The Christchurch Civic Crèche Case

News Reports Index


North and South
August 1996, Pages 54 - 69

Second Thoughts on the Christchurch Civic Crèche case:
Has justice failed Peter Ellis?
By David McLoughlin

Former day care worker Peter Ellis is now three years into the 10-year jail sentence handed down after he was found guilty of 16 counts of sexually abusing young children at the Christchurch Civic Creche. As time passes, the case against him looks less and less credible.

Ellis is supposed to have subjected large numbers of children to horrific abuse over five years. According to the prosecution he sexually violated children, defecated and urinated on them, stuck sticks and needles into their anuses and penises and forced them to stand naked while women creche workers danced around them.

All this and much more supposedly happened without a single child complaining or showing distress, without a single parent or other adult noticing anything wrong, without a shred of medical or forensic evidence to support the charges of such vile abuse.

Since his trial, a growing body of evidence has come to light suggesting the justice system failed Peter Ellis. That parental hysteria created many of the allegations of abuse. That seriously flawed methods were used to extract "evidence” from very young children. That the police lost all sense of perspective while investigating the case. That mistaken rulings by the trial judge, upheld by the Court of Appeal, cost Ellis a fair trial. That the evidence overall was so suspect, a jury in possession of all the facts could not have found Ellis guilty beyond reasonable doubt.

This is the full story of the Civic Creche case. It's at least as disturbing as the highly selective one the jury heard.

David McLoughlin is a North & South senior writer. He has been researching the Christchurch Civic Creche affair for 18 months. Last year he helped produce a TVNZ Assignment documentary on the case.


On Saturday, June 5 1993, a jury in the High Court at Christchurch returned 16 guilty verdicts against Peter Hugh McGregor Ellis, then aged 35. The verdicts related to alleged offences against seven small children and followed the six-week trial of Ellis on some of the most sordid charges laid in this country.

The jury of three men and nine women decided that Ellis, formerly a worker at the central-city Civic Childcare Centre (the "Civic Creche"), had urinated on two children, made one masturbate him, put his penis in the mouths of three of them, engaged in indecent touching of three and put his penis or an unknown associate's penis against the vagina or anus of three.

The offences allegedly happened at the creche, at an unknown address and at Ellis's home while he was babysitting a creche child.

Ellis steadfastly protested his innocence but he didn't impress either the jury or the trial judge, Justice Williamson. The jury heard evidence of Ellis's flamboyant personality, his heavy drinking and his delight in shocking his women co-workers by talking about practices like "golden showers", a term that means urinating on one's partner for sexual pleasure. While summing up the case for the jury, the judge called such practices "kinky" and emphasised the prosecution's claim they were similar to what Ellis was charged with doing to the children.

"Unlike almost all of those who have pubIicly feasted off this case, the jury actually saw and heard each of the children," Williamson told Ellis 17 days later when sentencing him to 10 years' jail. "They also heard your own evidence and that of the other former Christchurch Civic Creche workers. The jury disbelieved you. They believed the children and I agree with that assessment.'

Believe the children.
It's the key to this entire case, the reason Ellis might have been wrongly convicted of the horrendous crimes with which he was charged.

The required standard of proof in New Zealand criminal cases is that of “beyond reasonable doubt" . The onus is on the prosecution to prove guilt to that exacting standard. Mere suspicion that someone is guilty is not enough. The fact the police, the prosecutor, the judiciary or the jury might have found Peter Ellis an unsavoury character should not ordinarily have been enough to find him guilty without substantive evidence. As will unfold, there was none in this case.

But in charges of sexual abuse the law was changed 11 ears ago to allow juries to convict alleged child molesters on the uncorroborated

word of young children, even those barely out of nappies and barely able to express themselves. Their "allegations" might be nothing more than playing with bizarre looking "anatomically correct dolls" that have penises, vaginas and anuses. What the children say or do is interpreted for juries by psychiatrists and the Social Welfare staff who conduct videotaped "evidential interviews" with children suspected of being abused. Since 1989, these videotapes have been admitted in court as the evidence-in-chief of child complainants.

The philosophy behind the law change was laudable. It was designed to convict child molesters in cases lacking physical evidence or the absence of adult witnesses to heinous acts which, by their very nature, occur in private.

As will also unfold in this story, however, the law change opened the prospect of miscarriages of justice.

The child psychiatrists, psychologists and social workers who promoted the change were driven by an American theory that a child claiming to have been sexually abused must always be telling the truth, because such horrible acts are outside the experience of anyone who hasn't suffered them. With that theory goes the belief that a child who denies being abused, or who retracts a previous allegation, has still been abused but is in a psychological state called "denial". These experts even argue that parents who claim a child hasn't been abused are also "in denial".

The "children never lie" theory was used to support allegations of sexual abuse that broke out like plague at childcare centres in California and elsewhere from the early 1980s. Scores of American men and women childcare workers were jailed after highly publicised court cases eerily similar to the Civic Creche affair. Identical cases followed in Britain. As with Ellis, the American and British childcare workers found it almost impossible to defend themselves in the face of experts who claimed abuse had occurred whether the child alleged it or not and whether corroborating evidence existed or not.

More recent research on the credibility of child witnesses, however, has cast serious doubt on the convictions of many of those overseas childcare workers. Many have been freed from jail and pardoned or acquitted on appeal. But that research, at least until now, has been of no help to Peter Ellis, who to this day languishes in Rolleston Prison, south of Christchurch, still protesting his innocence.

Picture the Civic Creche case as a judicial iceberg. The 16 counts involving seven children that Ellis was found guilty of were just a third of the 48 counts involving 21 children he originally faced. In turn, those original charges formed but the tip of a far greater iceberg that seethed beneath the surface of the case, enveloping scores of Christchurch families in a trauma from which many will probably never fully recover.

At least some of the large team of detectives who investigated the creche case believed the Civic was a den of iniquity supplying children on demand for paedophiles and child pornography. They believed Ellis was merely the worst of many offenders involved.

In addition to Ellis, the police arrested four women creche staff during the investigation. Three, the supervisor Gaye Davidson and staff members Marie Keys and Jan Buckingham, each faced four indecency charges. The fourth, Debbie Gillespie, faced three charges: Like Ellis, they denied the charges against them at an early stage. They were discharged just before his trial.

At least 127 children were interviewed. Some detectives believed that dozens, perhaps up to 80, had been abused. They also believed far more adults than Ellis and the four women were involved. Interviews with the children carried on to the start of the depositions hearing (the district-court proceedings which decides if a prima facie case exists) in November 1992.

A leading member of the police inquiry team told me that more women creche staff and even Ellis's mother were facing arrest when a halt was called to allow the case to go to depositions. The same officer left me in no doubt that he and other detectives even now do not consider the case is closed.

The women were arrested late in the inquiry, six months after the first charge was laid against Ellis and only a month before the depositions started. Their arrest followed the police investigators asking themselves the obvious: How could abuse on the scale alleged have occurred without anyone noticing, unless Ellis's co-workers were either party to it or covering for him?

But even if the women were participants, how to explain why none of the parents noticed anything amiss? The Civic Creche was a Mecca for New Age liberal parents. Many were articulate middle-class professionals, some even social workers and sexual-abuse therapists highly likely to have alerted their youngsters to the danger of sexual abuse. For five years, none so much as suspected it.

Many of the alleged offences supposedly occurred in the toilets of the busy and popular creche. Others allegedly occurred at other addresses around Christchurch, during the walks around town the children were taken on by Ellis and other staff. Despite Ellis being accused of sodomising children, forcing children to eat his faeces, urinating on them, suspending them in cages, putting them in ovens or taking them on terrifying trips of abuse through tunnels, ceilings and trapdoors, none showed any sign of those ordeals. To the contrary, there was considerable evidence that the children loved not just the Civic Creche but particularly the outrageous Ellis, the darling of many staff, parents and children alike.

To explain this further incongruity, the police, supported by Christchurch psychiatrist Karen Zelas and the Social Welfare staff who interviewed the children, came up with the theory - similar to that used in many American creche cases - that Ellis so terrorised all those children, by threatening, for example, to kill their parents if they revealed the abuse, that the children remained silent until the evidence was coaxed out of them during videotaped interviews conducted by the Social Welfare staff.

Many of the parents and at least some of the police team also believed that the Civic Creche affair had all the hallmarks of the "ritual abuse" alleged in many of the American cases.  Like the Civic, many American cases involved allegations of children being supplied for organised paedophilia and pornography rings as well as for ritualistic abuse and Satanic rites similar to the allegation that Ellis and his Co-accused forced children to stand naked inside a circle of dancing adults.

 Allegations of ritual abuse swayed many juries in the United States in the 1980s but the Christchurch crown prosecutor, Brent Stanaway, had no intention of putting such bizarre claims before a conservative Christchurch jury. From his arrival on the case halfway through the depositions, to the end of the trial, Stanaway fought to keep the case narrowly confined to simpler accusations of abuse, based on the video-taped testimony of the most credible-sounding children.

The Crown's determination to spare the jury the bizarre allegations didn’t stop the defence from trying to raise them in defence. Defence counsel Rob Harrison sought to show the jury the tapes in which children who were the subject of more credible charges went on in subsequent tapes to describe such incredible events as being suspended in cages from rafters and having lighted paper stuck in their backsides.

Harrison wanted the jury to see tapes because he believed they would cast reasonable doubt on the more credible testimony. He regarded this as the crux of the defence case. But Justice Williamson ruled that such crucial tapes were not relevant to the charges against Ellis and he placed such strict limitations on playing them that Harrison believed the defence was mortally handicapped.

Court of Appeal judge Sir Maurice Casey, subsequently upholding Williamson’s ruling, said: “He was clearly right in seeking to prevent the trial becoming enmeshed in all the collateral and peripheral matters covered in the tapes not relied on by the Crown and about exposing the jury to the playing of many hours of irrelevant material, thereby distracting them from consideration of the real issues.”

This was a case it seemed the defence could not win whatever argument it put up. Ellis was confronted with the view of Karen Zelas, a veteran expert witness in numerous abuse cases, that virtually any behaviour exhibited by a child was consistent with sexual abuse.

The Court of Appeal adopted this ethos when confronted with a retraction by Child S, the Crown's prime witness. S was the oldest and most credible of all the children. Her evidence was the first heard by the jury, which convicted Ellis of all three charges involving her. A year later, during the appeal process, she told her parents she'd lied about Ellis. Nothing had happened. Picking up the theory that a retraction is merely “denial”, Sir Maurice Casey wrote, "We are by no means satisfied she did lie at the interviews, although she may now genuinely believe she did." He discharged Ellis on those three charges alone, only reluctantly, on the grounds it would  be unsafe to let those convictions stand, given her retraction.

Justice Casey retired from the Bench 10 months after his verdict on Ellis's appeal. In his retirement speech, he said the saddest aspect of his 21-year judicial career had been watching a massive increase in cases of child sexual abuse.

Christchurch in late-1991 was a city in waiting for a major child-abuse scandal. Peter Ellis, as many of his supporters have observed since, was inevitably going to be part of it, an accident waiting to happen because of his gay flamboyance.

In the few years immediately preceeding the first allegations of abuse at the Civic, Christchurch was home to several abuse scares - the so called Ward 24 case regarding highly suspect interviews of children at Christchurch Hospital; The mistaken mass-diagnosis of children at the Glenelg Children’s Health Camp; and the Spence family affair which concerned highly questionable claims that a father abused his children. Some of the professionals involved in the Civic case, including psychiatrist Karen Zelas, had a role in some of those too. And it was well known round town that various Christchurch police officers were hunting for a near-mythical pornography-paedophile ring alleged to involve judges, Freemasons and prominent businessmen, though it was never found.

For two months from early September 1991, there was more or less continuous publicity of sexual abuse and ritual abuse of children in the local press or in national media many Cantabrians would have seen. Much of it was prompted by a family-violence conference held in the city in early September 1991, attended by 250 people, including some from Australia and the United States.

One conference event in particular attracted considerable attention: a workshop on satanic ritual abuse run by Wellington abuse counsellor Anne-Marie Stapp. She was prominent in an organisation called the Ritual Action Group, another member of which was a Wellington policeman, Laurie Gabites.

Stapp told the Christchurch daily The Press, in an article published on September 4, that New Zealand was fast approaching the level of ritual-abuse awareness found in the US. She said ritual abuse took many forms but nearly always involved the cult of Satanisrn, with the victims being used in worship rituals where they were abused in a humiliating and sadistic way. Stapp's colleague Jocelyn Frances claimed cult members in New Zealand came from groups developed from fundamentalist Christian churches, the Freemasons and a sex ring that operated among businessmen. Stapp claimed to have interviewed three ritual-abuse survivors and said 20 more had sought help.

A variety of print and broadcast media stories on sexual and ritual abuse of children followed, including a big Sunday News report on November 3, quoting policeman Laurie Gabities, who alleged Satanism was rampant in New Zealand and linked to child pornography.

Seventeen days later, the first complaint about Peter Ellis was made to creche supervisor Gaye Davidson.

New Zealand’s first American-style daycare abuse scandal probably couldn't have happened anywhere but Christchurch, and even there it couldn’t have happened anywhere but the Civic Creche. No other city or town in the country had experienced the garden city's run of well-publicised sexual-abuse scares and the Civic Creche was one of a kind, the trendiest creche in Christchurch, if not in New Zealand.

The Civic, owned by the Christchurch City Council, was located in the Cranmer Centre, the former girls' high school complex adjacent to Cranmer Square in the central city. The creche wasn't the centre's only occupant. Marriage guidance services, rape counsellors and the like occupied offices all over the rambling complex. Then, as now, the notice boards along its many corridors carried messages and posters promoting a variety of feminist and liberal causes.

Most of the Civic's 80 or so full-time and part-time children came by car from all over the metropolitan area. Their predominantly middle-class parents shared in common the full range of liberal values from anti-racism to gender equality. Many had taken part in the Springbok tour protests of 1981. Many were teachers, journalists, social workers and the like.

"There were no plumbers or truck drivers there," says former creche worker Stephanie Hauiti, who came to the Civic after teaching at a kohanga reo. "I was accepted because I was taha Maori, but I'm from the wrong side of town and I was nothing like the parents. My kohanga parents didn't agonise over the correct gender for a nursery rhyme character, which was important to the Civic parents."

In November 1991, the very month the first allegation of abuse was made, the Education Review Office issued a report warmly praising the Civic.

By most accounts, Ellis was extremely popular with children and parents alike. His over-the-top, mincing campness, his habit of wearing make-up and his cutting sense of humour were pluses for many parents. It's said the chance to be exposed to a gay man was one reason some parents sent their preschoolers to the Civic. "He was the kind of male role model many parents wanted their children to look up to " says another former staff member.

On the other hand, some children disliked Ellis for his boisterous play which sometimes turned into inappropriately rough play, earning him a number of formal and informal warnings over the years. Supervisor Gaye Davidson had also had to reprimand him for drinking alcohol in his lunch hour. A smoker, he was in the habit of hiding in a toilet cubicle for a cigarette.

Ellis is the eldest of four children whose parents separated when he was nine. He picked tobacco in Motueka after leaving school, then went overseas for two years, on his return holding a number of jobs, some of them responsible positions which he apparently found stressful and set him drinking heavily. He had been unemployed for some time when he arrived at the Civic in September 1986 on a two week community service order for benefit fraud, his only prior conviction. He liked the creche and staff, children and parents liked him, so he stayed, completing a childcare certificate on the job between 1987 and 1989.

"Peter was an anarchist," says Mary Cox whose three children attended the Civic over a 10-year period. “He was the one who stirred things up. ,

During the depositions and trial, there was extensive unchallenged evidence that Ellis was boisterous in his play with the children. Some enjoyed this, others apparently did not. He frequently took groups of children for walks around the city, often to the nearby Botanic Gardens, once by his own admission to a house he'd lived at in Hereford Street on the other side of the city centre, where he'd shown children his prize collection of rabbits and other animals.

There was also unchallenged evidence that Ellis liked to shock his fellow workers, especially the less mature, younger ones, with talk of sexual practices. As well as the "golden showers" already mentioned, Ellis himself told the High Court he'd spoken of a practice he'd read about in a book whereby two males would insert each end of a straw in the other's uretha.

However damagingly all this was painted in the High Court, nobody at the Civic Creche thought for a moment that Ellis or any other staff member was a danger to the children. Nor do those parents and staff members who to this day believe in his innocence accept that he or his colleagues had the opportunity to abuse children.

"I used to drop in at any hour of the day," says dentist Bernie Wynn-Williams, whose surgery is less than five minutes' walk from the Civic. His three children attended the creche over eight years. "I didn't just go in and out to see how my child was getting on. If I had time between appointments, I'd stay for half an hour. They never knew when I was coming and you could get into the creche through any of several entrances.

"I was put off when I first saw Peter Wynn-Williams continues. "I thought he was a bit strange and that I'd better watch him. I was never totally at ease with him, because of the things he said to shock people. But nobody there had the opportunity to do anything to the children. A lot of the abuse supposedly happened in the toilets, but I never once saw the toilet doors closed.

"My impression of Peter was that he was blissfully unaware of the danger his tongue would one day get him into."

On the morning of November 20 1991, a creche mother phoned supervisor Gaye Davidson and alleged her preschool son had accused Peter Ellis of sexual abuse, specifically stating the child had said to her while in a bath: "I don't like Peter's black penis." The mother told Davidson the boy had made a similar comment to his grandmother.

The identity of the complainant children in this sad case may not, quite rightly, be revealed. Their parents' identities are also suppressed by law, to protect the children. For the purposes of this article, therefore, the children identified will be usually identified by the initial of their first name.  Their parents will be similarly identified by their first initial. For example, Child A or Mother B.

However, several children and several parents had the same first names. Three mothers central to this story had identical first names beginning with the letter S. They will thus be identified here as mothers S1, S2 and S3.

The parent who phoned Gaye Davidson that fateful November day was Mother S1, and her son, for the purposes of this story, is Child F. Mother S1's profession was as a counsellor specialising in sexual abuse. She told Davidson that F would not return while Ellis was there. Davidson was legitimately concerned. After she discussed it further with the boy's mother and a city council official, Ellis was put on temporary leave pending an investigation.

Five days after her complaint to David son, S1 contacted Christchurch detective Colin Eade, who asked to be kept informed. The same day, her son F underwent the first of the many scores of video taped interviews of children in this case. F was adamant he hadn't been abused and maintained that stance determinedly in subsequent interviews until (and including) his final one in September 1992. No charges were ever laid in respect of him.

Meanwhile, rumours of Mother S1's allegations and Ellis's suspension spread through the creche community. Journalists were sniffing about. Aware of this, the creche management committee met on November 28 1991. Present were Eade and Sue Sidey, the Social Welfare psychologist who conducted most of the 127 child interviews. It was decided to call a meeting of all current parents to explain what was happening. Mother S2 was delegated to organise it.

The parents' meeting took place on the evening of December 2.  That morning, an article in The Press had revealed Ellis's suspension. Malcolm Cox, Mary's husband, describes the meeting that night thus: "We were sitting there when the door opened and [Mother S1] came in and burst into tears. [Mother C] raced over and comforted her, sat her down.  A clutch of them hovered around her. You knew who the victim was supposed to be. The city manager, John Gray said there was concern, but Colin Eade said he hadn’t seen anything that convinced him there was abuse.

Malcolm Cox says the meeting wasn't given much detail: "Someone asked, `Are we talking about tickling or penetration?'  Sue Sidey from Social Welfare was asked about symptoms and said we should look for bedwetting, nightmares and tantrums. Her statement was greeted with a collective drawing in of breath. What children don't have some of those symptoms?"

Sidey told the parents they should not directly question the children about abuse. Parents who suspected a child had been abused could bring the child to her to be interviewed.

From that point, Cox says, creche parents divided into the two camps they remain in till today - those who believe abuse occurred, those who believe nothing happened: "Next day, a kid turned up at the creche and asked where the witches were. His dad had come home from the meeting and said there was a witch hunt going on. "

Following the meeting, some parents formed a support group for Mother S1. Two of them, mothers S2 and C, promptly sent their sons to Sidey for interviews. C's son said Peter tickled him and stole his food, but despite direct questions, neither child told Sidey they'd been abused. On December 20, Colin Eade wrote to the creche management committee saying he'd found no evidence of abuse but added he didn't believe Ellis was an appropriate person to be involved in childcare. He apparently reached his conclusions without having interviewed Ellis, who remained suspended.

Mother S1 didn't accept Eade's conviction that her son hadn't been abused. Parents who supported her continued to send their children to Sue Sidey for videotaped interviews. As 1991 became 1992, S1 approached an increasing number of creche parents to warn them Ellis might have abused their children. She became a virtual telephone-exchange of information, real and imagined, about the Civic and what supposedly went on there. Questioned during the depositions hearing, S1 agreed that Sidey had warned parents not to share information, because it could taint the evidence, but S1 said she believed such "secrecy" allowed abuse to continue.

S1 withdrew F from the Civic and put him in another Christchurch day care. Soon afterwards she alleged a male worker there had also abused him. An investigation of this charge too found it groundless, but this didn't stop the police from taking seriously her many subsequent claims, increasingly absurd though they became.

On January 30 1992, mother S2, the organiser of the December parents' meeting in took her daughter, R, who’d not even attended the Civic, to Social Welfare to be interviewed. R was diagnosed as having been abused by Ellis; the allegation was indecent touching. It supposedly happened in the few moments she was on the creche premises while her mother picked up her son.

The police immediately cranked up the investigation. Several other children whose parents had been in close contact with S1 were interviewed in February and March and what they told Sidey was also diagnosed as evidence of abuse.

The police and Social Welfare quickly decided they were dealing with an extremely serious case of multiple child abuse. They decided to call a meeting of current and past creche parents to alert them to the fear many more children than those already interviewed might have been abused. Invitations were mailed, asking parents to attend a meeting on March 31 at Knox Hall on Bealey Avenue, several blocks north of the Cranmer Centre.

The Knox Hall meeting took place in an atmosphere of alarm and hysteria among creche parents. The day before, Detective Colin Eade had arrested Peter Ellis on a charge of indecently assaulting R. Ellis's first appearance in court, the day of the meeting, fed a media frenzy that had begun a week before when word that the meeting had been called spread through the city.

An article in The Press on March 23 was headlined Parents In Terror Of Abuse Discovery and said up to 200 children might be involved in an investigation at the Civic. Holmes the same night claimed police had received 20 complaints of abuse and believed at least 50 children were involved. Karen Zelas appeared on Holmes to tell parents not to question their children but send them for specialist interviews. She also talked about behavioural symptoms that suggested sexual abuse.

Why did the police wait until the very eve of the Knox Hall meeting to arrest Ellis? They'd had R's videotaped testimony for two months. Without doubt, the timing of the arrest inflamed an already heated atmosphere. It seemed designed to cause Ellis maximum prejudice. And why choose the allegations by R, when they'd had stronger claims from other children for at least a month?

Eade was asked these questions in court almost 18 months later. He conceded he'd interviewed Ellis when he did because the Knox Hall meeting was the next day. He'd hoped to question him about other children but ran out of time. Asked why Ellis hadn't been interviewed earlier, given the police had the videotaped testimony, Eade replied: "The decision was made above my level for the interview to be done at that time."

It's interesting to note that R's unsupported allegation didn't proceed. The charge involving her was dropped without explanation even before the depositions hearing.

Knox Hall was crammed with angry, frightened and bewildered parents on the night of March 31. All the seats were taken and parents stood in the aisles and along the walls to hear the police, Sue Sidey, Karen Zelas and others talk about the investigation and alert them to possible signs of abuse. Parents were told that all children who'd attended the Civic during Ellis’s five years there should be interviewed.

They were given a pamphlet headed "What To Do When A Child Tells Of Abuse" (the word "when" was crossed out and replaced with a handwritten "if '). It outlined five basic rules: believe what children say; say you're glad they told you; say you're sorry it happened; tell them it's not their fault; let them know you'll help.

Parents were given phone numbers to call to arrange videotaped interviews, offered counselling for themselves, their children, even relatives, and given Accident Compensation forms to claim the $10,000 which at the time was automatically paid to anyone claiming to have been sexually abused.

"I'm not saying the thought of ACC payments motivated anyone," says Malcolm Cox, "but I'm sure it had at least an incidental effect. We were even visited at home by a council social worker with an ACC claims form who said we had to get in quick to claim the money because lump sums were being abolished. I said we had no fears our children had been abused, and she told us we should still claim now and change our minds later if we wanted to."

Acc paid more than $500,000 to around 40 parents of Civic children. Many payments were the standard $10,000, but in cases where Ellis faced multiple charges relating to a single child, some parents claimed for each alleged incident of abuse. One child's parents claimed five payments another four. ACC didn't require a conviction before paying out. It paid up without so much as charges being laid in respect of some allegations. The police even wrote letters to ACC supporting compensation claims.

The Civic inquiry quickly became one of the biggest police investigations Christchurch had experienced.  A team of detectives was appointed, led by Detective Sergeant Bob Hardie, under the overall responsibility of Chief Inspector Brian Pearce. Sue Sidey was unable to cope with interviewing scores of children by herself, so several other Social Welfare staff were assigned to help her.

The production-line interview process soon led to further charges against Ellis, who was on bail, living on the dole. Four new charges were laid on April 14 1992, five on June 6, one on June 25, six on June 30 and a massive 14 on September 25, a week before Gaye Davidson, Jan Buckingham and Marie Keys were arrested. Before laying each set of charges, a detective interviewed Ellis, played him the relevant videotapes and sought his comment. He denied each allegation.

The creche, meanwhile, remained open. For months after Knox Hall, Ellis was the only suspect. But gradually parents began withdrawing their children.

Apart from the investigation team, few people knew the exact nature of what Ellis had allegedly done. The new charges read out each time Ellis appeared in court were vague counts of indecent assault. This didn't stop the spread of rumours, but even the rumours were nowhere near as extreme as what finally surfaced when the depositions hearing began in November.

"The story we heard was that Ellis had abused a kid while baby-sitting," says Malcolm Cox. "When we finally discovered what he was really supposed to have done, it was so bizarre it was laughable. It would never have got that far if more people had known earlier on what was being claimed.”

The allegations and charges became more serious, or, depending on your viewpoint, more off the wall, as the interviewing went on. They started with claims Ellis had indecently touched children. Some of these indeed allegedly happened during baby-sitting. They moved on to acts of sodomy, oral sex, penetration with fingers and sticks and then vicious vaginal rape. Next, Ellis was said to have urinated on children and made them drink his urine or eat his faeces. Towards the end of the interview process, he'd supposedly taken children through tunnels, across rooftops and inside ceilings on safaris of abuse, as if the many other occupants of the Cranmer Centre wouldn't have noticed.

The later allegations featured Asian men dressed as cowboys, Masonic lodges, cemeteries, the Park Royal Hotel and private houses far from the creche. Also featured were women creche staff, not just the four who were charged. With them came a panoply of other adults, never found, with names like Spike, Boulderhead and Yuckhead. Ellis's mother also featured. Children were allegedly suspended in cages from rafters in a big hall in the Cranmer Centre. And, of course, there was the notorious "circle incident", where Ellis and his co-workers supposedly took a group of children to 404 Hereford Street on the other side of town and made them stand naked and kick each other while the adults danced around them.

Not part of the videotaped evidence, but alleged by one parent, was the sacrifice of a boy called Andrew. His body hasn't been found, nor has he been reported missing.

Anyone familiar with the American daycare cases would immediately recognise the Civic scenario as a carbon copy of numerous scandals in California and elsewhere. Most of those started too with a single charge of indecent touching. Most of them too then progressed via a similar interviewing process to allegations of sexual violation, urinating and defecating, child sacrifice and weird rituals involving women day-care staff and outsiders. Additionally, most of the American cases also involved middle-class parents and relatively less-affluent creche staff, male and female.

Many of the 127 Christchurch children who underwent videotaped interviews were seen only once, with Sidey accepting they hadn't been abused. But many others were interviewed time and again, often because of parental pressure, with some being subjected to as many as six interviews lasting between one and two hours. Some of these children denied in early interviews they'd been abused, then made allegations in subsequent interviews. Many allegations become more bizarre with each interview.

At the centre of this whirlpool of allegation and suspicion was, throughout, S1. Each time her son, F, was interviewed by Sue Sidey's team he steadfastly denied Ellis had done wrong. But, according to his mother, when speaking to her he produced a steady diet of lurid allegations. She, in turn, fed those tales to the police, many of whom spent a lot of time trying to confirm them. At least some of the detectives accepted what she said was true.

S1 discussed allegations with parent J, mother of L. Charges followed regarding L. Then J told the police she was concerned another child, SB, had been abused, which caused SB to be interviewed. S1, S2, J and C had frequent discussions. C spoke to parent G, leading to G's son A being interviewed. He led police across the Cranmer Centre's rooftops, though no evidence, fingerprint or otherwise, was found to support claims Ellis had taken children over the roof and through the ceilings. S1 spoke to AJ who passed on concerns to AD, mother of children B and C who went on to become complainants. And on it went.

This without doubt was major networking between parents who, convinced large numbers of children had been abused, passed on every rumour, allegation and suspicion to each other. At every new snippet, parents would ask their children if it happened. They planted the idea in young minds. Many of the charges against Ellis and the four women demonstrably stemmed from this.

It must be emphasised that not all parents were involved in this networking. The parents of the Crown's prime witness child S, had no contact with S1's group. Their daughter's allegations against Ellis , involving indecent touching at the creche and while baby-sitting, were made without any apparent contamination from other parents. That's why she was so compelling a witness. She was also the one who recanted during the appeal.

Most creche parents were well educated, rational and thoughtful individuals. I met many of them from both sides of the argument while working on last year's Assignment programme on the Civic creche. Most of those who believed Ellis guilty were as credible as most of those who believed him innocent. How many parents, assured by experts their children had been abused, would doubt what they had been told?

Probably the most suspect of all the "evidence" came relatively late in the investigation from a boy whose allegations led to the arrest of Davidson, Keys and Buckingham. This boy, N, and his mother, S3, were the ones who came up with the "circle incident", the "cages" Andrew's "sacrifice" and other absurdities that helped to put the women creche staff under suspicion. His mother became second only to S1 in promoting the belief the Civic was a lair of mass ritual abuse.

Before the Knox Hall meeting, S3 hadn't been much concerned at the abuse rumours. believing they didn't affect her family. Knox Hall changed her mind and she set about enthusiastically questioning N. "The impression I got [from the meeting] was that you don’t approach the child directly in regard to what may have happened,” S3 stated during the depositions. “I chose to ignore this. The way we worked with N was that if there was a  problem, we would ask a direct question.”

N underwent five videotaped interviews between May and October 1992. Each produced progressively more unusual claims. The first allegation Social Welfare extracted from the by-now six year-old N was that Ellis "wobbled my dick" while changing his nappy at the age of three. Yes, this led to a charge of indecent assault, though it was dismissed by a sceptical judge at depositions. At later interviews, N alleged Ellis defecated on him in a bath, stuck a stick in his anus, sodomised him and subjected him and other children to the "circle" ritual at the house in Hereford Street (he knew the address because his mother took him there on the way to see Sidey). Still later came the cages. His mother also claimed he'd been forced to kill "Andrew" with a knife during a sacrificial rite. Sidey, despite prompting him, couldn't get N to verify this palpable fiction.

The detail N gave worried Sidey and Zelas. Sidey declined to interview him on one occasion because she suspected too much maternal prompting. Zelas also noticed parental questioning of L.

"It is clear that L's parents elicited disclosures of abuse by Peter Ellis by highly leading questioning," Zelas wrote in a letter to Detective Sergeant John Ell on August 28 1992. "N's brother and parents did the same. In N's case, the parents subjected him to intensive interrogation pertaining to ritual abuse... N would then disclose in the next interview with Sue Sidey the information elicited by his parents the previous night."

But far from this casting doubt on the allegations, to Zelas it made them “extra important" because of supposed similarities between what N and L were saying. Small wonder that Ellis's lawyer, Rob Harrison, rhetorically asked Zelas during the trial if anything wasn't evidence of abuse.

At the depositions, N's mother demanded that an American ritual-abuse "expert", Pamela Hudson, be brought to Christchurch. S3 was familiar with Hudson's work. N's allegations contained all 16 of Hudson's "indicators" of ritual abuse; everything from being defecated and urinated on to being held in cages and partaking in sacrifices.

Despite the lurid nature of N's evidence and the blatant maternal coaching which produced them, Ellis was found guilty on three of the four counts concerning N that reached the High Court. The only one the jury rejected concerned the "circle" incident.

A year ago I sat in S3's living room with Assignment reporter Rod Vaughan and director Di Musgrave. For more than two hours we listened to S3, her partner and a friend describe in disturbing detail what she claimed had happened to her son and other creche children. N was at a table behind us the whole time. I kept glancing at this young boy, about the same age as my own eldest child. He sat rigid as a board, eyes bulging, as he listened to every word. It was hard not to believe N was indeed the victim of abuse.

Last October, S3 applied to Justice Williamson for the five videotaped interviews featuring her son. "I believe it is necessary to obtain copies of those tapes to assist N in his healing process," she said to the judge. "He has told me he would like to be able to see the tapes and I think that it is important that he see them in a safe environment."  She enclosed a letter from her son's therapist supporting her application. I'm told Justice Williamson agreed to give her the tapes before his untimely death after heart surgery six months ago.

Several meetings with S3 left me in no doubt she fervently believed the Civic Creche was a hive of abuse. She's now a leading figure in a ritual-abuse organisation, partly funded by the Lotteries Commission, that publishes a newsletter which regularly attacks those who doubt ritual abuse happened at the Civic or anywhere else.

The August 1992 interviews with N, citing the circle incident, put women creche staff under suspicion for the first time. Helping to confirm it was a report on the creche prepared for the Christchurch City Council by psychologist Rosemary Smart.

Smart's report assumed Ellis was guilty, despite it being written almost a year before his trial. It contained extensive accounts of his deviant lifestyle, allegedly related to Smart by his co-workers, though reading the report one sees the influence of mother S1 too. Smart quoted now-discredited research by the New Hampshire sociologist David Finkelhor, whose 1987 book Nursery Crimes became the bible for American believers in ritual abuse. Finkelhor's "research" highlighted the alleged involvement of women day-care workers in sexual abuse.

Detectives involved in the Civic case told me Smart's report was central to their decision to investigate Ellis's female colleagues.

To the police, the possible involvement of creche workers other than Ellis gave the case an alarming new dimension. The women N had named still worked at the Civic. It meant scores of children might still be at serious risk. They decided the creche must be closed for the safety of the children. This was done abruptly on September 3 1992. All 13 staff were made redundant. It was a controversial closure that much later resulted in the Employment Court awarding the sacked staff $1 million damages. The council's appeal against this decision is due to be heard in the Court of Appeal in August.

In late September 1992 the police decided they had enough evidence to charge Gaye Davidson, Jan Buckingham, Marie Keys and Debbie Gillespie. Before arresting them they decided to search their homes and that of another creche staff member, Jenny Wealleans. It's symptomatic of the state of mind of the police team at this time that they didn't go to the nearby district court to get a judge or registrar to sign the search warrants. Instead, they went to the suburbs to get the signature of an elderly justice of the peace, despite it being 10am on a weekday, when the court was open. The irresistible inference is that the police thought a judge might ask tough questions about the warrants.

After Ellis's trial, Marie Keys' husband, Roger, officially complained about the process used to get the warrants. Detective Superintendent Neville Stokes accepted it was valid criticism.

"There is a general police policy... that legal documents should be obtained from a court when these facilities are open and to use justices of the peace out of hours,” Stokes wrote to Roger Keys. "Detective Legat [who obtained the warrants] acknowledges he was aware of this policy but on this occasion made a decision to use a JP on the basis that the issue would be more secure. I do not support that explanation, but it was a judgement he made."

The houses were searched for everything from address books and pornography to babies' bodies. Nothing was found. The women were arrested amid massive televised publicity of the most prejudicial kind. Davidson, Keys and Buckingham were charged with assaulting children during the "circle" incident. Gillespie faced three preposterous indecency charges. The first alleged she had sexual intercourse with Ellis on the creche toilet floor while children watched. Its basis was welfare interviewer Cathy Crawford interpreting how a girl played with dolls. The second alleged Gillespie and Ellis had sat naked in the creche and put their fingers in a two-year-old's vagina. It resulted from insistent, leading questions from the same interviewer.

If anything stands out about the four arrested women it is their ordinariness. The thought of the kindly Marie Keys taking part in the sexual abuse of children is laughable. The charges against the four were so preposterous it is extraordinary they were laid, let alone survived almost to trial. All four were sent for trial after the depositions. The only charge against them dismissed at depositions was the one claiming Gillespie had sex with Ellis. Justice Williamson reluctantly discharged the women before Ellis's trial, on the grounds their chances of a fair trial would be prejudiced by their association with Ellis. He later rejected their application for costs, making it abundantly clear in his judgement that he supported the police decision to bring charges.

The police who decided to arrest the women believed creche staff socialised together, as if this proved criminal intent. They believed they watched pornography together, but not a shred of evidence was found. The police seized pornographic videos from shops and individuals all over Christchurch and tried to link them to the Civic inquiry, but there was no link. Mothers S1 and S3 were always approaching the police alleging fantastic conspiracies involving the creche staff. They adopted a number of their delusions as facts.

In Davidson's office at the creche, the police found two obviously joke employment letters, one purporting seriously to decline a job application, its twin stating the creche already employed a paedophile, so it didn't need another. At Gillespie's house they found an equally joking scrap of paper about children watching through a window as "Mr Playschool" had sex with women inside a house.

Senior police officers proffered these stupidities to me as proof of their allegations against the women, even though the documents concerned had been suppressed during the depositions and had not been presented at Ellis's trial. One senior detective showed me an intercepted letter, written by Ellis from jail to Jan Buckingham's daughter. It contained raunchy sexual comments the officer rightly intimated were unacceptable to send to a child. But the daughter concerned is 23. Ellis says he wrote the letter in response to one from her. This also raises the question, what right do the police have to intercept letters from prison inmates and show them to passing journalists?

It would be easy to allege bad faith on the part of the police involved, but having met the leading investigators and discussed the case with them, it's clear everything they did was based on their firm belief the Civic was as bad as they portrayed it. Like the parents, they were told by experts this was so.

The man in overall charge, Chief Inspector Brian Pearce, was vilified after appearing on Holmes in June 1992 to say the Civic was evidence society was reaping the fruits of mocking Patricia Bartlett, John Banks and God.  But Pearce and his colleagues genuinely believed the Civic was a den of Sodom. Society has reached a sad stage when good men like Pearce are attacked for their moral and religious beliefs rather than the evidence they put forward of a crime. The evidence was sadly wanting in the Civic case but not the investigators' convictions.

The most serious criticism of the investigators, police and Social Welfare staff, has to be the fact they lost all sense of perspective. Nobody, it seems, stood back and asked if the fantastic tales they were hearing were plausible, given all the circumstances. All appear simply to have accepted the theory that any allegation of abuse must be true.

Crown prosecutor Brent Stanaway was apparently the first to realise the limitations of the police case. He quickly narrowed the charges. In the absence of medical evidence, charges of sodomy and vaginal rape became Ellis placing his penis against vaginas or anuses. Stanaway strove successfully to prevent any whiff of ritual abuse from reaching the jury. Though the women's participation was crucial to the police case, Stanaway was able to convince the jury of Ellis's guilt without the women being in the dock too. He ran a case good enough to put Ellis away for 10 years despite the abundant doubts raised by the nature of the evidence, or lack of it.

Of 28 charges that reached the trial, the jury found Ellis guilty of 16 and not guilty of nine. Three others collapsed during the trial when children denied them. One child innocently told the court that "Cathy taught me what Peter did" before the interview began. Cathy Crawford was the interviewer whose interpretations led to Debbie Gillespie's arrest.

Since the trial of Peter Ellis three years ago, strong new evidence has emerged to cast serious doubt on the case against him.

While helping with the Assignment programme on the Civic affair I sent thousands of pages of court documents and transcripts of the videotaped interviews to one of the world's foremost experts on the credibility of children's evidence, Professor Stephen Ceci of Cornell University, New York State. Unlike the "rent-an-expert" typically wheeled into trials by the prosecution and defence, Ceci has never taken the stand for one side or another in a child-abuse case, but he's conducted extensive academic research into them.

I asked Ceci to review the material I sent him and to come back with whatever he made of it, whether he thought Ellis guilty or innocent. His considered response was heavily damaging to the Crown case.

"This case entailed an array of factors that give me cause for concern," Ceci said.
"Children frequently reported highly implausible events that were never checked, for example the presence of the defendant's mother during baths, repeated sodomy occurring only minutes apart with other children who were said to be present, and they were never reined into reality. That some of their claims were plausible is no assurance that they did not emanate from the same sources as the implausible claims."

Ceci said the child interviews were not the worst he'd seen, nor the best. They were typical of those in the US in the late 1980s. He didn't believe Sidey and her team engaged in a witch hunt, but nor was their style designed to maximise accuracy. They didn't query allegations to test their veracity, but they repeatedly pressed for details of abuse.

"There wasn't any effort to falsify the [interviewers'] hunches [that Ellis was an abuser]. Often there was a repetition of questions: it was almost as though the interviewers were surprised that the child said `that's all' Mr Ellis did to them and therefore they would repeat the same question over and over again. There wasn't an effort to rein the children back into reality when they roamed into these fabulous claims. Whether or not the interviewers' minds were made up prior to the interviews I can't say, but what I can say [is] there was no serious attempt to test an alternative hypothesis to the [Crown's] claim that Mr Peter Ellis molested these children."

Most compelling was Ceci's verdict on whether the alleged abuse had occurred.
"Some of the things the children said I would be exceedingly sceptical that they ever occurred. It, in my experience, is exceedingly unlikely that you can coerce a group of children this age into silence for prolonged periods of time when the following were allegedly involved... anal penetration, forcing children to walk over precarious ladders perched high above buildings, defecating and urinating on children... these are events which cause almost instant revulsion in children, night tremors, unwillingness to go to school, fear of the perpetrator.

“In my view it is very very unlikely that you could persuade children to be silent about that for long periods and also to exert affection for the perpetrator which many of these children did. So on that level I'm exceedingly sceptical. I don't think the bizarre stuff happened. Does that mean nothing happened?  Well, I simply don't know. No one else knows either except God and Mr Peter Ellis."

I sent Ceci's 14 close-typed pages of reasoned comments to Attorney-General Paul East. He passed them to the Crown Law Office, which apparently gave them to Social Welfare. Neither organisation has publicly challenged Ceci's findings.

Since then, new New Zealand research has emerged to cast doubt on the multiple interviews of the Civic children. For a research project, Hamilton child psychologist Jane Rawls had 30 five-year olds play dress-up games alone in a room with a man called Trevor. The games consisted of the child and Trevor selecting two items of clothing from a box and putting one item of clothing on themselves and one on the other.  No clothes were removed to do this.

Later, each child was interviewed four times in a manner similar to Social Welfare evidential interviews. Some were asked open questions about what happened, others were asked leading questions such as, "Did Trevor put his hand over or under your dress?" Seven children reported they had been abused, either by genital touching, having their bottoms touched or being made to touch Trevor. Of course, no abuse happened, as could be proved by videotapes of the dress-up games.

"Their errors seemed to evolve over time with repeated interviews," Rawls wrote in the April 1996 edition of Law Talk.

Her research casts obvious doubt on the accuracy of the multiple interviews and leading questions used to elicit evidence from the Civic children. Social Welfare reacted critically to Rawls' findings, but a police response was exceedingly interesting.

" Our concern about the research is the number of interviews that were carried out," wrote police legal adviser Mark Copeland and child-abuse interview training coordinator Wendy Burgering in a letter to the Dominion on May 31 1996 after the paper published an article on Rawls' research.

"Dr Rawls stated she did four interviews with each child. For specialist interviewing this is well above the standard practice of one interview. Specialist interviewers are very aware of the problems of the reliability of the disclosure through overinterviewing."

Just one interview? Tell that to the Civic Creche prosecutors.

Did Ellis get a fair trial?  Auckland University evidence expert Scott Optican says there's nothing unusual in the way Justice Williamson managed the trial, nor in the Court of Appeal decisions which upheld the convictions. That doesn't mean they are beyond question.

Optican says the large number of charges Ellis faced made his case hard to defend: "It's the argument about where there's smoke, there's fire. The jury thinks there must be something there."  Similarly, Williamson's decision to admit evidence that Ellis had talked of practices like golden showers was extremely prejudicial. "It would have had a very powerful effect on the jury." But probably any judge would have allowed that evidence.

On the other hand, says Optican, the fact the charges came from the uncorroborated evidence of young children should have made the prosecutors and the judge more cautious. It would have been fairer had Justice Williamson allowed the defence to play the videotapes which Rob Harrison believed cast reasonable doubt on the more credible witnesses.

"The Court of Appeal upheld the convictions on the grounds it believed the judge's evidential rulings were correct and the jury was entitled to believe the kids. What sustained that was the way the jury discharged Ellis on some of the more bizarre counts. One way to look at that is it meant the system worked. The other way to look at it is that the bizarre ones taint the whole thing."

Optican notes the Appeal Court only reluctantly quashed the charges relating to S, the prime witness who recanted, because the court accepted the "denial" argument. But her retraction, and the dropping of many other charges, could also be viewed as casting doubt on the remaining testimony.

" Ceci would say that a lot of sexual abuse can go undetected for a long time. On the other hand, given the number of allegations in this case over such a long period, it's difficult to imagine they could go undiscovered. It's really hard to imagine how it took five years to come to light. What is disturbing about the case is, it may be that Ellis was guilty of some things, but not what he was found guilty of.”

At Rolleston Prison, Peter Ellis says he's much more concerned for the children than himself. He wants his convictions quashed for them, not him.

"The real child abuse is what has been done to the children in the judicial, process," he says. "I'm angry that all those kids will go through life with everyone thinking they were abused. If they get to I 19 and have some problems, everyone will say, `Oh, he was at the Civic Creche.' They need to know that nothing happened there."

It's telling that Ellis hasn't suffered the fate of many convicted child molesters: being beaten up by fellow inmates. The prison guards who sat through his trial let it be known they didn't believe the evidence against him. Word soon spread. He's had no trouble at all.

Ellis was initially in Paparua Prison but was transferred to Rolleston after taking up complaints about prison life on behalf of other inmates, even calling in the Ombudsman. He has a determined support group outside, led by Winston Wealleans, husband of former creche staff member Jenny Wealleans, and Roger Keys, husband of Marie Keys. They beg him not to create a stir at Rolleston, because if he's transferred away from Christchurch it will be even harder to help him.

Those who know Ellis say he's not as outrageous as he was. Three years in jail has matured him. But he is still wearing mascara the cold Saturday morning I talk to him in the visitors' room and his sense of humour is still cutting: at Paparua he ran into David Bain, the Dunedin man convicted of killing his parents and siblings, and Ellis said to him, "Hello David. I'll be your friend, but don't treat me as family."

Ellis has never said the children lied, just that what was alleged didn't happen. He's spent a lot of time thinking where some of the stories came from, parental input apart, of course.

"I was supposed to have driven them in my car to houses to abuse them, but I don't drive, I've never had a car. But at the creche we played in make-believe cars made out of boxes. That's where it would have come from. And talk of the staff being bad, it came from the Wizard Of Oz. For weeks [after it was on television], all the kids would play Dorothy, Toto and the Wizard. But none of them of course would play the wicked witch, so Marie had to. Creche staff always played the baddies in the games."

Ellis says he's aware that many lawyers and law students believe he's not guilty. He's angry they sit around discussing his case at morning tea but don't support him publicly.

"I'm angry that I could be prosecuted for abusing unknown children at unknown places at unknown times. Where are all the children L said were abused with her?  Where is Andrew's body?  Why hasn't he been reported missing?  The interviewers assumed if a Peter was mentioned it was Peter Ellis. N had bruises on his knee at one interview and told Sue Sidey, `Peter did it.' But it couldn't have been me. I'd been arrested long before. She didn't ask him the obvious: `Peter who?' They never did."

Footnote: The four former creche workers arrested with Ellis are trying to rebuild their lives. Gaye Davidson is selling real estate, Debbie Gillespie is at university, Marie Keys is doing voluntary work and Jan Buckingham is looking after her house and children.

The civic premises in the Cranmer Centre are again being used as a childcare centre.