NJ Bans ‘Shaken Baby Syndrome’ Evidence in Nationwide First (1)

Criminal prosecutors can’t build murder cases on medical diagnoses that the mere shaking of a baby, without further evidence of trauma, resulted in a child’s death, the New Jersey Supreme Court ruled Thursday.

The landmark ruling made New Jersey the first state in the country to block a “shaken baby syndrome” theory when investigations into child deaths don’t show other injury, such as bruising or neck damage. It’s the first high court to openly question this diagnosis, which has come under intense scrutiny in recent years as scientists and lawyers have pushed back against what they say is medical guesswork and has resulted in countless convictions.

“Regardless of the severity or viciousness of a crime, however, the bedrock foundations and protections embedded within our system of criminal justice must prevail,” Justice Fabiana Pierre-Louis wrote for the court’s majority. “Under one of those core principles, only evidence that is sufficiently reliable and has probative value not outweighed by the evidence’s prejudicial effect may be presented to the jurors to inform their consideration of the charges against the accused.”

Justice Rachel Wainer Apter’s solo dissent said the court was discarding the majority opinion of the country’s medical establishment.

“This Court has previously warned that a ‘court should not substitute its judgment for that of the relevant scientific community,’” she said. “Judicial humility requires us to accept that we are judges, not scientists. Nor are we experts in the scientific method.”

The case caught the eye of national groups, like The Innocence Network and Center for Integrity in Forensic Sciences, which urged the court to take a hard look in light of evidence questioning this theory.

“It’s a great ruling on the specific issue with this specific diagnosis, but it’s also a great ruling for us in enforcing the gatekeeper rule of the courts—that only good science is being allowed to be presented to juries,” said Kelly Strange Crawford, a co-chair of Riker Danzig and lawyer for those groups.

‘Sufficiently Reliable’

The ruling attacked the science head on, saying there is a higher “sufficiently reliable” standard for evidence that can be presented to a jury than for doctors deciding whether to create a diagnosis and treat a patient or report potential child abuse.

The majority’s 109-page ruling scrutinized the theory and its history, going back to the origins in 1968 with neurosurgery experiments in car accident whiplash injuries involving monkeys. Though this research didn’t involve infants, it became a foundation for a diagnosis bolstered by neurosurgeons in the decades that followed, leading the American Academy of Pediatrics to release a statement in 1993 that there was a medical “presumption of child abuse when a child younger than 1 year of age has intracranial injury.”

Moreover the court found significant disagreement among scientific communities, and especially that many in the biomechanical community doubt shaking alone can produce the symptoms doctors say come with shaken baby syndrome.

The consensus on that evidence is too weak to potentially send someone to prison based on a doctor’s testimony that the only explanation for a child’s injuries in these cases is abuse, the court ruled—a step praised by the New Jersey Office of the Public Defender.

“The Court’s decision underscores a foundational principle: a person’s liberty cannot depend on expert evidence that lacks scientific grounding, cross-disciplinary support, or methodological reliability,” Cody Mason, the office’s managing appellate attorney, said in a statement.

Child Abuse Charges

Prosecutors pointed to the long list of medical associations that backed up the theory, arguing the medical consensus lines up behind the reliability of “shaken baby syndrome.”

These groups pushed back against a host of bio-mechanical experts who questioned the diagnosis and have raised alarm bells in recent years.

Wainer Apter said the court was going beyond its proper gate-keeping role, eliminating the state’s ability to charge people with child abuse for “shaking or slamming an infant unless external evidence of impact is present.”

“No case has ever concluded that evidence of SBS/AHT is unreliable. And no case has ever found its reliability sufficiently questioned to preclude its admission at a civil or criminal trial,” she said.

The case is New Jersey v. Nieves, N.J., No. A-26-23, 11/20/25.