Peter Ellis’ latest attempt to overturn his convictions for child sex abuse has once again highlighted how society handles such allegations. By Jonathon Harper
Any day now, the Supreme Court is likely to deliver its judgment on a landmark appeal. The Peter Ellis case is the first time in New Zealand that a court has been asked to make a judgment on someone after their death.
Just in case anyone needs reminding, in 1992 Ellis was charged with sexually abusing 20 young children who attended the Christchurch Civic Crèche, where he worked. Four of his female co-workers were also charged. The women were discharged pre-trial, but Ellis was eventually convicted of sexually abusing seven of the children. One later recanted.
Ellis spent almost two decades fighting to clear his name. Three guilty verdicts were quashed on appeal in 1993. However, a second appeal failed.
Ellis was released from prison in 2000 and died two years ago, aged 61, after battling advanced bladder cancer. The Supreme Court in September 2020 allowed the latest appeal – heard last October – to proceed posthumously, and although it hasn't yet published its reasons, the appeal is regarded as an important test of how tikanga Māori influences our justice system, and whether the system accepts that someone's mana can survive them.
It is not just Ellis' mana that is at stake, however. The mana of panicked parents, social workers, therapists and police has also, once again, been thrust into the spotlight.
The convictions relied on accounts of sexual abuse from pre-school children, and it now appears to be accepted by even the Crown that some of the children's evidence was "contaminated" after they were repeatedly and suggestively questioned.
If the children's original allegations were correct, then four female crèche workers watched as sharp sticks and burning paper were inserted into one child's anus; Ellis' mother hung children in cages attached to the ceiling of the centre; and there were ovens, a murder and secret passageways. The alleged abuse was originally thought to have involved 11 women, several men, and five teenagers, and included something known at the time as "satanic ritual abuse".
The officer in charge of the police investigation, Colin Eade, was later revealed to have suffered extreme stress, and has since admitted relationships with women involved in the case.
A Crown expert at the trial, child psychiatrist Dr Karen Zelas, warned parents before the trial about sexual abuse "signs". She then supervised the forensic interviewing, and appeared as an expert witness for the prosecution. During the original trial, she couldn't remember discussing the allegations against Ellis on a prime-time TV programme, and apparently forgot about a letter she sent to police warning them about serious problems with two children's evidence.
Author Lynley Hood, in her book A City Possessed, has argued that the trial was preceded by a moral panic. In the book, Hood strongly criticises several people's personal and professional conduct; no one has taken legal action over her claims.
Robert Bartholomew, a medical sociologist and author from New York, is now an honorary senior lecturer in the Department of Psychological Medicine at the University of Auckland. He has published two textbooks on public panics. He compares the Ellis case to the Salem witch trials in the 17th century, which resulted in 19 people being hanged for witchcraft in colonial Massachusetts.
Ellis, he suggests, was the quintessential "deviant" – quirky, homosexual, flamboyant, a misfit male in a predominantly female industry. "In Salem, the first people accused were deviants," he observes. "One girl had stopped going to church, another smoked a pipe and had some loose sexual morals."
In Salem, ordinary warts and skin blemishes were claimed to be evidence of a "witch" having consorted with the devil. In the Ellis trial, a new section (23G) of the Evidence Act allowed for common behaviours such as nightmares, bed-wetting and disobedience to be admitted as evidence "consistent with" sexual abuse.
Ellis, Bartholomew believes, was in the wrong place at the wrong time. "The context of this case, which occurred during the satanic ritual abuse panic, cannot be overestimated.
"The panic was driven by techniques such as regressive hypnosis, dream interpretation, guided imagery, leading questions, and implanting ideas in children's minds – each of which has been linked to the production of false memories. The more you know, the more convinced you become he was not guilty."
Tony Ward is a professor of clinical psychology at Victoria University of Wellington and a former director of the Kia Marama sex offenders' unit at Rolleston Prison. Although sex offenders are a diverse group, that isn't appreciated by the public, says Ward.
"To a mostly elderly, middle-class Pākehā jury in 1990s Christchurch, it is a fair bet that there was plenty of prejudice against homosexuals, including the mistaken belief that homosexuality is synonymous with child molestation," he says.
During the latest appeal, seven psychologists assessed the forensic interviews in Ellis' original trial and spoke at length about the unreliability of the children's accounts.
Deirdre Brown, of Victoria University of Wellington, is both a clinician and research scientist who trains specialist child interviewers. By today's standards, the interviews conducted back in 1992 would be described as "not yet competent", she told the court. "We do our most vulnerable children no favours by insisting that the interviews were not problematic. I would not like to see an interview conducted today in that way."
The experts all agreed that repeated interviews of children lead to less accurate reports. Usually in New Zealand, children are interviewed only once. Two children in the Ellis case were interviewed five times, and another two were interviewed six times.
The most prominent expert psychologist at the Supreme Court was the University of Otago's former vice-chancellor, Harlene Hayne. She is a leading researcher in memory development in both children and adults, and her original analysis of the interviews conducted with children in the Ellis case helped support calls for it to be reopened.
Now vice-chancellor at Curtin University in Perth, she told the court she believed the children's accounts of what happened may have been "contaminated" before they were interviewed.
The children were asked more than 20 "suggestive questions" about abuse-related matters prior to each allegation that led to a conviction, said Hayne. They were not given permission to say, "I don't know" or "I don't understand", and were not given permission to correct the interviewer if she made a mistake.
"As a memory expert, it is my strong opinion, that I believe is shared by every other expert in this case … that there is a clear risk their accounts were contaminated prior to the interviews. I am not aware of any technique or court procedure that would have been able to undo the impact of that contamination if it occurred."
Gail Goodman, of the University of California-Davis, a specialist in children's eyewitness testimony, appeared for the Crown. Goodman told the court that it was rare for children to make false claims. In fact, she suggested, "in taking thousands of children across various types of events, I would estimate less than 1 per cent of these children have made such false claims".
Nevertheless, she admitted there might have been "powerful sources of contamination" in the case. "It was my conclusion in the Ellis case that contamination … preclude[d] a confident evaluation of witness credibility … I don't feel I know what really happened."
Only two of the seven psychologists argued that the interviews were still potentially informative. One was Auckland-based clinical psychologist Fred Seymour, whose affidavit stated: "We do not wish to speculate about the impact of parents' cross-talk or other parents' interactions in this case. There is no evidence available to show that this was actually the case."
After questioning from Ellis' lawyer, Rob Harrison, Seymour accepted there was evidence of contamination.
In particular, Harrison referred to a letter sent by Karen Zelas to police before the original trial. It went missing before the trial, and counsel for the Crown could not explain how or why it was not disclosed.
The letter referred to "highly leading questioning" by the parents of at least two children. In one case, the parents subjected their child to "intensive interrogation" about "ritual abuse" between interviews, which were on consecutive days, she wrote. According to Zelas, in each subsequent interview the child disclosed the information elicited by his parents the previous night. However, she failed to tell the court of her concerns.
Clinical psychologist Suzanne Blackwell, of the University of Auckland, also appeared for the Crown and acknowledged "there is evidence of parental contamination". However, she questioned whether there was an "absolute causal" link. "I don't believe that we can know that," she argued.
The five judges also viewed videotaped excerpts from the children's evidential interviews, provided by Ellis' lawyer. Crown lawyer John Billington had counselled against it.
Chief Justice Helen Winkelmann and two other members of the bench, Joe Williams and Susan Glazebrook, expressed concern about what they had seen. The other two, Mark O'Regan and Terence Arnold, made little comment.
A ruling on the appeal is due soon. Regardless of the outcome, the Geoff Scott case may be another candidate for a review, because it is so very similar. Scott was a Wellington Hospital crèche worker who was sentenced to seven years in prison in 1994 for convictions relating to five children who attended the crèche in 1991 and 1992.
• Jonathon Harper is a freelance journalist and musician who once worked as a clinical psychologist, teacher and social worker.
