Jeffrey A. Singer
On November 20, the New Jersey Supreme Court ruled 6–1 that the hypothesis behind Shaken Baby Syndrome/Abusive Head Trauma (SBS/AHT) — that shaking a baby’s head can cause the clinical “triad” of intracranial hemorrhage, brain swelling, and bleeding behind the retina — is controversial within the scientific community and cannot serve as a basis for convicting someone of child abuse. Writing for the majority in the case of State vs. Darryl Nieves (charged with aggravated assault by shaking his 11-month-old son and causing neurological damage), Justice Fabiana Pierre-Louis stated in the opinion’s syllabus:
[T]he evidence presented at the Frye hearing—including the testimony by the State’s single expert witness—showed that there was no test supporting a finding that humans can produce the physical force necessary to cause the symptoms associated with SBS/AHT in a child. There is evidence of general acceptance by many in the medical community, but the State must also establish general acceptance in the biomechanical community, and it has failed to do so.
As I wrote about a different case here, Shaken Baby Syndrome gradually gained popularity after British pediatric neurosurgeon Norman Guthkelch hypothesized that violently shaking a baby could cause the triad of findings in a 1971 article in the British Medical Journal. The hypothesis evolved into a diagnosis, and the triad of findings evolved into a diagnostic triad over the ensuing decades.
However, beginning in the 1980s, several bioengineers conducted biomechanical studies that concluded it is physically impossible for a baby to be shaken to the degree that would cause such an injury without any other physical impact.
A 2016 systematic review conducted by Sweden’s Agency for Health Technology Assessment and Social Services found only limited, low-quality evidence linking the triad—or any of its individual findings—to traumatic shaking. The reviewers also reported that the available research was too weak to determine how accurately the triad can diagnose shaking injuries.
Even in the face of this growing uncertainty, the medical establishment continued to defend the diagnosis.
As the controversy grew in 2009, the American Academy of Pediatrics (AAP) officially recommended replacing the term shaken baby syndrome with abusive head trauma in both clinical and legal settings. The shift from “shaken baby syndrome” to “abusive head trauma” allows its supporters to have it both ways: they continue to claim that shaking alone can cause the triad, but by dropping the original label, they no longer need to defend that mechanism or the weak evidence supporting it. Instead, they categorize any unexplained triad-like findings under the much broader—and less testable—umbrella of “abuse.”
In 2018, the AAP and several other American and international medical organizations, including the Centers for Disease Control and Prevention (CDC), published a consensus statement endorsing the validity of the SBS/AHT diagnosis. The Consensus Statement asserted that SBS/AHT is a medical diagnosis, not a legal finding of murder. It declared that “[t]here is no controversy concerning the medical validity of the existence of AHT.”
The medical community has routinely discounted biomechanical findings—claiming they don’t qualify as medical research—to avoid confronting what those studies imply about the reliability of the diagnosis.
However, the New Jersey Supreme Court concluded otherwise. In the opinion’s syllabus, Justice Pierre-Louis wrote:
It is therefore evident that the foundation of SBS/AHT lies in biomechanical science and engineering. A scientific community is either relevant or not for purposes of determining admissibility of scientific evidence at trial—degrees of relevance are not weighed. As in Olenowski II, there can certainly be more than one relevant scientific community for purposes of Frye. Here, the relevant scientific communities for purposes of determining the reliability of SBS/AHT expert testimony are both the medical/pediatric community and the biomechanical engineering community.
Put simply, science is science. If the core theory behind SBS/AHT remains contested across relevant scientific fields, that uncertainty belongs in the courtroom—and it cannot be the foundation for sending someone to prison. And in a criminal case—where the state must prove guilt beyond a reasonable doubt—unresolved scientific controversy is the very definition of reasonable doubt.
New Jersey is now the first state to bar expert testimony claiming a defendant caused SBS/AHT in criminal cases.
Meanwhile, in Texas, Robert Roberson, whom I wrote about in a USA Today article, remains on death row after being convicted of killing his young child by shaking her too hard. His execution was initially scheduled for October 2024 but was postponed after the Texas legislature intervened. When rescheduled for October 2025, the Texas Court of Criminal Appeals put it on hold pending further hearings.
At an October 2024 Cato online policy forum, I discussed the validity of the “shaken baby hypothesis” with a panel of forensic medical and legal experts.
Will the New Jersey Supreme Court decision have an impact in Texas or other states? I asked my colleague, Mike Fox, a legal fellow at the Cato Institute Project on Criminal Justice. His answer:
Obviously, one state isn’t bound by the law of another. But junk science doesn’t know jurisdictional boundaries, so I’d imagine it certainly can—and hopefully will.