The Fells Acres tragedy

With a dramatic ruling, the truth behind allegations of mass sexual abuse

Freedom Watch by Harvey Silverglate

By now, everybody has his or her own idea of how and why three members of the Amirault family were convicted of child sex abuse and sentenced to long prison terms in the now-infamous Fells Acres Day School case. The most recent view -- and the most authoritative to date -- was expressed last Friday in a lengthy opinion by Middlesex Superior Court judge Isaac Borenstein. In it, he granted a new trial to Cheryl Amirault LeFave on the grounds that the children's testimony had been fatally compromised by suggestive and even coercive interview techniques. Borenstein's decision will now be reviewed by the Supreme Judicial Court.

But the decision is far more than just another development in a case that, after 14 years, seems unwilling to end. It is, in all likelihood, the turning point. Appeals courts are generally reluctant to overturn factual findings (as opposed to procedural or legal rulings) by a lower court. And Borenstein's decision, based on new scientific evidence of how easily children can be induced to testify falsely, is a searching -- and damning -- finding of facts. In this sense, his decision is highly unusual: it considers, directly, the central question of guilt versus innocence.

His carefully documented 139-page opinion is also a fascinating look inside one of the more bizarre phenomena of the last decade and a half. Across the country in the '80s, the front pages exploded with horrifying allegations of mass, ritualistic sexual abuse of children at day care centers. There was the Grant Snowden case in Miami, the McMartin Preschool trial in California, the Little Rascals case in North Carolina, the Kelly Michaels allegations in New Jersey, and others. In each case, emotions have run high among the defendants, the children, and the parents. Yet, over time, these prosecutions have been collapsing. The Borenstein decision helps us understand how cases like these come to be -- and why, finally, they are crumbling.

In the Amirault case, the appeals process has dragged on for years, focusing largely on such technical and legal issues as whether the defendants were denied a fair trial because the child witnesses did not have to face them. The SJC has steadfastly refused to overturn the decisions of the juries that convicted first Gerald "Tooky" Amirault and then his sister, Cheryl, and his mother, Violet, who has since died of cancer.

This appeal, though, was prompted by new evidence pointing to the suggestibility of child witnesses. Defense lawyers presented an affidavit from Maggie Bruck, a well-known and highly respected doctor of experimental psychology who is a faculty member in the departments of psychology and pediatrics at McGill University in Montreal. Bruck had long been horrified by what she considered to be false accusations presented to juries in the day-care sex-abuse cases across the US, and in many of those cases, she had participated in successful efforts to undo the convictions. (Full disclosure: at one point I gave Bruck some legal advice.)

According to research by Bruck and her colleagues, the theory on which the prosecutors had operated was simply wrong. The prosecutors' assumption was that children who had suffered abuse would be reluctant to talk and needed to be made comfortable in order to do so. In violation of common sense, prosecution "experts" insisted that the child's initial "no" might really mean "yes," and that only through suggestive questioning would the truth emerge. In fact, it turns out that children will testify to real abuse -- but if the questioners push hard enough, children will also freely invent abuse that never happened.

Even at the trials, the leading way the Fells Acres children had been questioned was controversial. But Borenstein found that the Amirault case was a textbook example of no being taken to mean yes until, finally, the children gave in. As the judge outlined:

Social workers and police officers engaged in "repetitive questioning about a particular topic or event, regardless of a child's response."

Interrogators asked the children specific rather than open-ended questions, thus "providing the answer in the question."

They used peer pressure to elicit disclosures, suggesting to children that they should tell investigators the same things their friends and fellow students were saying.

Interrogators "reinforc[ed] statements which confirm[ed] the interviewer's beliefs."

When children made claims so bizarre and incredible that they threatened to discredit everything else the child said, those statements were ignored. Indeed, as time went on, the children were testifying that they had been visited by talking robots, that they'd seen animals tortured, that they had been tied naked to a tree (this on a public street in Malden).

Judge Borenstein's finding also pierces some of the other myths that have held the case together for so long. Perhaps the most important has been the notion that though the "professionals" were perhaps overzealous, their results were credible because the kids made their first disclosures to their parents.

That turns out to be untrue. What really happened, according to Borenstein, was this: Soon after word spread of an allegation made by one child, authorities in Malden closed down Fells Acres and "the community was thrown into a state of panic." In September 1984, Malden police convened a public meeting, at which "law enforcement officials made the irreversible and critical mistake of delegating the delicate and important task of interviewing to parents." The parents, Borenstein wrote, "were in essence deputized" and told to question their children about what would become the substance of police and prosecutors' allegations, including talk of abuse taking place in a "magic room" at the hands of someone dressed as a clown. The parents, no doubt terrified -- and armed with the police department's preconceptions as to what happened -- were sent home to question their children both aggressively and suggestively. The results were a fiction.

The original disclosure -- the story that touched off the panic -- itself has suspicious roots. Shortly after the child enrolled in Fells Acres, he started having difficulty sleeping, and he cried a lot. He was clearly disturbed. Yet these symptoms, Maggie Bruck points out, were not unusual for a child beginning school. (He also had just lost a pet, moved to a new neighborhood, and developed hearing problems.) Still, the mother immediately suspected sexual abuse and began persistently questioning her son about it. He denied it. She told him that his uncle had been abused as a child. He denied it. Even when the uncle questioned him, the child at first denied any abuse. Five months after the first questions, he finally gave in.

Those who believe in the prosecution will continue to insist that there is other behavioral and physical evidence to back up the stories. Borenstein, however, was quite clear on this point. There is, he wrote after extensive review of the case, simply "no independent evidence . . . to support [the child witnesses'] claims."

The physical evidence was flimsy and inconclusive. And where prosecutors cited evidence of suspiciously sexualized behavior on the children's part -- excessive masturbation, preoccupation with their own and others' genitalia, purportedly inappropriate sexual play -- Borenstein thought such behavior could either fall within the bounds of normality or be explained by other factors (not least of which was the investigators' use of anatomically correct dolls in the questioning, which gave the children an instant education in sexuality).

"The newly discovered evidence," the judge concluded, "shows beyond a reasonable doubt that what was done in this case was to risk convicting innocent people for acts that probably did not occur."

But weren't the facts for the jury to decide? The jury was, as the conviction's defenders often point out, aware of the interview techniques at the time. What they didn't know, though, was that one of the prosecution's central assumptions -- that abused children must be "led" to testify -- turns out to be, as a matter of scientific fact, completely false. The jury was given reams of horrifying testimony without the proper context to interpret it.

Indeed, then-Supreme Judicial Court chief justice Paul Liacos, in a dissenting SJC opinion, argued that the conviction should be thrown out simply because Judge John Paul Sullivan, who presided over the original trial of Violet Amirault and Cheryl LeFave, had allowed the jury to hear "highly inflammatory and prejudicial" testimony from a prosecution witness, US postal inspector John Dunn. Dunn, who was well known in Massachusetts law-enforcement circles for his almost obsessive interest in and rather idiosyncratic views on the traffic in child pornography, told the jury that the sex acts described by the children's testimony -- allegedly captured in Polaroid photographs -- were precisely the kinds of acts that one finds in pornographic photographs of children between three and six years old. He further informed the jury, on the basis of his "expertise" in child porn, that there was a large underground market for precisely the types of photographs the children had described. Yet no photographs were found in Fells Acres.

Edmund M. Powers, a juror in Gerald Amirault's 1986 trial, told the Boston Herald last week that he now thinks "the jury was duped -- misled."

He is not alone. One of the more interesting side dramas in the case has been to watch the judges involved become more and more dramatically disturbed by it. Liacos's early dissent in the SJC now appears prescient. After Violet Amirault and Cheryl LeFave were denied parole, Judge Sullivan tried to reduce the prison terms that he had himself had imposed. (The SJC reinstated the sentence.) After Sullivan retired, he was replaced by Judge Robert Barton, who ordered a retrial. When this order was overturned by the SJC, he angrily denounced the decision of the state's highest court -- a very unusual move given the conservative mores of the bench -- and recused himself from the case. Sullivan himself then declared that the SJC should have upheld Barton's order -- admitting to the world that he did not believe in the integrity of his own trial. Finally, Judge Borenstein took over from Barton. The result: last week's devastating indictment of the case's entire foundation and conclusion.

Still, the prosecutors have refused to fundamentally rethink the issue. When the Amirault cases were brought and tried, Scott Harshbarger (now attorney general) was Middlesex County's district attorney, and Thomas Reilly (now the DA) was his first assistant. Their reaction to Borenstein's ruling? Reilly and the chief of his child abuse unit, Lynn Rooney, filed an immediate appeal. Harshbarger issued another routine statement defending the work of his office. Both men are running for higher office -- Reilly for attorney general, Harshbarger for governor -- and both men owe the public more honest answers to the questions the case raises.

It now seems that, in many ways, the entire episode was a byproduct of the investigation itself. Those who were doing the looking knew what they were looking for, and they wouldn't rest until they found it. And the children, eager to please, told them what they wanted to hear.

Borenstein, for example, points to a psychologist's report from Children's Hospital, in which one of the children reported that he saw "balls of light coming at him to kill him." In what Borenstein calls "an obvious effort to get him to say something about photographs," the therapist "asked him if the balls of light were like flashbulb lights." The child, predictably, agreed with the therapist's proposition.

To Borenstein, it seems clear that both law-enforcement officials and Susan Kelley, the pediatric nurse who interviewed the child witnesses in the trial of Cheryl and Violet, "had decided from the start that the Amiraults had committed these crimes." As a result, he notes, investigators "chose to ignore" evidence and testimony that made it clear that the children's accusations were untrue.

Prosecutors have long claimed that the children's testimony was "chilling in its consistency." Now we know why. "In fact," writes Borenstein, "the only thing chilling in its consistency in this case is the manner in which the children were interviewed and pressured, all in the face of denials."

Writing on the front page of the Boston Globe, Eileen McNamara argued that these children have been forgotten. "Beyond describing their testimony of rape and sexual assault as `forever tainted' by the leading questions of `overzealous' investigators, Judge Isaac Borenstein does not have much to say about those children," McNamara wrote indignantly.

She could not have been more wrong. Judge Borenstein said something very important about them: he pointed out that they were "victimized" by the "inappropriate, suggestive interviews."

And that is the true tragedy of Fells Acres that McNamara and so many others do not appreciate. If there was no abuse -- and all the evidence supports this view -- then we must now face the horrible probability that a mass of concerned (and perhaps panicked) adults convinced a group of youngsters they had been victimized. That the state sent three innocent people to jail -- and stole the innocence of trusting children.

For a different view of the Fells Acres case, see "Suffer the Children," by Dan Kennedy, News, April 28, 1995.