False Sexual Abuse Allegations - Lessons from the Peter Ellis Case - Part 1B

Introduction

This article is the result of Ross Francis’s research into the Peter Ellis case. Part 1 examines the expert opinion evidence proffered at Peter Ellis’ second appeal hearing. The opinion of Michael Lamb is discussed in detail. The allegations of one of Peter Ellis’ accusers are evaluated in the context of the relevant research findings. In addition, Cabinet’s decision to hold a ministerial inquiry is briefly explored.

Part 2 examines the outcome of that inquiry, the conduct of officials and the subsequent inquiry by Parliament’s justice and electoral select committee. Options that could provide long-awaited finality are discussed.

This article cites official documents, affidavits, Cabinet papers and the latest research on child sexual abuse. Many documents have been made available, some with the Ombudsman’s assistance, only within the last year. The new evidence raises questions about the efficacy of the criminal justice system. In particular, it raises questions as to whether Peter Ellis’ convictions are safe.

Background

In 1992, Ellis was charged with sexually abusing 20 children who attended the Christchurch Civic Crèche. Four female crèche workers were also charged with sexual offences, but their charges were later dropped. In 1993, Ellis was convicted on 16 counts involving seven children. After twice refusing parole, he was released from prison in 2000.

At Ellis’s 1994 appeal, R v Ellis (1994) 1 NZCrimC 592 (CA), the oldest child recanted her allegations. However, the appellate Judges were unconvinced that her retraction was genuine. Casey J contended that it was not uncommon for young victims of sexual abuse to withdraw their allegations. Nevertheless, convictions relating to the child who recanted were quashed, but the appeal was otherwise dismissed.

In 1999, a reference by the Governor-General was heard by the Court of Appeal, R v Ellis [2000] 1 NZLR 513. The Court’s position was that certain matters had already been heard and disposed of. New evidence was provided in the form of expert opinions. Affidavits were proffered by Michael Lamb, Barry Parsonson, Maggie Bruck – all supporting Ellis – and Constance Dalenberg.

Expert Review of evidence

Michael Lamb is possibly the leading authority on the interviewing of child abuse victims. The Cambridge University professor has published numerous articles in scientific peer reviewed journals. In 1998, he co-authored Investigative interviews of children: A guide for helping professionals. The book’s intended readers were police, social workers and forensic interviewers, whom Lamb has advised and trained. Its guidelines have been cited by experts testifying in Court and have become the standard in several countries.

In 2004, the American Psychological Society presented Lamb with a lifetime contribution award. The society said that his work had “fundamentally advanced the interests of young children and their families”; it called him a scientist and scholar of “the highest integrity”.

Lamb argued that young children could be competent witnesses but could also be susceptible to errors when interviewed because they:

1. infer that the interviewer wants a particular response;
2. want to help but do not understand the questions;
3. retrieve information recently acquired about the event in question; and
4. become confused as to the source of their memory about the event.

When evaluating forensic interviews of children, Lamb’s preference is to focus on the interviewer rather than the child. This is because an interviewer’s behaviour, particularly vocabulary, complexity of their questions and their ability to elicit useful information from children, “profoundly influences the course and outcome of their interviews”.

Lamb cited research by Gail Goodman showing that a significant minority of young children were error-prone when asked specific abuse related questions. When questioned in a laboratory setting, between 20–35 per cent of three to four-year-olds falsely assented to questions such as “Did he try to kiss you?”, “Did he keep his clothes on?” and “He took your clothes off, didn’t he?”. Studies have found that false and potentially troublesome claims can also be elicited from pre-school and school-age children, even when asked non-leading questions.

Lamb noted that only 6 per cent of the questions in the Ellis interviews were suggestive (he classified a further 36 per cent as leading). He said that the use of suggestive questions was “not remarkable”. However, the Ellis interviews were conducted weeks and months after informal interviewing of the children began. Children were exposed to interviews and conversations “that are known to contaminate children’s accounts of either experienced or imagined events”. Furthermore, the average delay between formal interviewing of the conviction children and the alleged events was 18 months. Lamb wrote that given such a delay, the children were likely to have adopted recently acquired information about the events in question.

Free recall, non-leading and open-ended questions elicit the most accurate and detailed responses from children. Children should explain in their own words what has happened to them (eg “Tell me what you did today.”). This, according to Lamb, seldom occurred in respect of the children’s formal interviews. In the early 1990s, UK and US interviewers obtained more than twice as much information from open-ended questions as did interviewers in the Ellis case. By the mid-to-late 1990s, the gap had increased: when employing open-ended questions the Ellis interviewers elicited only 14 per cent of what UK and US interviewers elicited.

The Ellis interviewers eschewed the most desirable types of questions in favour of riskier alternatives.

Children should be questioned as soon as possible after the target event. The longer the delay, the greater the potential for children’s memories to be contaminated by misinformation. Lamb argued that once contamination had occurred, it was “often impossible” for young children to distinguish between real and suggested events. This was especially true if details were reinforced over time by repeated suggestive questioning.

Unintentionally false reports can be elicited from children even when the event in question is recent. Lamb cited research by Garven et al, “More than suggestion: The effect of interviewing techniques fromtheMcMartin Preschool case” (1998) 83 J of Appl Psych 347, which found that 58 per cent of four to six-year-olds accepted false or misleading information about a week-old event. Moreover, 44 per cent of children falsely assented to questions about touching after just five minutes of improper questioning. This raised concerns about the accuracy of the allegations in the Ellis case, given that children had been informally questioned over a period of weeks and months about events that had allegedly occurred months or years earlier.

According to Lamb, there was no serious effort to test the complainants’ claims. Those involved with the investigation were “singularly focused” on any evidence consistent with the hypothesis that abuse had occurred. However, psychiatrist Karen Zelas, who reviewed the children’s evidence for police, knew that some parents were questioning their children improperly. In a letter to Detective John Ell, she explained that two children, including Tommy Bander, had been exposed to “highly leading questioning” by their parents. Further, Zelas wrote, Tommy’s parents had:

subjected him to intensive interrogation pertaining to ‘ritual’ abuse … [i]t is an extremely difficult inquiry with such young children and it is most important that their statements are neither dismissed as fanciful nor accorded unwarranted weight primarily because of parental anxieties.

Collecting physical evidence consistent with the abuse hypothesis proved troublesome. In October 1992, Detective Ken Legat handed Lesley Ellis (Peter’s mother) a search warrant, giving police access to her Buffon St flat. They expected to find “instruments or sexual aids used in sexual offending”. None were found. Legat claimed, in an affidavit supporting the search warrant, that overseas studies and investigations showed that “this type of abuse on children have (sic) occurred in various crèches and play schools”. Michael Lamb agreed that the Ellis case shared “startling similarities” with daycare cases overseas. Social scientists generally believe that such cases were the product of a moral panic; doubt remains as to whether any children were actually abused.

Legat’s belief that crèches and play schools were havens for paedophiles was influenced by Rosemary Smart’s report into the Civic Crèche. The Christchurch City Council hired Smart, a qualified social worker, to review the performance of senior crèche staff. Smart, who began her review just days before Ellis’s arrest, repeatedly cited the findings of American sociologist David Finkelhor, a self-proclaimed expert on (and believer in) satanic ritual abuse. Smart accepted Finkelhor’s claims that child sexual abusewasmore prevalent in childcare centres than elsewhere and that 40 per cent of abusers in childcare centres were women (a large New Zealand study recently found that women committed only 1 per cent of alleged child sex offences). Smart was apparently unaware that Finkelhor’s claims were based on unsubstantiated cases of sexual abuse. She asserted that children attending the Civic Crèche had been sexually abused by a male staff member qualified in early childhood education. It was obvious who she was referring to. She questioned how the abuse had remained undetected for several years and why it did not “arouse serious concern on the part of staff”. Four experienced female childcare workers were subsequently charged with sexually abusing children in their care.

Lamb noted that children in the Ellis case made similar allegations at or near the same time. This suggested contamination, “not validation”. Although delayed disclosure did not imply deceptive disclosure, the fact that children were no longer in contact with Ellis reduced the likelihood that they would remain silent (about abuse). No child, said Lamb, alleged abuse when first questioned by their parents. The vast majority did not make allegations when formally interviewed.

Social influence probably affected the disclosure process. An evidential interviewer or police officer would talk to a suspected victim shortly before the child was formally interviewed. Such conversations were not recorded. Colin Eade, who led the police investigation, monitored many of the formal interviews. He spent up to thirty minutes with children prior to their interviews and had, according to Lamb, “ample and unchecked opportunities” to shape their claims. He also made unscheduled visits to the children’s homes “to try and [help them] overcome [their fear] prior to evidential interviews” (Police Report Form, 19 March 1992).

Interviews with young children

It was not very long ago that many social workers and clinicians believed that children were incapable of making false allegations of sexual abuse. But in the late 1980s and early 1990s a series of mass allegation daycare cases raised the possibility that children’s “memories” of sexual abuse could be distorted bymere suggestion. Experimental research has since confirmed this conclusion; moreover, researchers have found that a single interview can have powerful and lasting effects, producing false reports in later non-suggestive interviews.

Lamb’s affidavit listed nine conditions under which suggested information was likely to be adopted by a young child. Among the conditions were:

• details are suggested repeatedly;
• an air of accusation is established;
• the questioner responds positively to some statements and ignores others;
• the child is told that others have reported the details in question;
• some details are rehearsed;
• conversations with sources of contaminating information (including parents, peers, counsellors, and police proceed unchecked);
• any real memories are weak.

All nine conditions were present in the Ellis case, said Lamb, making it highly likely that the children’s reports were (unintentionally) tainted. The risk of contamination was so high:

and the failure [by investigators] to explore alternative hypotheses so obvious that it is almost impossible for either an expert or a tribunal of fact to determine which if any of the complainants’ accounts were valid. (Michael E Lamb, R v Ellis (CA 120/98))

In May 1992, evidential interviewer Lynda Morgan formally questioned child complainant Tommy Bander, aged six years and two months. Below is an excerpt from that interview:

A: He smacked my bum.

Q: And he smacked your bum, yeah?

A: Real hard.

Q: Real hard. I wonder why he smacked your bum.

A: I can’t remember. I don’t. I remember he smacked it.

Q: Right, so where, did you have clothes on?

A: He pulled down my pants because I had to get changed.

Q: Oh, why did you have to get changed?

A: Because I done poos in my bum, that was when I was really, really very little.

Q: Well, do you think there’s anything else…that that you need to tell me about crèche and about Peter?

A: No.

Q: So that’s the things that you told … too, aye?

A: Yep.

Q: Mmm, okay, so you think that’s absolutely everything about the things you told Colin [Eade] and Mum about Peter and the crèche. Can you remember any other things happening that you didn’t like?

A: There was no other things anyway.

Q: There was no other things?

A: Nope.

During the same interview Tommy claimed that while he was being changed, Ellis “wobbled my dick”. He was later asked the following leading question:

“He [Peter] didn’t pull his pants down, so you didn’t see any of his rude bits?”

“No”, Tommy replied, before adding that “he might have done it to other children … but not to me”.

Despite Tommy’s denials, Lynda Morgan told Zelas that she believed Tommy had been indecently assaulted.

Tommy’s mother, Joy Bander, testified that she spoke with Lynda Morgan after her son’s first formal interview. She said Morgan did not tell her that Tommy had apparently soiled his pants, which was why Ellis had had to pull them down. When cross-examined as to whether she believed, after talking with Morgan, that Tommy had more to disclose, she replied: “Absolutely”.

Prior to Ellis’ trial, Karen Zelas advised Brent Stanaway that Tommy’s evidence was consistent with “a cleaning up procedure”. Furthermore, she asserted that “the investigation of Tommy’s circumstances were (sic) considered complete after his first interview”. (She failed to inform jurors of this fact.) Lynda Morgan’s opinion of Tommy’s abuse status was not shared by investigators. So why was he interviewed again, months later? According to Zelas, it was all down to his parents. It was “hard to believe”, she said, that they would have “accepted an opinion that Tommy had not been abused”. (Affidavit presented to Brent Stanaway, 22 March 1993)

Denial of child sexual abuse

Contrary to popular belief, abused children are unlikely to deny abuse when asked. (However, retrospective studies show that most adults did not disclose their childhood abuse at the time, presumably because they were never asked.)

In possibly the largest study of its kind, 26,325 Israeli children were formally interviewed between 1998 and 2002. (Hershkowitz, Horowitz and Lamb, “Individual and family variables associated with disclosure and non disclosure of child abuse in Israel” in Pipe, Lamb et al (eds) Child Sexual Abuse: Disclosure, Delay, and Denial (Lawrence Erlbaum Associates, 2007)) They were suspected victims of physical or sexual abuse. The disclosure rate for sexual offences was 71 per cent. Older children were more likely to allege sexual abuse than were younger children. In comparison, 77 per cent of suspected victims alleged sexual abuse in a recent American study. Among the youngest children, 63 per cent of four- to five-year-olds and 77 per cent of six- to eight-year-olds made allegations. (Pipe, Lamb et al “Factors associated with non disclosure of suspected abuse during forensic interviews” in the same book.) A disclosure rate of 68 per cent was observed in a recent New Zealand study. Some of the 4060 suspected child victims were interviewed though they had made no prior allegation of sexual abuse. The disclosure rate for children who had made a prior allegation was 88 per cent. (Wilson, “Forensic interviewing in New Zealand”, ibid)

Also contrary to popular belief, children generally do not need to be prodded to disclose abuse. In the Hershkowitz et al and Pipe, Lamb et al studies referred to above, experienced interviewers trained in state of the art interviewing methods were employed. Suggestive and leading questions were generally avoided. Free recall and open-ended questions predominated and few children were interviewed more than once. Neither anatomical dolls nor body diagrams were used.

Tommy Bander

Tommy Bander denied having been abused when (repeatedly) questioned by his parents and oldest brother. He did not make a clear disclosure during his first formal interview. However, although most sexually abused children do not deny the abuse, younger children do not appear to be as forthcoming as older children. On the other hand, false allegations can be elicited from young children within five minutes of improper questioning (Garven et al). On one occasion Tommy was questioned by his parents for two and half hours. See Joy Bander, A Mother’s Story. (Howling at the Moon Productions, 1997)

Tommy’s memory of the alleged events was weak. During his fourth formal interview, he said that three female crèche workers had stuck needles into his penis. When asked “so what stopped you from telling me [about that] yesterday?”, he replied: “Oh, I just remembered today”.

Tommy had at least four therapy sessions prior to his second formal interview. Gayle Taukiri, his therapist, confirmed that she showed him satanic signs and asked him to identify them. She claimed, however, that she did not talk to him about Peter Ellis or the Civic Crèche, unless he raised these matters of “his own volition”. Tommy’s allegations of ritual abuse appear to have occurred only after talking with Taukiri (who took no notes during their sessions).

Taukiri subsequently advised Joy Bander that American ritual abuse “expert” Pamela Hudson should be brought to New Zealand to assist police.

Pipe, Lamb et al found that among children aged six to eight years, 70 per cent made allegations of sexual abuse when the alleged offender was an immediate family member (eg biological father). When the alleged offender was familiar but unrelated, 81 per cent of this age-group disclosed. Tommy had had no contact with Peter Ellis or other crèche staff for more than a year when his parents and eldest brother began questioning him. According to Lamb:

abused children are most likely to keep secrets when still in contact with (and presumably fearful of) the alleged abuser andwhen their parents are skeptical or unsupportive.

Neither was true in Tommy’s case. In fact, Tommy’s mother was skeptical of his denials and encouraged him to disclose.

During the police inquiry, parents were instructed not to ask their children direct questions about Peter Ellis. Joy Bander admitted at the depositions hearing that she found this advice “rather odd”. She said she asked Tommy direct questions because she always dealt with him in this manner. She did not tell the Court that when the police investigation began she was supporting five children on a sickness benefit of $289 a week. Nor did she say if Tommy’s claims of abuse had caused her to apply for lump sum compensation from ACC.

If Tommy was not abused, why has he not recanted? The obvious answer is that he believes he was abused. To quote leading child sexual abuse expert Debra Poole:

Studies have shown that children will vehemently defend the veracity of implanted memories. They recall reporting them, and those reports produce mental images of the events that these individuals cannot distinguish from their real experiences. But the kids are not responsible for that. The interviews are. (Quoted in the Boston Herald, 8 July 2001)

In 2003, Tommy, then 17, reportedly told journalist Linley Boniface that “all my parents ever said to me was that I should tell the truth”. He added:

I stand by everything I said when I was little. I didn’t make anything up. But back then I believed everything I was told. … when you’re a little kid, you think adults are always telling you the truth. (“I am sick of being called a liar”, Dominion Post, 16 August 2003)

It is believed that between 1995 and 2003, Tommy was able to watch his videotaped evidential interviews. His mother told the High Court, when requesting copies of the tapes, that Tommy could not heal unless he saw the interviews. The Court agreed.

In 1992, the then six-year-old reportedly said that: four female crèche workers had watched as sharp sticks and burning paper were inserted into his anus; Ellis’ mother had hung five cages, in which there were children, from cables attached to the ceiling of the crèche; three female crèche workers had inserted needles into his penis; he had been forced to kill a boy. He revealed the first names of 11 women who he claimed had physically or sexually abused him. He made similar accusations against several men and five teenagers.

It is worth noting that in Wilson’s (2007) study, 96 per cent of suspected abuse victims were formally interviewed only once. Tommy, however, was questioned five times over five-and-a-half months. He did not make any unambiguous disclosures during his first interview. Investigators, however, seemed to be under the impression that abused children were reluctant to talk about their abuse. Suspected victims needed to be encouraged to disclose. Ellis was convicted on the basis of allegations made in Tommy’s later interviews.