Written by Rosa Klein-Baer
Last month, Jennifer Del Prete was released on bond from prison after serving nearly half of a 20-year sentence for the murder of an infant in her care.
Del Prete – a daycare worker with no history of violence and who has maintained her innocence since being accused – was convicted of shaking 4-month old Isabella Zielinski when the child fell unconscious while in her care. Isabella never woke up from her coma, and died in the hospital 10 months later. It was, prosecutors insisted, a classic case of violent and intentional shaken baby syndrome (SBS), and they had medical experts to prove it.
However, Matthew Kennelly, the U.S district judge who ordered Del Prete’s release, is not at all confident of the SBS verdict, or even that the typical symptoms used to diagnose SBS are sufficient to prove anyone guilty of child abuse or murder. In a comprehensive 97-page opinion, Kennelly draws attention to the controversial science behind SBS, going so far as to suggest: “a claim of shaken baby syndrome is more an article of faith than a proposition of science.”
Kennelly’s ruling is groundbreaking, as it condemns for the first time the SBS diagnosis – which has held up in court and been used to convict hundreds of people on charges of child abuse and murder – as unreliable. Since the 1990s, over a thousand cases of SBS have been heard in court. Somewhere between half and three quarters of these cases have relied on an SBS-associated triad of symptoms as the sole medical evidence.
For decades, this triad – subdural hemorrhage (bleeding surrounding the brain), retinal hemmorhage and brain swelling – has been used by prosecutors as irrefutable evidence that determines virtually all important aspects of the case: how the child died (from violent shaking), who did it (the last person with the child), and what the murderer’s state of mind was at the scene of the crime (enraged and intentional). It has been assumed that – with the exception of high-impact accidents such as a car crash or a fall from a building – the only thing that could lead to such internal bleeding and brain
swelling could be extremely violent and purposeful shaking on the part of the last person with the child.
Other physical symptoms, such as bruises or cuts that one might expect to occur in cases of extreme child abuse, did not need to be present for a guilty verdict. Nor was it necessary for there to have been any witnesses. An overwhelming amount of faith in the scientific principles that established the triad meant that it alone had the power to determine one’s guilt.
This is troubling for several reasons, not the least of which is that the science behind the triad is coming into question. Many doctors now acknowledge that the same symptoms that once were used to definitively conclude that SBS had taken place could actually be caused by other conditions, such as infections or bleeding disorders. There have also been findings that dispute that the assumption the shaking need necessarily be violent and intentional for these symptoms to take place.
More importantly, however, Del Prete’s case reminds us of the vulnerabilities of the criminal justice system and makes us question our understanding of scientific theories as “facts.” Central to science is the notion that ideas are always under dispute and subject to change. Yet this core tenet was not reflected in Del Prete’s initial trial, or hundreds of others like it, which deemed the SBS theory was enough to send someone to jail. By trusting the triad as irrefutable evident of guilt, we put the lives of hundreds in the hands of a single idea – a single idea that is now coming to be seen as false.
Del Prete’s case can teach us about much more than the controversy surrounding SBS. It shows us how crucial it is that evidence not be based on one idea alone. Theory is important, but over-reliance on it can be a bit like trusting a single witness when the rest of them have nothing incriminating to say. We need to be comprehensive in our trials, otherwise “innocent until proven guilty” loses its meaning.