeding, a family history of this syndrome must be considered. Again, the proof was disallowed. As the appellate court related: “The trial court ruled that such evidence was inadmissible unless (1) the defense produced an expert in the syndrome who could explain the mother’s diagnosis; and (2) the defense produced an expert who could diagnose J.A. with the syndrome ‘to a reasonable degree of medical certainty.’” (Id. at *15–16.)
Both of these rationales were wrong. First, experts need not be “expert in the syndrome” to be able to apply their knowledge of that condition to their conclusions.
Doctors Ophoven and Tawansy were qualified by the court to offer expert opinions on the State’s differential diagnosis of shaken baby syndrome/abusive head trauma. . . . Contrary to the trial court’s belief, the doctors did not need to be experts on Ehlers-Danlos Syndrome to reasonably rely on this medical information. Nor did they need to be experts on the syndrome to offer their expert opinion that this maternal history was something that “merited further investigation” and should be considered in any differential diagnosis of J.A.’s death.
(Id. at *17.) Put more simply, an expert in differential diagnosis may opine that “before you reach a conclusion of how the child died, the possibility of Ehlers-Danlos syndrome must be weighed in light of the family history.”
As to the second error, the trial court conflated certainty with a relevance standard.
Allison was not required to present an expert who could diagnose J.A. with Ehlers-Danlos Syndrome “to a reasonable degree of medical certainty.” We recognize that the trial court had concerns about what it considered to be “the speculative nature” of the diagnosis. But the State bore the burden of proving Allison’s guilt beyond a reasonable doubt; it was not Allison’s burden to prove his innocence. . . . The evidence offered by Allison satisfied this test; that is, it tended to create a reasonable doubt as to Allison’s guilt. The State’s proof in this case rested on its experts’ conclusions that there was no reasonable explanation for J.A.’s death other than physical abuse. Evidence that there were other possible medical explanations for her excessive bleeding was something that the jury should have heard.
(Id. at *18–19 (footnote omitted).)
The appellate court was making two points. There was enough proof that the child might have suffered from Ehlers-Danlos Syndrome to challenge the state’s expert for failing to weigh that when conducting a differential diagnosis. Such proof simply needed to meet the minimal relevance standard of tending to make a fact of consequence more or less probable.
Are there essential takeaways from these two cases? The due process guarantee of the accused’s right to present a defense must be factored in before excluding forensic evidence. Where qualified experts have information that will be helpful to the jury in answering the question(s) of culpability, courts must tread with care before keeping witnesses or knowledge out of the courtroom.
There is one additional lesson, separate from the due process concern. These two cases are ones where counsel dug deep, asking what else could explain conditions that made their clients appear guilty and then finding qualified experts who could answer that question. That is a model for effective lawyering.