Judging Parents as Murderers on 4 Specks of Blood

Months had passed, and still Rick Crowl couldn’t purge images of the Lehmer baby from his mind.

There’d been no obvious wounds on the 3-month-old, nothing you could see. No signs of massive trauma; no signs of any trauma. No skull fracture, no collarbone bruises, no head injuries, no bleeding in the eyes, no gross bleeding under the scalp. Yet Thomas Bennett, the state medical examiner, had diagnosed shaken-slammed baby syndrome. Thomas Bennett had called Jonathan’s death a homicide.

So that’s what Crowl had called it too. The prosecutor for Pottawattamie County had gone along with Bennett, even though he’d thought the autopsy report looked mighty thin. He’d filed first-degree murder charges against the Lehmer baby’s parents. He’d put Joel Lehmer and Teresa Engberg-Lehmer in state prison for 15 years.

He’d never felt good about it, though. Almost any other prosecutor would savor his victory. Crowl stewed. He had abiding questions. The truth was, he’d almost been waiting for this day.

It was late on April 6, 1998. All afternoon, a defense attorney from halfway across the state had been standing over the prosecutor’s copy machine, duplicating the Lehmer file. Crowl watched him now. Stephen Brennecke meant to appeal the Lehmer case. He looked to be a straight-arrow, levelheaded sort.

“You ever handle a case like this before?” Crowl asked.

“Yes, I have,” Brennecke said. “The Weaver case.”

Crowl knew of the Weaver case. It was another one that involved Thomas Bennett--and another one that raised questions.

Crowl liked and respected Bennett, knowing the state medical examiner to be expert at recognizing the type of indirect brain injuries caused by the shaking of infants. Recently, though, other doctors had started to question publicly some of Bennett’s diagnoses. There’d been letters to the editor, letters to state officials. Some had been pretty disturbing.

There’d been claims that Bennett overreached, that he over-diagnosed, that he was overzealous. Where Bennett had found obvious signs of violent abuse, other doctors had found nothing. In one case up in Charles City, they’d stopped a murder trial on the third day of jury selection when prosecutors couldn’t find any medical experts to support Bennett’s diagnosis. In another over in Decorah, the prosecutor, faced with half a dozen doctors willing to challenge Bennett’s diagnosis, had declined to file charges.

The problem, it seemed to Crowl, reached well beyond the state medical examiner. Shaken baby syndrome had become a controversial subject across the country. Certain researchers surmised that 5% to 10% of deaths attributed to sudden infant death syndrome (SIDS) were actually shaken baby homicides, which meant there might be several hundred undetected baby killings each year in the United States. Yet which of the SIDS cases were the murders?

They debated that question in Crowl’s office all the time. Only a small number of babies who die suddenly, without explanation, are homicide victims. To spot abuse without accusing blameless people--that was a tough balancing act, that was a real dilemma.

Crowl was 39, and had been county attorney for eight years. During his tenure, he’d handled five child deaths that involved Bennett’s diagnoses. He couldn’t help wondering whether he’d been doing the right thing. He certainly didn’t want child abusers to walk free. Yet he couldn’t bear the notion that he might be putting innocent people in jail.

Crowl studied Brennecke. Slowly, the visiting lawyer fed sheets into the copy machine. There were the police interviews . . . there the hospital log . . . there Tom Bennett’s autopsy report.

Just four specks of blood in the skull--that’s what had convinced Bennett this was a killing.

“So what do you think about this one?” Crowl asked.

Brennecke looked up. “I don’t think this was a shaken baby.”

Mother Puts Baby to Sleep, Then Finds Him Not Breathing

Teresa Engberg-Lehmer’s story never changed. She fed Jonathan between 7 and 7:30 on the evening of Friday, April 4, 1997, then put the baby down to sleep on a blanket in the back bedroom. She and Joel went to their bedroom. About 11:15 p.m., both parents got up. She made coffee for Joel while he dressed. He left for work, a job delivering bundles of newspapers. She went into the baby’s room to give him a bottle. Jonathan was cold, non-responsive. Jonathan wasn’t breathing. They didn’t have a phone, couldn’t afford one. She screamed. She rushed to a neighbor’s apartment, asked her to call 911.

The police and hospital logs pick up the story from there. When the baby reached the Jennie Edmundson Hospital emergency room in Council Bluffs, the medical staff saw no signs of physical injury. They also saw no signs of a pulse. They pronounced Jonathan dead at 12:28 a.m. They notified Dr. Scott Blair, the county medical examiner, at 12:53. Blair requested that the baby be sent to Douglas County Hospital--minutes away, just over the Nebraska border in Omaha--for an autopsy.

That plan soon changed and, with it, the entire course of the Lehmer affair. At 2:10 a.m., Richard Stehly, a detective with the Council Bluffs Police Department, called the hospital to say the autopsy instead would be done on Sunday by Dr. Thomas Bennett in Des Moines, 130 miles away.

Crowl arranged this middle-of-the-night change, he would later explain, because he had “great respect” for Bennett and because the state was doing autopsies at “no charge to us.” Bennett, in other words, was the pathologist prosecutors turned to when a baby died; he was the pathologist most willing to “work with” prosecutors.

The drive from Council Bluffs to Des Moines, eastbound on wide-open Interstate 80, takes a good two hours, even with a heavy foot on the accelerator. Bennett met Stehly at the Broadlawns Medical Center morgue shortly after 4 p.m. that Sunday.

He entered the room as he usually does, with hearty greetings and good cheer. Bennett is handsome and charismatic, with a soothing voice and an amiable personality. He stands just over 6 feet. His mustache has started to turn gray, but at 45 he remains fit. He is considered a compelling courtroom witness, a superb “explainer” who talks in vivid, understandable terms. He is also considered a charmer, who will chat up everyone from courtroom clerks to opposing defense attorneys. He grew up on an Iowa farm. He puts on no airs. He projects unlimited self-confidence.

This last trait particularly irks his critics. Dr. Earl Rose, a retired forensic pathologist and former Dallas medical examiner who taught Bennett in medical school, recalls finding his student one day holding forth to classmates and an attending physician about the cause of a patient’s death--even though Rose felt Bennett had it all wrong. “I think he does that,” Rose said. “He speaks beyond his knowledge, but people believe him because he’s so affable and convincing. He has an assurance not backed up with knowledge.”

Others derive comfort from Bennett’s certainty. His admirers--among them various investigators, prosecutors and pediatric specialists--describe him as “devoted” and “caring.” They talk of how willing he is to clarify medical matters and get involved in possible child abuse cases. “I have complete confidence in Tom Bennett,” declares Floyd County prosecutor Marilyn Dettmer.

Bennett plainly inspires extremes of feelings: One doctor in Iowa offered the word “sociopath” when talking of him, while a prosecutor compared him to “Ben Franklin and electricity.” This disparity reflects not just Bennett’s character; it reflects also the emotional, volatile nature of the world in which he works.

That world has changed dramatically in recent years. There was a time not long ago when a child’s unexplainable death aroused little suspicion or attention. It was considered a private family matter. Babies were buried with causes of death listed as “undetermined” or “SIDS.” Then came several seminal articles in medical journals.

Among the most influential, published in 1972, was “On the Theory and Practice of Shaking Infants,” by pediatric radiologist John Caffey, who declared that many injuries attributed to disease or infection are in fact caused by the “whiplash shaking” of babies. Such “generally disregarded” whiplash shakings, Caffey wrote, “are precarious, pervasive, prevalent, and pernicious practices . . . . Whiplash shaking is widely practiced in all levels of society.”

The shaking, though, produced no obvious direct-impact injury, only indirect stress caused by the acceleration and deceleration of the brain within the skull. So there was a striking lack of overt evidence such as bruises and fractures. The two most consistently telling markers, Caffey reported, were subdural and retinal hemorrhages: bleeding in the brain and behind the eyes.

Caffey’s article didn’t produce immediate changes. In some quarters his findings were never embraced. There are still prominent physicians who insist it’s not possible to shake a baby to death. Yet by the mid-1980s, awareness across the country was mounting that not all babies died natural deaths. In Des Moines, Polk County authorities formed a child abuse trauma team consisting of social workers, police, prosecutors and doctors.

Thomas Bennett, named state medical examiner in 1986, soon became a critical resource for this team. Doctors hadn’t generally been inclined to work with law enforcement or, for that matter, to diagnose child abuse. Bennett was.

“There was a big difference between Tom Bennett and the doctors before him,” recalls Paul Houston, a Polk County attorney’s investigator who served on the trauma team. “Other doctors might think a case indicated abuse but then change their mind in court. The key was to bring in a physician willing to tell us whether it was an inflicted injury.”

That’s what Bennett did. From 1989 to 1997, his autopsies included 17 diagnoses of shaken baby syndrome. He also saw shaken babies in other cases where he served only as an expert witness. Understandably, the trauma team’s estimate of Bennett burgeoned. Four team members would in time declare him “one of the finest medical examiners in the United States.”

Others weren’t so sure. Here and there, a Bennett diagnosis started to raise eyebrows. Here and there, it looked as if Bennett had diagnosed shaken baby syndrome based on microscopic evidence--tiny tears and leaks in the brain--rather than the customary gross bleeds visible to the naked eye.

On Christmas Eve 1996, after hours of showing flu-like symptoms, 6-month-old Brennan Hutchinson of Decorah died in his mother’s arms. Doctors had seen no retinal hemorrhages when the baby first reached Winneshiek County Memorial Hospital. An autopsy revealed no gross visible bleeding in the brain. Bennett, though, detected minute drops of cerebral blood--and diagnosed shaken baby syndrome.

Then he did it again. On Feb. 8, 1997, 3-month-old Levi Sansgard of Charles City was found dead in his crib. There were no retinal hemorrhages, skull fractures or gross visible bleeding in the brain. The family physician, who was also the Floyd County medical examiner, diagnosed SIDS. Yet when Bennett reviewed the autopsy report, he noted evidence of small spots of blood in the brain--and thought it proof of a homicide.

Now, two months later, Bennett had been called to another baby’s suspicious death.

In the autopsy room at Broadlawns, with Jonathan Engberg-Lehmer’s body lying on a steel table, Det. Stehly greeted Bennett warmly; the two had often worked together. At 4:30 p.m., gowned and gloved, Bennett approached the baby’s body.

Pathologists are medical doctors trained to examine deceased and damaged organs. A forensic pathologist such as Bennett has additional training in the application of laboratory medicine to matters of law. They are accustomed to examining sudden, violent, unnatural, unexpected and suspicious deaths. Nationwide, there are only about 500 forensic pathologists; in Iowa, there are just five. This is understandable: The work of a forensic pathologist is not glamorous. With knives and saws, you slice open torsos, roll back flaps of skin, crack skulls, plunge hands into viscous innards. You pull out spleens and livers and kidneys, you weigh them, you dice them up. Black bile oozes, blood drips, odors assault.

“The truth is,” Bennett likes to say, “no one really wants to do autopsies. Especially on kids. I got involved with kids’ autopsies because no one else wanted to do them.”

The body before Bennett was tiny--just 15 pounds, 29 inches. Jonathan had blue eyes, blond hair. He looked unscathed. As Stehly recalls it, Bennett in these first moments said he thought this was a SIDS case. Then the medical examiner wielded his knife.

He carved open the chest and abdomen, removed the organs, seeing nothing significant. At the far back of the baby’s head, he sliced from behind, starting at one ear, ending at the other. He folded back the scalp. He picked up an oscillating saw. He was cutting through bone now, across the forehead to the ears, then around the rear of the head. He pulled off the skull cap--forensic pathologists call it “popping the skull.”

He studied the inside of the skull that he held in his hand. He saw four small dark gelatinous blotches--blood--each about one-quarter of an inch in diameter. He looked down at the exposed brain. He saw one cubic centimeter of blood, a small clot about the size of a garbanzo bean.

Most forensic pathologists would not think that such spots signified anything. Most would call them “artifact”--blood that leaked during the popping of the skull. Most would need to see 10 to 20 times that much blood before calling it a subdural hematoma. For Bennett, though, it wasn’t even a close call. The trauma to this baby was not only severe but obvious.

“Gentlemen,” he informed the watching detectives, “you have a homicide.”

County Prosecutor Starts to Have Doubts About Case

Pottawattamie County Atty. Rick Crowl took his time with the Lehmer matter. That’s his custom. He thinks before he speaks. When he does open his mouth, he offers direct, unaffected sentences. He has soft features and a high receding hairline. He favors sweaters instead of coats and ties in the office. His demeanor is relaxed. Bland might be a fitting description, except that his calm hints at a certain resolve.

Studying Jonathan Engberg-Lehmer’s death certificate, Crowl could see that Bennett had listed “shaken-slammed baby syndrome” as the immediate cause of death. Some doctors, facing ambiguous situations, preferred using this umbrella phrase that encompassed not just shaking, but also the forceful striking of heads against a surface. In the autopsy report, though, Bennett had made it simply “slammed baby syndrome.”

Was the baby shaken or not? What struck Crowl most was the absence of the usual retinal hemorrhages seen in such cases. “If the baby was forcibly shaken,” Bennett had written, “the shaking was of short duration (no more than a second or two).”

Crowl had never seen a pathologist make such an assessment before. Bennett’s “second or two” seemed odd. He’d asked Det. Stehly to seek clarification.

“It has become clear,” Bennett had responded, “that any individual has the potential to be violent. . . . It only takes a couple three seconds or so to produce these injuries. That does not take away from the severity of the injuries, because these are so severe.”

Bennett insisted: The injuries came “within minutes to perhaps a half-hour” after Jonathan’s last feeding. “To a reasonable degree of medical certainty . . . these are nonaccidental injuries.”

Crowl often attended autopsies. He now regretted that he hadn’t gone to the Lehmer baby’s. What, he wondered, to make of all this?

He’d been starting to hear something of a drumbeat against Bennett, but sitting out on the far western edge of Iowa, just across the border from Nebraska, he remained insulated from his own state’s events. Bennett was the state medical examiner, Crowl told himself. Bennett was saying this baby died of shaken-slammed baby syndrome. You had to go with that. You had to trust Bennett’s opinion.

Or did you?

Crowl’s question and the discord in Iowa, although particularly heated and personal, reflect a continuing, deepening controversy over shaken baby syndrome. Police, prosecutors and pediatric experts have increasingly employed the syndrome as a way to explain the deaths of children when there are no outward signs of physical injury. Yet the forensic pathologists who actually rule on causes of death remain divided.

Some think shaken baby is a rare syndrome; others think it’s common; and many fall somewhere between.

Robert Kirschner at the University of Chicago, a former deputy medical examiner for Cook County, says, “The fact is, babies are shaken to death. In good hands, this diagnosis is being made regularly and accurately.”

Yet Cyril Wecht, coroner for Allegheny County in Pennsylvania, says the syndrome is so over-diagnosed he wouldn’t allow his kids to baby-sit other children because “accidents happen, and these kinds of cases tend to take on a life of their own.”

Vincent DiMaio, San Antonio’s chief medical examiner, goes still further. He says he’s never seen a shaken baby death in nearly 30 years of practice.

Last August, the respected British medical journal Lancet addressed this situation in a lead editorial: “Since the ‘shaken baby syndrome’ was first described by Caffey in 1972 . . . there has been disagreement about which features comprise the syndrome and how shaking might cause them. . . . With such an uncertain definition, it is not surprising that the syndrome has been misdiagnosed. . . . The consequences of such an error are appalling. . . . If 26 years after Caffey’s description, doctors are still undecided about the ‘shaken baby syndrome,’ the difficulties faced by experts in presenting medical evidence in court, and by the judge and jury in making sense of it, are readily imaginable. . . . Let us hope that increased awareness is tempered with caution against over-diagnosis.”

The reason for such confusion is apparent: Arguments over shaken baby syndrome turn as much on matters of ideology as on hard science.

On the one hand, there are those who feel cautious about using uncertain evidence to issue opinions that send people to prison, possibly for life. On the other hand, there are those outraged by the continuing--and often unpunished--abuse of children. Both sides claim high principles; both sides express genuine dismay at what others are doing. A concern for civil liberties ends up clashing with the urge to protect children. It’s not surprising that the discourse gets nasty.

Some pathologists are accused of being “prosecution advocates,” some are called “defense whores.” Those who think shaken babies are commonplace draw charges of being “cultists.” Those who don’t, draw charges of being callous--if not child abusers themselves.

It’s an oversimplification, but not by much, to say that the pediatric specialists who care for children are more inclined to diagnose abuse than the forensic pathologists who rule on causes of death.

“You can’t be an advocate and medical examiner,” maintains Peter Stephens, a forensic pathologist in eastern Iowa who challenged Bennett’s diagnoses. “Pathologists are desperately trying to evaluate each case, to not miss it when it’s there or see it when it isn’t.”

“You can argue all you want,” counters Randell Alexander, a pediatric specialist who, while at the University of Iowa Hospitals and Clinics, worked closely with Bennett. “Meanwhile, the child is in my ER with broken bones.”

It’s that child in the ER--or morgue--who finally makes the shaken baby controversy most complicated. Say what you will, it’s hard to challenge someone who’s campaigning against baby killers. To do so isn’t politically palatable, or, for many, personally appealing. Dismay over child abuse trumps concern for civil liberties.

The nature of those usually accused in child deaths further confounds the situation. They most often are poor and uneducated. Often they’re not the best caregivers. Often the boyfriend of an unmarried mother is involved. Often the boyfriend has a history.

As it happens, the Lehmer family in Council Bluffs fits this profile fairly well. In weighing what to do about Jonathan’s death, Rick Crowl faced not just a difficult scientific dispute and a charismatic medical examiner. He also faced a pair of parents who were far from pristine.

Both Joel and Teresa, as Crowl understood it, had problems with drugs. They were so poor they couldn’t afford a crib, so their baby slept on a blanket spread on the floor. They didn’t marry until after the baby’s death. Joel Lehmer previously had turbulent relationships with two other women. In 1993, he’d pleaded guilty to assaulting a girlfriend. Twice he’d relinquished parental rights to babies he’d fathered. He’d been urging Teresa to put Jonathan up for adoption. The day before the baby’s death, they’d been arguing over that issue. Later, Joel, distraught, told a relative that he’d been “mean” to the child.

Crowl knew he could use Joel’s comments in court. Coupled with Bennett’s diagnosis, that certainly made it a case to prosecute. Against Teresa as well as Joel: If a caretaker allows abuse to happen, that’s child endangerment, whoever did the shaking. Maybe she did it, for that matter. Their story left her alone with the baby, not him. It could have been Teresa.

Just one thing troubled Crowl: The Lehmers wouldn’t give each other up.

He’d been making offers back and forth, first to Joel, then Teresa. Working each against the other. Almost always, you get one to turn. Not this time. Crowl went so far as to offer dismissal of charges for whoever would testify against the other. Neither would.

“I never shook Jonathan,” Teresa told detectives over and over. “I never personally saw Joel do it. . . . I never saw Joel shake him. . . .”

“We have the medical evidence,” the detectives kept saying. “There is no reason that Dr. Bennett would rule this if it didn’t happen. He deals in facts only. . . . There is no reason for Dr. Bennett to lie.”

“No. . . . no. . . . no. . . . no,” Joel would insist. “NO! No, sir, no, sir. I didn’t shake the baby. Teresa didn’t shake the baby. I know Teresa better than she knows herself, and she didn’t do this.”

Crowl had to admit that’s just not the way these things usually worked. No gross bleeds, no retinal hemorrhages, no parent flipping on the other. . . . This wasn’t one of his strongest homicide cases

Still, the state medical examiner was calling it an obvious murder. Tom Bennett thought this a graphic example of shaken-slammed baby syndrome. For so long--too long, Crowl knew--prosecutors had failed to recognize child killings. You had a dead baby here, you had an offensive dad, you had an established expert talking with polished assurance. How could he buck that? How could he ignore a baby’s death?

In the end, he decided he couldn’t. After conferring with his investigators, Crowl in July 1997 filed first-degree murder charges.

Teresa drew county public defender Greg Steensland, Joel a local court-appointed lawyer named Drew Kouris. Neither attorney deposed Bennett, so neither questioned him about how he reached his diagnosis. Kouris did write John Plunkett, a Minnesota forensic pathologist, asking if he would review the case for the defense. Before he got a response, though, Crowl offered a deal: Plead to involuntary manslaughter and child endangerment; we’ll agree to 15-year sentences.

The couple and their attorneys pondered. If they went to trial, a jury could find them guilty of first-degree murder. Under Iowa law, even second-degree murder would mean at least 42 years in prison. How could they expect to win? Especially since, as Drew Kouris put it, Thomas Bennett was a “superb courtroom witness with a national reputation.”

The couple’s lawyers nudged them toward accepting Crowl’s plea. If they accepted, Steensland told Teresa, “I would consider that an intelligent decision given the findings of the medical examiner.”

No, the couple said. We won’t say we’re guilty.

Case law offers an alternative: an “Alford plea,” in which you maintain your innocence but admit that the county attorney can prove his case. On Oct. 2, 1997, the couple entered such a plea. “We were literally scared into taking the plea,” Teresa Engberg-Lehmer told a reporter. “We were both told we could get a life sentence.”

Two weeks later came surprising news out of Des Moines: Bennett, facing an ongoing state investigation into the administration of his office, had resigned as Iowa’s medical examiner.

Medical Examiner Bennett’s Critics Gain Strength

Officially, Tom Bennett’s resignation resulted from an investigation into his habit of using state resources when he testified and performed autopsies for private corporate clients. This had long been his custom, though. That it suddenly drew scrutiny and rebuke reflected the growing influence of Bennett’s critics.

The gloves first came off in March 1997, at the third trial of a small-town baby-sitter, Mary Weaver, charged with the killing of an 11-month-old. The dispute wasn’t whether Melissa Mathes had been shaken and slammed--she had a massive skull fracture and extensive bleeding--but when it happened and whether it was the fatal injury. Bennett and seven other state experts thought Melissa died not of the earlier skull fracture but of a later injury inflicted by Mary Weaver. Rising to challenge their judgment was Dr. Thomas Carlstrom, an outspoken Des Moines neurosurgeon who’d written Weaver’s attorney, volunteering his services. It seemed clear to him that Melissa died of her days-old skull fracture; he believed Mary Weaver innocent.

It’s hard to underestimate Carlstrom’s impact, not just on the Weaver case but on all that followed. The brain specialist proved as potent a courtroom weapon as Bennett. He helped the defense shape its opening argument and questions. Then he took the witness stand, to great effect. On March 3, the jury acquitted Mary Weaver.

Bennett couldn’t contain himself; his usual amiability gave way to something that sounded to some like harsh arrogance. Within a day, he was publicly sniping at Carlstrom and urging that complaints be brought before the Iowa Board of Medical Examiners: “When a neurosurgeon can testify that sitting down hard can seriously bruise the base of the brain, that needs to be looked at closely. . . . I think that perhaps criteria have to be established for who can and who can’t be an expert witness.”

Carlstrom fired back: “I think there are a lot of people who are flaunting their opinions who don’t know what they’re talking about. But I’m not one of them. . . . I think sometimes Bennett might be a little bit overzealous in his pursuit of a shaken baby case.”

In this way, a relatively private professional dispute came unlatched. Those forensic pathologists who’d been seething silently over Bennett’s methods now stirred. Dr. Francis Garrity, the Polk County medical examiner, wrote Weaver’s lawyer, saying he wished he’d come forward to help also. Peter Stephens, who is a former deputy state medical examiner, made it known that he would review other Bennett cases without charge and, if merited, serve free as an expert witness.

“After Weaver, there was very much an aspect of the ‘emperor has no clothes,’ ” Stephens later explained. “Some of us were encouraged to speak out.”

Within weeks, both Stephens and Garrity agreed to look at the Christmas Eve 1996 death of 6-month-old Brennan Hutchinson of Decorah. So did Earl Rose, the retired forensic pathologist who’d once taught Bennett. So did John Plunkett, the forensic pathologist in Minnesota. So did Henry Krous, a San Diego specialist in pediatric pathology. Each, it turned out, had much the same response: Brennan had died not of shaken baby syndrome, as Bennett diagnosed, but of brain swelling due to a viral infection or metabolic disorder.

Wrote Rose, who besides being a forensic pathologist is also a general practitioner and lawyer: “I am reluctant to believe that Dr. Thomas Bennett would risk the credibility of his office as the Iowa state medical examiner by drawing the serious and far-reaching conclusions of ‘shaken baby syndrome’ and ‘homicide’ based on these meager findings.”

In the end, faced with five expert witnesses saying Brennan died of natural causes, the Winneshiek County prosecutor never filed charges. This decision didn’t forge a pattern, though. Handed a similar situation in the death of Levi Sansgard, Floyd County Atty. Marilyn Dettmer responded differently: She charged Levi’s mother, Melissa Funte, with murder.

The by-now familiar challengers to Bennett arose, each insisting there was no evidence of any injury or trauma. Three times, Funte turned down plea bargains, the last one being an offer of no prison time in exchange for a no-contest plea to a misdemeanor charge. Jury selection began in January 1998. Then, on the second morning, a lawyer from the prosecution team visited Funte’s attorney, Cheryl Weber. He had information that Weber might think exculpatory; he felt obliged to hand it over.

The state, it turned out, had been canvassing experts around the country, looking for one to support Bennett. They’d failed, finding instead only critics. Among them was St. Louis chief medical examiner Mary Case, a leading authority who was almost always a prosecution witness. Weber promptly called Case. The St. Louis pathologist offered to testify for the defense. “I do not find any evidence of injury to this child’s head,” she wrote. “These [microscopic findings] do not represent any type of trauma.”

Put up or shut up, the judge told the prosecutors on the last day of jury selection. The prosecutors asked for more time. The judge said no. The prosecutors moved to dismiss without prejudice, meaning they could refile later. They dropped the case, they announced publicly, because “new evidence had surfaced.”

That explanation was just too much for those now openly challenging Bennett. Both Francis Garrity and Peter Stephens wrote letters to Iowa’s commissioner of public safety, who oversaw Bennett’s office.

The Funte case, Garrity argued, “underscores the urgent need for the Courts and State Medical Examiner system to further distance themselves from Dr. Bennett. . . . The state should carefully review Dr. Bennett’s opinions in similar cases before additional, embarrassing ‘new evidence’ surfaces. . . .”

Months passed. Silence fell. A review of Bennett’s cases never came.

He was a lawyer, not a doctor, Public Safety Commissioner Paul Wieck eventually explained. It was not possible for him to evaluate this huge discrepancy among forensic pathologists. It was not possible for him to make a judgment between competing medical examiners. To do so, at any rate, wouldn’t be appropriate. These were county cases, filed by local prosecutors, not the state. “I have no business intervening,” he declared. “It’s not my job.”

A Letter From Prison and a Plea for Help

On Nov. 14, 1997, Teresa Engberg-Lehmer, sitting in her cell at the Iowa Correctional Institution for Women in Mitchellville, took pen to paper and began a letter to an attorney from Grinnell whose name she’d noticed in the newspaper.

“To Mr. S. Brennecke,” she wrote. “My name is Teresa Lehmer. . . . We, my husband and myself, both feel that we are a victim of circumstances, probably due to Dr. Bennett. . . .”

Stephen Brennecke read Teresa’s letter with some interest. As it happened, he was a friend of the baby-sitter Mary Weaver. He’d served as her lawyer at the second trial and had assisted at the third. He’d written a letter of complaint about Bennett to the Des Moines Register. That’s what had drawn Teresa’s attention. Now here was her handwritten plea.

We want to thank you for your article in the newspaper. We believe it was inspired by God. . . . If there is any information that you may have that can help us. . . We would greatly appreciate anything. . . . May God bless you. . . .

In the courtroom during the Weaver case, Brennecke had found Bennett personable, but also annoying. Bennett would approach him and Mary Weaver and chat them up, even though he could see they were trying to prepare for the day’s session. “Hi, how you doing?”--stuff like that. He’d also told Brennecke his client was guilty. Brennecke thought that unprofessional. Bennett sounded to him like an advocate, not a medical expert.

More than an advocate. In such forums as the Iowa State Medical Examiner’s news bulletin, Bennett was now issuing alarums about shaken baby syndrome: “A major obstacle in the identification and adjudication of these cases had been denial by professionals and public alike that people could act violently against children. . . . It’s well documented that frequently there is no justice, criminally, in SBS cases.”

Bennett, it seemed to Brennecke, truly believed in what he was doing. So much so that he flatly dismissed critics and never wavered from his findings. He wouldn’t change any of his diagnoses, he told reporters. He invited anyone to review his work. He had piles of supporting literature. He was in the mainstream of medicine.

Why, Bennett even felt able to look at the X-ray of a dead child and say how much anger the mother had. That, Brennecke thought, was ridiculous.

OK, he informed Teresa. I’ll take your case.

I understand there’s a forensic pathologist who will look at Bennett’s cases free, Teresa told him. Peter Stephens, that’s his name.

Stephens was interested; Stephens asked Brennecke to get him the full Lehmer file. He wanted everything: the autopsy report, depositions, slides, photos, police records. In early March, Brennecke wrote to the state medical examiner’s office, seeking these items. For two weeks, he drew no response, despite a series of follow-up phone calls. Then, on March 18, the attorney general’s office called back. Inadvertently, the office explained, all slides and tissue samples from the Lehmer case had been destroyed back in January.

Not just the Lehmer case: In the wake of Bennett’s resignation, someone at the state crime lab had improperly destroyed samples from 356 autopsies conducted by Bennett from 1995 to 1997. Five involved infant deaths. Four of those five were shaken baby homicides.

The four shaken baby cases, as it happened, received special treatment: In those, the crime lab destroyed not just tissue samples but all biological materials.

“Autopsy Evidence Destroyed in Violation of State Rules,” declared a Des Moines Register headline on Page One three days later.

“I am chagrined,” Public Safety Commissioner Wieck was quoted as saying. “It was a poor decision that was contrary to policy.”

“This is outrageous,” Brennecke declared. “They deliberately destroyed these materials to cover up glaring omissions by Bennett.”

There is no way of knowing what happened. Iowa’s governor ordered the obligatory investigation but named Wieck to head the probe of his own department’s conduct; Wieck soon was talking about “looking at general issues” and “assessing the way we do business.” All that can be said for certain is that Brennecke ended up receiving from the state only 14 color autopsy photos.

“What more can you get me?” Peter Stephens asked.

In early April, Brennecke called Pottawattamie County Atty. Rick Crowl. “If it’s OK with you,” he said, “I’d like to come visit.”

Crowl Characterized as Being Fair

For a while, watching Brennecke duplicate the Lehmer file on the afternoon of April 6, Rick Crowl managed to appear resistant. This is a good conviction, he told his visitor. I sure think this is a shaken baby. Yet Crowl didn’t sound the least contentious. He sounded open-minded.

In truth, Crowl was once again wondering about a case he’d won. This custom of his sets him apart from most prosecutors. The archetypal prosecutor digs in his heels when challenged and insists he only files charges against those he’s certain are guilty. Crowl, on the other hand, admits that he can’t honestly say he’s never targeted the innocent. He isn’t always positive. He doesn’t know how any prosecutor could be.

This Lehmer case had left him even more uncertain than usual. It had simmered in his mind ever since the couple went off to prison.

“Why don’t you think this was a shaken baby?” Crowl asked Brennecke.

“For one thing,” Brennecke said, “there weren’t any retinal hemorrhages.”

That was true, Crowl had to admit. There had been a plea bargain, though. The Lehmers were sitting in jail because they’d pleaded. Had he set up pressures that forced the Lehmers to do this? That was surely possible.

“You know,” he told Brennecke. “I’m willing to listen. I don’t mind taking a second look.”

In late June, Brennecke called Crowl and read to him from Peter Stephens’ written report. It was unequivocal: “The autopsy findings do not indicate that Jonathan sustained any form of head injury. . . . I do not believe that any responsible pathologist would certify death as being due to trauma based on the evidence that I have seen to date.”

Crowl hesitated. He didn’t know Peter Stephens; they worked and lived on opposite sides of the state. He did know Stephens was among the more active, vocal critics of Bennett. It would be so much better, Crowl reasoned, to hear from a neutral expert whom he knew and trusted. Someone who had always told him straight out whether he had a case.

“I want you to send the Lehmer file to Jerry Jones in Omaha,” Crowl told Brennecke. “Let’s see if you can convince him.”

Pathologist Who Reviews Case Is Stunned by Report

Over the course of 30 years, Jerry Wilson Jones has performed about 5,000 autopsies for prosecutors in Douglas County and a host of other jurisdictions in Nebraska and southwest Iowa. He is board certified in anatomical, clinical and forensic pathology. He teaches in Omaha at the University of Nebraska Medical Center and the Veterans Affairs Medical Center. He’s slow and heavy-set. His hair, what’s left of it, is graying. His instinct, at all times, is to teach.

“Take a look at this,” he urges, pushing a document across his desk with disdain. It’s yet another example of the discord over shaken baby syndrome--an article published in the New England Journal of Medicine in June 1998, written by a team of doctors at Children’s Hospital of Philadelphia headed by Ann-Christine Duhaime. The much-discussed Duhaime paper questioned whether shaking alone could cause the type of injuries associated with shaken baby syndrome. Better to call it “shaken-impact” syndrome, Duhaime suggested, for it’s the forceful striking of the head against a surface that’s responsible for “most if not all” severe inflicted brain injuries.

“Absolute nonsense,” Jones mutters as he glares at this article. “I doubt the authors are forensic pathologists or see actual autopsies. That their article is distributed by such a prestigious journal, that it gets promulgated. . . . It muddies the waters. To my way of thinking, there’s no question you can shake kids to death, no question they are being shaken to death. It’s nonsense to say they can’t.”

Jones has diagnosed shaken baby deaths eight times. He’s interested in the shaken baby syndrome. He has researched retinal hemorrhages as a marker for the syndrome. He has made presentations on the subject at international conferences. He believes the diagnosis for shaken baby to be exactly as John Caffey described it. “You can see it,” he declares. “If you adhere to the criteria, you can make the diagnosis.”

For all these reasons, Jones did not respond warmly when Brennecke called him in September. If anything, he felt disposed against him. He agreed to review the Lehmer case, Jones would later say, “only because Rick Crowl asked.”

Then, one morning in late September, he opened the Lehmer file for the first time. He read it once, twice, three times. Bennett’s autopsy report stunned him. He couldn’t put into words how much it stunned him. Jones saw no evidence of a shaken baby, no evidence of any trauma.

Single Centimeter of Blood Called Critical

Bennett was calling a single cubic centimeter of blood a subdural hemorrhage. That was one-fifth of a teaspoon. Give me a break, Jones thought. One cubic centimeter just didn’t demonstrate an acute subdural. That amount could leak as you pop off the skull cap. You pull the cap, you can tear vessels.

Bennett described an area of bleeding beneath the skin in the back of the right side of the scalp. Jones studied the photograph taken of that area. It was badly out of focus, but even with a blurry picture, he could see a bleed if there were one. At best, this photo showed a normal, gray-yellow patch to be partially gray, a little red.

Jones lifted up another photo. This one showed small drops of blood on the inside of the skull cap. These also could routinely be collected when you removed the skull. Again, Jones wasn’t convinced.

Photo after photo, Jones turned through the pile before him. He couldn’t see how Bennett possibly had made his diagnosis. Jones wasn’t at the autopsy, but he had the photos, and gross brain bleeds would be easily visible. There just was no evidence here to support a “shaken-slammed baby syndrome” diagnosis. He knew for sure this wasn’t a shaken-slammed baby.

Jones felt disappointed. It seemed to him that Bennett’s zeal cast doubt not just on that man’s own diagnoses but also on legitimate shaken baby cases. He feared that Bennett’s work could raise doubts about whether doctors can diagnose this syndrome.

Here finally was the most effective catalyst behind the challenge to Bennett: Not the self-serving outrage of a defense attorney, not the scoffing dismissal of those who insist they’ve never seen shaken baby deaths, but the frank dismay of one who knows that caretakers do kill children.

Jones reached for the phone. It was late on Sept. 24, a Thursday. He punched in Rick Crowl’s phone number.

“This is not trauma,” the pathologist told the prosecutor. “This absolutely is not shaken-slammed baby syndrome. This is SIDS. These people are innocent. They should not spend another day in jail.”

‘Unique’ Way of Diagnosing

For a time, as the shaken baby controversy unfolded in Iowa, it was possible to regard the challenges to Thomas Bennett as evidence of personal animosity. His two former deputies, Francis Garrity and Peter Stephens, had often clashed with him. They complained that Bennett ran his own fiefdom. Garrity can’t stand Bennett; Stephens says, “I don’t respect the man.”

Yet in the end, this undeniable rancor couldn’t explain the conflict. There were too many experts disputing Bennett from all perspectives, many with no personal connection to him. Even the staunchest believers in shaken baby syndrome were balking.

One of those, the respected St. Louis medical examiner Mary Case, let it be known in August--just weeks before Jerry Jones reviewed the Lehmer case--that she felt Bennett had a “unique” way of diagnosing shaken baby syndrome. “He’s looking at normal brains at autopsy and seeing tears in the tissue that he says resulted from shaking. As far as I know, he’s the only one who is doing it that way. . . . I see the microscopic tears he’s talking about. Those are artifacts. You see them all the time. They mean nothing.”

For Rick Crowl, Jerry Jones’ report was the crowning blow. Not even Bennett’s customary emphatic defense--”I stand by my findings. . . . It’s not like I’m making it up”--could sway him any longer. He still liked and respected Bennett but felt he had to go with what the experts were saying. The fact was, the state no longer had probable cause. Even if Joel Lehmer were mean to that baby, even if he wanted to put it up for adoption, that baby still didn’t die of trauma.

Or, at least, Crowl couldn’t prove it. He couldn’t show that anything the parents did caused their baby to die. No way could the state win at a retrial. The defense would have Jerry Jones, plus the destroyed slides. By law, it would be assumed that missing evidence was beneficial to the defendant.

It’s hard to imagine any other prosecutor doing what Crowl now did. It was one month before election day, and Crowl was up for reelection. The Lehmers were a dubious couple at best. They’d pleaded; he had them in jail. The easiest thing would be to let Brennecke file a motion for post-conviction relief. Go through the lengthy appellate procedure, wait out the election.

Instead, Crowl, on the evening of Sept. 24, just hours after hearing from Jones, called Judge Timothy O’Grady of the Iowa District Court in Pottawattamie County. “We have a problem,” Crowl began. “I think we have some people in prison who shouldn’t be there. I don’t think I have a provable case.”

The hearing commenced at 2:42 the next afternoon at the Pottawattamie County Courthouse in Council Bluffs. For the county appeared Rick Crowl. For the Lehmers appeared their original lawyers, Drew Kouris and Greg Steensland.

Brennecke was out of the country on an 11-day tour of China. Crowl, not wanting to wait, had called Kouris midmorning, without advance notice. We’re going to court, he’d advised. You need to prepare a motion for a new trial.

“These motions,” Judge O’Grady began, “allege that there has been evidence discovered since the judgment in these cases. Mr. Crowl?”

The county attorney rose. “Yes, your honor. The state is not going to resist the application, but in support of the state’s position, we would like to call Dr. Jerry Jones.”

On the stand, Jones again was unequivocal: Completely unconvincing . . . I could not interpret this in any way, shape or fashion as evidence of subdural hemorrhage. . . . I completely disagree with the findings and the interpretations by Dr. Bennett.

When Crowl finished his interrogation, Judge O’Grady posed his own question: “Dr. Jones, . . . How is it that your opinion differs so much with Dr. Bennett’s? Do experts commonly disagree with interpretations of those photographs?”

Jones shifted in his chair, plainly uncomfortable at the notion of openly criticizing a colleague. “Well,” he said. “I cannot explain Dr. Bennett’s interpretation of this. I believe it is completely incorrect, and I don’t believe that other competent pathologists in reviewing this would come to the same interpretation as Dr. Bennett. . . . I don’t know how else to answer the question. . . .”

The standard for meriting a new trial is whether “important and material” evidence has been discovered that the defendants “could not have discovered before trial.” There’d been no new evidence at all, of course, but since that’s what was needed, that’s what was recognized.

“I find that the standard has been met,” declared O’Grady. “The motion for a new trial will be sustained.”

Crowl rose. “Your honor, in light of the newly discovered evidence, the state moves to dismiss both cases in the interest of justice.”

First the judge made sure neither defense attorney had an objection. Then, at 3:45, 63 minutes into the hearing, he said, “The state’s motion to dismiss will be sustained.”

Unaware of what had transpired, Brennecke arrived home from China late the next night, a Saturday. When he opened his Sunday newspaper, he saw the banner headline atop Page One: “Jailed Parents Cleared in Death.”

Bennett Defends His Report

After the Lehmers won their case, it took a few hours for reporters to track down Bennett. By then he’d moved to Billings, Mont., where he works now as a private pathologist. That Saturday afternoon, Bennett, presented with the news out of Council Bluffs, insisted that he still had truth on his side. He not only defended his findings--he added to them the retinal hemorrhages that were notably absent in his autopsy report.

In what he later explained was a lapse of memory, he told a Des Moines Register reporter that Jonathan had bleeding in the brain and eyes: “You cannot diagnose SIDS when there is injury to the brain, including gross bleeding and retinal hemorrhages.”

He issued a warning to Francis Garrity and other doctors: “The statements . . . made by various pathologists . . . are irresponsible, venomous and guaranteed to wind them up in court because of the libelous nature.”

He declared his nonchalance over the Lehmers’ release: “I don’t care about that. If Pottawattamie County wants to give them a medal, that’s OK, as long as they do it using all the facts.”

Three days later he leveled even stronger fire in a Des Moines Register opinion-page essay: “My respect for and belief in our judicial system have led me to maintain my silence during these attacks. These malicious attacks now have overstepped all bounds of professionalism and demand a response, in the media and the courts. . . . Why should any medical person subject himself to this abuse? . . . There is a sickness in Iowa. . . . Children in Iowa will suffer. . . .”

For weeks after that, Bennett remained on the attack.

He took aim at the news media: “Glitzy news columns . . . Geraldo Rivera style . . . Pack of dogs.”

Taking Aim at Prosecutors, Doctors

He took aim at weak-willed prosecutors: “To go into court and offer your opinion, you have to have conviction and courage. When those using your services don’t have the courage to stand with you, you end up standing alone. Politicians will do that. . . .”

He took aim at other doctors: “Every time you go into a courtroom, there will be experts on both sides. Many testify outside their fields. Many testify with no basis. . . . “

Most intriguing, Bennett continued to profess bewilderment over the startling disparity between his judgments and those of so many colleagues. It seemed to baffle him as much as it did Jerry Jones that he saw acute abuse where others saw nothing.

“I don’t know why,” he said more than once when asked about this disparity. “I just don’t know why. . . . I ask the same question. . . . I don’t know why. . . .”

Not until early December did Bennett finally drop this stance.

For all his contentious impact in Iowa, Bennett had always tried to hold himself above the fray. He’d regularly provided vivid quotes to reporters, but only in bursts that defended more than explained. These weren’t matters to be argued in the newspapers, he declared more than once. These were matters for the courtroom and the medical journals.

For weeks, Bennett had been hearing pleas to engage in a more extended discussion. For weeks, he’d resisted, saying, “I’m so tired of being hammered.” Finally, near the end of the year, he reluctantly agreed to receive a visitor in Billings.

He lives there with his wife, Melodee Hanes, a former assistant Polk County attorney who, as it happens, specialized in child abuse and helped organize Polk County’s child abuse trauma team. The first night, Hanes came to dinner alone, since her husband had been called to an out-of-state hearing. The next morning, Bennett himself showed up for breakfast. Neither displayed the angry tone evident in some of their public statements. It’s hard to demonize someone who, like Bennett, grows moist-eyed with hurt when his critics’ comments are read to him. It’s hard also to confront someone who speaks passionately, as Bennett does, about “the worth of a child’s life.”

Bennett and Hanes resemble the affluent professionals they are: They enjoy cooking, they like good wine, they ski, they share a large comfortable home with the three youngest of their combined six children. He teaches and performs autopsies; she is thinking of signing on with a local prosecutor as head of a new child abuse unit. In conversations, they jokingly referred to Bennett as the “defrocked pathologist.” They also laughed at suggestions--heard often in Iowa--that she has influenced his diagnoses. “Right,” Hanes agreed. “I told him no sex unless he calls them shaken babies.”

Most striking, they now made no pretense that Bennett’s contested diagnoses reflected mainstream thinking in forensic pathology. “Tom’s on the cutting edge,” Hanes declared at the first night’s dinner. “He wouldn’t say it, though. He’s too much the plain country boy.”

The next morning, Bennett did say it.

Over cups of coffee in a cafe, as light snow fell in downtown Billings, Bennett sketched diagrams, produced journal articles, recited history and told his story.

It began with the Polk County trauma team, he explained. Then it really coalesced around 1996. That’s when the pediatric specialist Randy Alexander asked him to come talk to Iowa state child protection workers. Bennett started to study Alexander’s work. He started thinking and reading and attending seminars. He found himself conferring with doctors outside his own field--doctors who were inclined to diagnose shaken babies using criteria other than gross bleeding in the brain and eye.

There were radiologists who thought microscopic tears in the brain signaled a shaken baby. There were pediatricians who thought brain swelling alone, without signs of bleeding, signaled a shaken baby. Some were involved with Bennett’s own multidisciplinary team in Iowa. Was he going to just ignore these other specialists? Some were on the American Academy of Pediatrics’ committee on child abuse and neglect. Had you seen their pivotal 1993 paper?

Bennett came to think them right. He wasn’t the leader here. He was “standing on the shoulders of giants and leaders.” He was “synthesizing.” He was saying you can make diagnoses of shaken baby syndrome based on the microscopic evidence of tiny tears in the brain. He was saying that a swollen brain with microscopic tears was an injured brain.

“Not all shaken babies are going to have visible bleeding in the brain,” Bennett maintained. “Not all shaken babies are going to have bleeding behind the eyes. You need 18 hours after the injury to see a subdural hemorrhage develop. But the kids aren’t living that long. That’s why you didn’t see those hemorrhages.”

Here was a Tom Bennett not evident before in the Iowa furor. No longer was he blaming the brouhaha merely on personal antagonisms or subtle professional differences. Now he was saying, yes, I am pushing the envelope. Now he was proposing nothing less than an expansion of the basis for diagnosing shaken baby syndrome.

Yet he didn’t have even the staunchest shaken baby experts with him on this--not even those experts he’d urged reporters to call. Some said he flat-out misused the literature; some said his theories were preposterous.

There was Mary Case in St. Louis: “Bennett’s on the cutting edge? Well, that’s not true. Why would he have unique knowledge that no one else has? He hasn’t always done this. Bennett seemed to change. His job got tied to law enforcement. He became personally involved in a way that’s not good.”

There was Robert Kirschner at the University of Chicago, after reviewing the Lehmer file: “It’s not a convincing shaken baby case. . . . There’s no retinal hemorrhage. . . . The blood in the skull is probably artifact. . . . The Jerry Jones report is very reasonable. . . . Tom Bennett probably has over-called cases.”

Bennett, reminded of such comments, sounded undaunted. OK, he responded. It was true: Not many forensic pathologists saw it his way. They were wrong, though. They were so entrenched, they weren’t able to see different points of view. In 10 years, they’d see it his way. In 10 years, his view would be accepted as the standard threshold.

He was out of the cafe now, striding down Billings’ snowy downtown streets. Bennett’s animation over his ideas mounted. He was going to present his views at a conference in Florida. He was fully aware of the consequences of his diagnoses. He wouldn’t be doing this if he didn’t truly believe. That’s the question: What do you believe in? He believed in science. He worshiped at the altar of science. He deeply and truly believed that within a decade, his ideas would be accepted as the standard threshold. All truth passes through three stages. First, it’s ridiculed. Second, it’s violently opposed. Third, it’s accepted as self-evident. That’s why there was so much opposition to him now. That’s why.

At a street corner, Bennett stopped, rocking on his heels, waiting for a red light to change. “History moves slowly,” he explained, “and is very cruel to those who try to move fast.”

Needless to say, it requires a resolute act of faith to charge people with murder, and imprison them for up to life, based on a scientific theory whose general acceptance will not arrive for a good 10 years. Listening to Thomas Bennett unveil himself in a Billings coffee shop, watching him exuberantly proclaim himself a medical pioneer on a Billings street corner, it became apparent just why Iowa’s furor burns so brightly and endures unresolved. It also became apparent why some might liken Bennett to Ben Franklin, while others might stir through his autopsy reports with astonished dismay.

Midnight Call to Prosecutor’s Home

In all the extended dispute, there has been no more telling exchange than a midnight call Bennett made to Rick Crowl’s home on the day he learned they were releasing the Lehmers. The former state medical examiner sounded furious.

“If you’re going to have a hearing,” Bennett demanded, “why not call me?”

Crowl cringed a little. He still had questions about the Lehmer case, and always would. Who knows, maybe Bennett was on the cutting edge. Maybe Bennett was right and everyone else wrong. Crowl couldn’t say; Crowl knew only that the medical examiners were the key to these cases. He had to rely on them; he had nothing else.

“Tom,” Crowl replied, “I didn’t have the guts to call you.”

“There was trauma. I saw trauma. . . . Jones is wrong. . . . Jones isn’t up on the reading. . . .”

“Nobody’s perfect, Tom. I respect you, but nobody’s perfect.”

“How do they explain the swollen brain if there was no subdural hemorrhage? That’s an injured brain; that’s trauma. . . .”

“Nobody’s perfect, Tom. . . . How can you know for sure? . . . Nobody’s perfect. . . .”

“Jones wasn’t at the autopsy. He didn’t have the microscopic slides. I don’t know what he saw. . . .”

“Tom, you want people sitting in jail for life who maybe didn’t do it?”

“My job is to stick my neck out for you, and here you are stabbing me in the back.”

“No, Tom.” For once, Crowl felt certain. “This is not a personal attack on you,” he said. “That’s not what this is about.”

*

The accounts in this story are drawn from numerous sources. They include:

* Transcript of the hearing of the Lehmers’ motion for a new trial

* Judge Timothy O’Grady’s court order releasing the Lehmers from prison

* Autopsy reports, autopsy photos, death certificates, hospital records, police reports and fire department logs

* Medical reports and depositions by consulting pathologists

* Public and private correspondence among forensic pathologists, pediatric specialists, lawyers and state regulators

* A review of medical and law journal articles concerning shaken baby syndrome

* Local news reports in Iowa

* Direct interviews with many of those involved

(BEGIN TEXT OF INFOBOX / INFOGRAPHIC)

The Signs of a Shaken Baby

Shaken baby syndrome is the medical term used to describe the violent shaking and resultant injuries sustained from shaking. Those injuries can include brain swelling, subdural hemorrhage, mental retardation, blindness and death. Knowledge about the syndrome continues to develop amid controversy. There is disagreement among physicians about the frequency of shaken baby syndrome, and the extent to which shaking alone can damage a baby.

THE EFFECTS OF SBS

For those who recognize SBS, violent shaking is considered especially dangerous because the neck muscles of infants and young children are undeveloped, and their brain tissue is exceptionally fragile. For many years, doctors failed to identify the cause of brain bleeding in small children who’d been shaken because the children usually didn’t exhibit any external signs of injury. Doctors saw only the results--convulsions, brain damage and death. The two most consistently telling markers for SBS, doctors eventually concluded, were subdural and retinal hemorrhages--bleeding in the brain and behind the eyes.

Sources: Des Moines Register, the Ultimate Visual Dictionary