Susan C. Anthony

Shaken Baby Syndrome Resources

Here are links to Shaken Baby Syndrome (SBS) / Abusive Head Trauma (AHT) resources on this web site.

A Tragedy and a Mystery

Article sections:  Timing  |   Differential Diagnosis  |   Character and History of Defendants  |   Criminal Justice System   |  Conclusion  |   Let's Network   |   Bibliography

I began researching Shaken Baby Syndrome more than a decade ago because of a tragedy and a mystery.

The tragedy: One of the gentlest, most patient and most mature women I’ve ever known was babysitting a sick baby. The child had a seizure and died in the hospital two days later. The babysitter was charged with manslaughter. It took three years for the case to come to trial, in part because of perjury on the part of prosecution witnesses, considered by prosecutors an "obstacle" to be "overcome." Despite a weak case and little investigation or testing, the sitter was convicted of manslaughter and sent to prison.

The mystery: Of all the people I’ve ever known, this woman is the least likely to abuse a child at all, let alone so severely that death results. She has an unusual tolerance for noise (screaming and crying), an unusual ability to deal with children’s behavior firmly but lovingly, and an unusual tenderness and empathy for children. But if she didn't hurt the baby, why did he die?

Shaken Baby Syndrome (SBS), officially called Abusive Head Trauma (AHT), is based on the premise that a particular constellation of injuries—subdural hemorrhage (bleeding in the brain), retinal hemorrhage (bleeding in the eyes), and cerebral edema (swelling of the brain)—are reliable indicators that an infant has been violently shaken. For decades, doctors have been taught to diagnose child abuse when they find these symptoms unless caretakers offer an acceptable alternate explanation for their existence.

The syndrome first appeared in the medical literature in the 1970s.1,2 During that decade, people seem to have been reluctant to believe that a parent or caretaker would harm an infant. In the 1980s a good deal was published on Shaken Baby Syndrome, and it became clear to a majority of doctors that shaking could indeed cause infant death. My impression from reading articles published during this period was that researchers were searching for truth and advancing hypotheses rather than jumping to conclusions. They were concerned and open-minded. In the 1990s, however, several unproven hypotheses hardened into dogma.


The timing of the injury is one example. It is dogmatically assumed that the last person alone with the child is the perpetrator. Whatever that person says to the contrary is taken to be a lie, and whatever a previous caretaker says (i.e. "the child was fine when I left") is taken to be truth. In our case, even repeated perjury by the previous caretaker did not shake the confidence of prosecutors that everything else she said was true! This despite a dearth of investigation or hard evidence.

Few articles specifically address timing. In 1995, Nashelsky3 found only three cases in the literature where the interval between lethal infant shaking and the onset of severe symptoms (seizure, apnea, coma) was documented. In two of the cases, symptoms arose immediately after shaking. In the other case, there was a delay of four days between shaking and the onset of seizures. In 1998, Gilliland4 reported an interval of less than 24 hours in 80% of 76 cases reviewed and an interval of more than 72 hours in four children. It is generally agreed that a child is not asymptomatic in the interim, but symptoms may be non-specific: lethargy, irritability, loss of appetite, vomiting. A caretaker unaware of what occurred prior to their watch may attribute such symptoms to illness. Without a good history, even doctors don't always diagnose life-threatening brain damage based on nonspecific symptoms!

In 2002 a letter to the editor was published in the American Journal of Forensic and Medical Pathology5, Dr. Robert Huntington III described a case that clearly had a delay between injury and alarming symptoms. A 13-month-old girl was brought to the hospital on the morning of September 18, 1999. She had been vomiting for 24 hours and was irritable and lethargic. Because she had bruises, abuse was suspected. She was admitted to the hospital and police were contacted. Her chart indicates she was fussy and clingy, but interactive and responsive. It was not until 2:00 the following morning, more than 14 hours later, that a nurse on regular rounds noticed decreased respiration and realized something was seriously amiss. The baby was taken to intensive care where she died the following evening. The parents fled. (They were found and extradited from Mexico more than nine years later. The father pleaded guilty to a minor charge, went to prison for several months, and was subsequently deported.)

In this case, had the nurse who first noticed the alarming symptoms been a babysitter, she would likely have been charged with manslaughter or murder based on an unproven assumption—that severe symptoms are immediately apparent when a life-threatening injury has been sustained, therefore the last person caring for the child is the abuser. People around the world have been convicted on expert opinion to that effect. It would take concentration camp "experiments" to procure accurate statistical information on the length of time between shaking and onset of severe symptoms in fatal cases. For obvious reasons, research to conclusively prove or disprove any hypotheses on timing cannot be conducted.

Despite a dearth of evidence, however, the National Center on Shaken Baby Syndrome (NCSBS) insists that "true experts" have reached a consensus that severe injuries are not inflicted hours earlier. "Instead, brain swelling which progresses rapidly, causing breathing difficulties, loss of consciousness and seizures is a clear indicator that the cause of the brain swelling immediately preceded the onset of symptoms by only minutes to possibly an hour. Once this basic tenet of medical science is understood, the answer to the ‘whodunit’ is a matter of obtaining a minute-by-minute account of when the baby was fine or okay and when the baby became symptomatic or sick. Those who inflict severe injuries on children most often give the answers to those questions inadvertently, since they don’t know we can pin down the time of injury as closely as we can."6

Prosecutors and police officers like such simplicity. According to dogma, the last person with the child is the perpetrator, period. No need to waste time investigating further. It’s easy to build a case. There are no witnesses and no alibi. Organizations such as the National Center on Shaken Baby Syndrome are ready and eager to help prosecutors by providing "true experts" who can convince juries to accept the dogma, as well as assistance in debunking any claims made by "irresponsible experts." (Any expert who questions the dogma is branded an "irresponsible expert".)

I know of a case in which a babysitter (the last person with the child) passed a lie detector test. The previous caretaker, the mother, failed. Prosecutors still managed to get the babysitter convicted and sentenced to 53 years, after the first trial resulted in a hung jury with a majority voting for acquittal. In at least one other case, the previous caretaker had a history of child abuse; the person accused did not. Dogma is defended, regardless of individual circumstances of a case. Without it, some fear, true abusers may escape justice. Unfortunately, that is possible. On the other hand, clinging to dogma may consign non-abusers to prison. When that happens, the well-being of living innocent children are sacrificed to avenge the death of an innocent who will not return to us, no matter what. Some errors are inevitable. We should do what is possible to minimize the risk of error, but in America at least we are supposed to err on the side of acquitting a possibly guilty person rather than risk convicting a possibly innocent one.

Differential Diagnoses

The presumed "fact" that severe symptoms occur almost immediately is combined with a dogmatic assumption that the symptoms always indicate Shaken Baby Syndrome. Doctors may quickly conclude abuse and neglect to test for differential diagnoses. In some cases, doctors have leapt to the conclusion that the last person with the child shook him even when the child exhibited just one, two, or even none of the symptoms.7

I believe strongly that certain medical tests should be required before charges can be brought against a suspected perpetrator. Rare metabolic disorders such as Glutaric Aciduria type 1 have symptoms very similar to Shaken Baby Syndrome, and have in fact been misdiagnosed as such a number of times.8 Hemophilia or other clotting disorders may not manifest during the first year of life and may be contributing factors.9 In a number of cases I’ve read, the baby had recently received vaccinations. It could be that some individual children have allergic or heightened reactions to vaccinations.10 A baby may be deficient in Vitamin K, necessary for proper blood coagulation, especially if he/she is breast fed. If analgesic drugs have recently been taken, they could exacerbate existing problems.

One mother lost custody of three children after authorities assumed she had shaken her baby, who died some years later. Criminal charges were not filed because one person in authority didn't "cooperate" in assuming her guilt. She later had another baby who manifested the same symptoms. This time, she was lucky. A doctor tested for alternatives and discovered that the child had a rare genetic condition, Menkes' Disease. This led to the testing of tissue from the child who had died. He had the same condition. Unfortunately, she lost the battle to recover the children taken from her. They had been adopted and the statue of limitations for reversing an adoption had expired. Some authorities still insist she is guilty. Just because the baby had a disease with the same symptoms, they say, doesn't prove she didn't shake him!11

Why don't doctors test for alternative causes? I am told it's because alternative causes are rare. That means absolutely nothing in an individual case! I know a young woman with an extremely rare genetic disease, afflicting only a few other people on the planet. Just because it's rare doesn't mean she doesn't have it! However rare a condition may be in the general population, it is not as rare in the small subset of individuals who have the symptoms of the condition.

I suspect that it is easier to build a case against an accused person without testing for alternatives, although it is difficult to understand how truth would hurt the prosecution in cases of actual abuse. Persons intent on "defending children" by prosecuting and convicting their caretakers may consider their noble motives more important than truth. The lack of medical tests for alternatives should provide reasonable doubt, but jurors too often vote with their emotions. Assuming no short fall could result in such horrible injuries, they may be convinced by a skillful prosecutor to declare an innocent person guilty in order to avenge the death of a defenseless infant.

But incarcerating an innocent parent is another form of child abuse, sentencing that person’s innocent children to a childhood without a parent, perhaps in unnecessary poverty. Truth and justice should be the objectives, never revenge.

Character and History of Defendants

In some cases, it is quite clear that a child was abused. The perpetrator may have had a history of violence or hostility, a neighbor may have overheard the incident, or there may be other apparent injuries. In other cases, it is not so clear. A Texas prosecutor bemoaned the fact that it is difficult to get long prison terms for defendants. Defendants too often do not fit a criminal stereotype. "Some of it is based on the fact that you have these defendants who many times have had no prior criminal history and no past run-ins with the criminal justice system. They also usually have good character witnesses."11

If defendants in these cases "usually" have good character witnesses and no history that would indicate a tendency toward violence, perhaps something is going on that doctors don’t yet understand! Perhaps not every defendant is factually guilty.

Many articles deal with possible motives for shaking. Dr. John Caffey believed the most common motive is to correct minor misbehavior. "Crying" is often cited as the motivation, despite the fact that all babies cry. No distinction is made between defendants who historically have shown great tolerance for crying and those who are immature and expect children to meet their needs rather than vice versa. Much about defendants’ character and history is assumed without validation, i.e. that they are under stress, that they were abused as children, that they are emotionally unstable, that they have inadequate emotional support.

Criminal Justice System is not Perfect

If an initial investigation by police is shallow, superficial or inadequate, valuable clues as to what really happened are lost. Police are under a lot of pressure and are often overworked. Nevertheless, the consequences of shoddy work may be that a true perpetrator remains free to offend again while an innocent person is convicted, with all that implies.

Although our criminal justice system is better than many, it is far from perfect. The book Actual Innocence by Peter Neufeld and others12 looks into the phenomenon of innocent people convicted of heinous crimes and later cleared by DNA or other new technologies that prove beyond doubt that the convicted criminal is innocent. Hundreds of people have been released from long prison sentences and even death row based on absolute proof of their innocence made possible by technological advances. Many, many more cases are in process of being reviewed. The National Commission on the Future of DNA Evidence states, "The strong presumption that verdicts are correct has been weakened." Judges, lawyers and jurors should be aware of this and maintain humility as they search for truth. Innocence must be presumed until guilt is proven.

This is not to say that most people convicted of crimes are innocent! But would we be content to know that "most" of a particular airline’s planes arrived safely at their destinations? It’s important to be diligent in investigations and not to base criminal convictions on unfounded assumptions. Proverbs 17:15 states: Acquitting the guilty and condemning the innocent—the Lord detests them both. The objective should be truth, not winning.

There is no DNA evidence in shaken baby cases. There are generally no witnesses. Character and past history are often considered irrelevant. Convictions may be based completely on "expert" opinions of doctors who did not even test for differential diagnoses, or were hired specifically to help convict a defendant. How can an innocent person defend him/herself?


I am clearly convinced of the innocence of the woman whose indictment led to my research on Shaken Baby Syndrome. This opinion is not held lightly. I know her well and was impressed with her parenting skills from the day her first baby was born. I’ve spent hundreds of hours watching her interact with kids in every imaginable situation, and have been ever more impressed with her maturity, competence and empathy.

The case against her was based entirely on medical evidence. No official, other than the police officers who initially interrogated her and were already convinced of her guilt, ever bothered to speak to her. No doubts were ever entertained by her accusers. In their view, based on theory, assumption and dogma, she is guilty. Everything she may say to the contrary is a lie in their estimation.

As the days and months and years passed while we waited for her trial, my heart broke to think of the loss her little children would sustain if she were convicted. From the day they were born, I’ve thought how lucky they are to have such an unusually loving and functional family. There is no question but that she was under more stress than ever before in her life after the indictment, yet she chose to live each day as it came. She appreciated the little blessings each day offered. She steadfastly maintained her innocence despite threats and offers of plea bargains.

Why do I believe many allegations of baby shaking are false? For more than a decade, I've been reading appeals, news articles, medical papers, anything I could find about Shaken Baby Syndrome. Patterns have emerged that raise these questions:

  • Why do so many caretakers, all over the world, having no contact with each other, tell such similar "lame" stories about what preceded the child's collapse?
  • Why do so many suspects pass polygraphs?
  • Why do so many defendants have in common an excellent history and character?
  • Why do so many refuse to plea bargain?
  • Why are so many accused of not showing remorse?
  • Why do so many refuse to allow the jury an option to convict them on lesser included charges?
  • Why do so many appeal after their conviction?
  • Why do so many continue to deny wrongdoing even after conviction, when they have nothing to lose and could actually benefit by being granted earlier parole?

It is my strong opinion that there is something important doctors don't know yet about infants and infant brains. Unfortunately, no one is investigating. To question Shaken Baby Syndrome theory is heresy and could end a doctor's career.

In 1830, Thomas Sharkie wrote, "The maxim of the law is…that it is better that ninety-nine…offenders shall escape than that one innocent man shall be condemned."

In cases of suspected Shaken Baby Syndrome, suspects are presumed guilty and forced to prove their innocence beyond a reasonable doubt. They are prohibited from talking to others who cared for the baby earlier. They are not allowed to conduct an investigation as to what might actually have led to the baby's collapse.

It is my hope and prayer that honest researchers will question the dogma and continue to search for the truth, the whole truth and nothing but the truth rather than relying on unproven assumptions when people’s lives and futures are at stake. In the meantime, I hope and pray that doctors, police officers, prosecutors and jurors will maintain an attitude of humility. No one should ever be convicted on medical evidence alone in the absence of a thorough investigation documenting who was with the child and everything that transpired for at least 72 hours prior to the onset of severe symptoms. Thorough testing should always be conducted to decisively rule out differential diagnoses.

Nothing can change the tragedy of a baby’s death. But two wrongs do not make a right. Wrongfully depriving innocent children of a parent and a solid, secure childhood home only compounds the tragedy.

Let's Network

If you're reading this because you or someone you know has been wrongly accused, please contact me:

I have compiled hundreds of stories from news sources and the Internet in an effort to discern patterns that might offer clues to what is really happening with some of these little ones. I fully expect discoveries in the future to cast more and more doubt on the validity of convictions based primarily on medical opinion.


  1. Guthkelch AN: Infantile subdural hematoma and its relationship to whiplash injuries. BMJ 11:430-431, May 1971.
  2. Caffey J: On the theory and practice of shaking infants. American Journal of Diseases in Children 124(2):161-169, August 1972
  3. Nashelsky MB, Dix JD: The time interval between lethal infant shaking and onset of symptoms. American Journal of Forensic and Medical Pathology 16(2):154-157, June 1995.
  4. Gilliland MGF: Interval duration between injury and severe symptoms in nonaccidental head trauma in infants and young children. American Journal of Forensic Sciences 43(3):723-725, 1998.
  5. Huntington, Robert W. III: Symptoms Following Head Injury. American Journal of Forensic and Medical Pathology 43(1):105, March 2002.
  6. Parrish R: Proof is in the details, investigation and prosecution of shaken baby cases.
  7. Hanson M: Why Are Iowa's Babies Dying? ABA Journal: 75-78. August 1998.
  8. Hartley LM, Khwaja OS, Verity CM: Glutaric aciduria type 1 and nonaccidental head injury, Pediatrics 107(1):174-175, January 2001.
  9. Baehner R, Strauss H: Hemophilia in the first year of life. New England Journal of Medicine 275(10):524-528, Sept 1966.
  10. Scheibner V.: Shaken baby syndrome: the vaccination link. Nexus Magazine 5(5):
  11. Sutherly, Ben: Dayton Daily News, January 12, 2005.
  12. Neufeld P, Scheck B, Dwyer J: Actual Innocence. New York: Doubleday, 2000.

Go on to read Wrongly Accused of Shaking a Baby?
Source:, ©Susan C. Anthony