The Christchurch Civic Crèche Case

A City Possessed - by Lynley Hood - Index

A City Possessed - Reviews


New Zealand Law Journal
September 2003

Christchurch Crèche Case: an author possessed?
Emma Davies and Jeffrey Masson question the psychological research in A City Possessed

A City Possessed:
The Christchurch Civic Creche Case
by Lynley Hood.
Publisher: Longacre Press 2001

Lynley Hood's book A City Possessed appears to have inspired a petition, a $100,000 reward for information and a paid advertisement of excerpts from interviews of child complainants (Sunday Star-Times, 8 August 2003). Unlike some of the petitioners, we cannot comment on the guilt or innocence of Peter Ellis, but if the research of the first three chapters of Lynley Hood's book is anything to go by, one should be cautious about the neutrality of her analysis. Weaving a story and selecting facts to fit the story is not the same as carefully documenting the history of events.

Hood describes herself, over and over, as objective. She promises to write (p 31) "as fairly, accurately and sensitively" as she can. She claims (p 32) that she "steered a steady course through the shoals and tempests of the controversy, illuminating the context along the way providing a balanced picture of the apparently conflicting, but presumably equally valid, points of view". She declares that she especially does not want to "oversimplify the world around us".

Our comments relate to the first three chapters of her book, the theoretical foundation for her later statements. , These chapters are devoted to explaining the issues around child abuse. Hood writes (p 34): "At the core of these arguments are questions that must be addressed: Just how widespread is child sexual abuse, and just how serious are the consequences? (Or, to put it another way: what exactly does the research say, and what exactly does it mean?)". We agree. These questions are crucial.

She asks: "Is the true prevalence of child sexual abuse higher than one in four, lower than one in 1,000 or somewhere between?... Clearly the only answer is: we don't know (italics in original)." But in 1994, Professor David Finkelhor published a review of prevalence studies with large samples in 19 countries, including New Zealand ("The international epidemiology of child sexual abuse", (1994) 18 Child Abuse & Neglect 409). He found that the variability was between seven and 36 per cent for women and three and 29 per cent for men.

In the 1990s, two New Zealand prevalence studies were published in peer reviewed scientific journals. In a University of Otago study by Jessie Anderson and colleagues, 32 per cent of a random sample of 3000 adult women (up to 65 years old) stated they had unwanted sexual experiences before they turned 16 ("Prevalence of childhood sexual abuse experiences in a community sample of women", [1993] Jo of American Academy of Child and Adolescent Psychiatry 911). Twenty per cent reported unwanted touching of their genitals, or attempted or completed sexual intercourse.

Professor David Fergusson and colleagues' research on a birth cohort of 1,019 18-year-olds in the Christchurch urban area found that 17.3 per cent of girls disclosed unwanted sexual experiences prior to sixteen with 12.6 per cent reporting genital stimulation and 5.6 per cent vaginal, oral or anal intercourse ("Childhood sexual abuse and psychiatric disorder in young adulthood", [1996] JAACAP 1355). Fergusson and his colleagues describe their data as a "lower level estimate", explaining "it may be that the estimates in this study were influenced by underreporting and it may be that, at age 18 years, a number of those exposed to abuse may not have been emotionally ready to disclose CSA [child sexual abuse]". Despite this, one in 20 young women described attempted or completed vaginal, oral or anal intercourse.

Two years before the publication of A City Possessed, David Fergusson and Paul Mullen published an evidence- based review of childhood sexual abuse research (Childhood Sexual Abuse - An Evidence Based Perspective 1999. Thousand Oaks: Sage Publications p 28). They concluded that despite methodological limitations of individual studies, "exposure to some form of unwanted sexual attention during childhood is all too common.... The majority of studies suggesting prevalence [for females] somewhere in the interval between 15% and 30%...with most estimates for males lying within the interval between 3% to 15%. Of perhaps greater concern is the number of children who report being exposed to acts of sexual abuse that involve sexual penetration. The estimates suggest that up to 13 % of females and 11.3% of males may report abuse of this severity, with the majority of studies reporting prevalence figures in the range of 5% to 10%".

Data from incidence studies (commonly the number of reported cases in a set period of time) are not directly comparable to these prevalence surveys (commonly the proportion of population recalling abuse over their lifetime). They each offer different lenses on the extent of child sexual abuse. Hood chooses not to clarify the distinction, highlighting whichever produces the lower number. She selectively emphasises the 1981 American National Study of Incidence and Severity of Child Abuse and Neglect (p 69), which found the incidence of sexual abuse among American children to be 1 in 500 (Washington DC: US Dept of Health and Human Services). Why 1981? After all, she was writing in 2000. Why did she not visit the US Government Department of Health and Human Services website and find the 1996 study, which showed "substantial and significant increases in the incidence of child abuse and neglect since the last national incidence study was conducted" (Sedlak & Broadhurst, Third national incidence study of child abuse and neglect. US DHHS).

Hood states that it is not known whether "most women who were molested as children experience some difficulties in their adult life, or is sexual abuse just one of many adverse childhood experiences that may or may not cause lasting harm?" A review by Kendall-Tackett and colleagues of 45 studies of the effects of child sexual abuse shows that sexually abused children have more symptoms than non-abused children, with abuse compared to other adverse childhood experiences accounting for 15-45 per cent of the variance in symptoms ((1993) 113 Psychological Bulletin 154). Symptoms included fears, behaviour problems, sexualised behaviours and poor self-esteem with no one symptom characterising abused children.

In the longer term, although some survivors show no ill effects according to the measures used, many experience serious problems throughout life, particularly when the abuser is a family member, when there are multiple incidents of abuse and when the abuse involves attempted or actual intercourse. Even in studies that control for other adverse childhood experiences, child sexual abuse has been linked to suicide, depression, psychiatric admissions, drug abuse and eating disorders (See Fergusson & Mullen 1999; Neuman et al. "The long-term sequelae of childhood sexual abuse in women: A meta-analytic review", Child Maltreatment 1996; 6-16 and Putman,"Ten-year research update review: Child sexual abuse" [2003] JAACAP 369).

Abuse survivors are ill more often and have surgery more often (Felitti et al. "Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults", [1998] Am Jo of Preventive Medicine 245). As Anna Salter, a prominent forensic psychologist, recently commented: "I had not read the literature on the impact of child sexual abuse carefully for some time ... I initially planned on reading ALL the literature on the sequelae of child sexual abuse, but that grandiose plan faded as I read for months on end without being able to tap into all the research. At the end of several months, however, I was convinced of one thing. Child sexual abuse was like getting bitten by a rattlesnake: Some kids recovered completely, and some didn't; but it wasn't good for anybody" (Predators, Paedophiles, Rapists, and other Sex Offenders 2003. New York: Basic Books, p 63).

While Hood is of course entitled to her personal opinion, she cannot justifiably claim her conclusions to be robust or balanced. She seems intent on minimising the extent and consequences of child sexual abuse. Her caricatures of the work of Diana Russell, Roland Summit and Miriam Saphira are unhelpful to those seeking a fair and reasoned understanding of child sexual abuse.

Russell's study of the prevalence of sexual abuse of girls was based on a random sample of 930 adult women in San Francisco ("The incidence and prevalence of intrafamilial and extrafamilial sexual abuse of female children", [1983] Child Abuse & Neglect 133). This was one of the first such surveys on this scale. The results and methodology were so compelling, that university-based researchers attempted something similar in many other countries.

Russell's definition of sexual abuse within the family was: "any kind of exploitative sexual contact that occurred between relatives before the victim turned 18. Experiences involving sexual contact with a relative that were wanted and with a peer [age difference of less than five years] were regarded as non-exploitive". Her results: 28 per cent of the 940 women reported at least one experience of sexual abuse within the family before the age of 18; 12 per cent reported same before 14.

Twenty-three per cent of all incidents of abuse within the family were classified as Very Serious: there were defined as completed and attempted vaginal, oral, anal intercourse, cunnilingus, analingus, forced and unforced. In just one instance, Russell classified "unwanted but non-forceful kissing by a cousin" as sexual abuse. Hood's often-subtle bias becomes starkly transparent when she writes: "Russell's definition included everyday aspects of family life like unwanted hugs and kisses". Hood's misuse of one isolated case seems inconsistent with her stated goal of objectivity.

For well over a hundred years no mental health system, never mind society at large, was willing to recognise the reality and seriousness of child sexual abuse (see Seymour, "Memory and childhood abuse", [1996] Waikato L R 155). However, Hood (p 30) is sceptical about "the bad old days when victims were disbelieved" and says that this is "a self-serving myth". Her source? Sixty-five years of the New Zealand Truth shows that "children's complaints of sexual molestation were taken seriously".

Summit was the head physician for the Community Consultation Service and clinical assistant professor of psychiatry at Harbor-UCLA medical center when he wrote a paper on the process of disclosure of child sexual abuse ("The child sexual abuse accommodation syndrome", [1983] Child Abuse & Neglect 190). Hood states that Summit described this paper as "highly impressionistic". The paper is actually based on clinical experience "derived from the collective experience of dozens of sexual abuse treatment centers in dealing with thousands of reports or complaints of adult victimisation of young children". Such clinical experience contributes another lens onto our understanding of child sexual abuse.

Hood, continuing her unhelpful game of heroes and villains, introduces Dr Alfred Kinsey (p 52), as a "husband and father of irreproachable personal conservatism". Kmsey's biographers James Jones (Alfred C Kinsey: A Public/Private Life New York: WW Norton 1997) and Jonathan Gathorne-Hardy (Sex the Measure of All Things: A life of Alfred C. Kinsey Indiana University Press 2000) do not agree. According to Jones (p 512), Kinsey allegedly used paedophiles to bring children to orgasm and then record the number they could achieve: "Children gasping for breath, sobbing, screaming in pain, fainting, and desperately struggling to fight off the assailants Kinsey dignified as 'partners' - these were descriptions of hapless victims". While Kinsey was convinced the victims derived pleasure, Jones commented: "how reliable was the testimony of paedophiles about the children they molested? Not very."

Much attention is given to discrediting the work of Hood's arch-villain, Dr Miriam Saphira. The statistics from the 1970's Women's Weekly survey are ridiculed. Yet the key messages that "the man will usually know the child" and "one in five child molesters will be the father, stepfather or foster father", are not as far from academic research findings as Hood chooses to believe. Nor is the widely quoted "one out of four girls will be molested before she turns eighteen". Twenty-five per cent (one in four) is within the accepted range of prevalence data (Fergusson & Mullen, 1999; Finkelhor, 1994). While this may seem unbelievable, it includes one-off incidents at the lower end of the spectrum of sexual abuse. The University of Otago and Fergusson's study found that men known to the children perpetrate most child sexual abuse (Anderson et al. 1993 and Fergusson et al. 1996). In Fergusson's study, 23.5 per cent of the perpetrators were family members. In Anderson's study, 38.3 per cent were family members and 16 per cent of abusers were relatives living with the child. In 18.5 per cent of children's allegations of sexual abuse reported to Child, Youth and Family in 1999, the alleged abuser was a father or a stepfather (Basher, "Children who talk on tape: The facts behind the pictures", (1999) 12 Social Work Now 12). Stepfathers are at a higher risk of being perpetrators than biological fathers (Fergusson & Mullen, 1999 p 51).

Hood writes (p 95): "To encourage the child protection movement to close ranks, the Geddis-led National Advisory Committee on the Prevention of Child Abuse recommended that all abuse allegations be investigated by "specialised personnel, trained and working together". The literature is consistent in the importance of the use of multi-disciplinary trained teams, not to close ranks, as Hood puts it, but to improve practice. Each discipline has different knowledge bases and skills to contribute to the investigative process (Myers, "Keep the lifeboat afloat", (2002) 26 Child Abuse & Neglect 561). Such teams help to increase professionals' and agencies' accountability for their practice, by being more open to the scrutiny of their colleagues from other agencies, the opposite of what Hood proposes.

While glamorising the good old days, Hood is equally misleading in her statement that the passing of the Evidence Amendment Act 1989 meant that (p 113): "child complainants no longer had to appear in Court in person. They were no longer required to deliver their evidence-in- chief without leading or coaxing. Their evidence was no longer tested by unhampered cross-examination. Instead juries viewed pre-recorded interviews with child complainants conducted by interviewers trained in the beliefs and methods of the child protection movement, and observed cross-examinations conducted via videolink".

Hood fails to mention the advantages of videotaped evidential interviews of children for the Court. Specialised interviewers are under scrutiny. If they ask poor questions, those questions are there for all to see and hear in the courtroom. Videotapes also provide a record of what the child said much nearer to the disclosure than the trial, which may be more than a year later. They also enable the jury to see the child complainant on tape at an age nearer to when the events allegedly took place. In one or two years, children can look quite different. As children are less stressed, their testimony is likely to be clearer (See Cashmore, Children's evidence research paper 1: The use of closed-circuit television for child witnesses in the ACT. Sydney: Australia Law Reform Commission, 1992; Davies & Noon. An evaluation of the live link for child witnesses. London: Home Office 1991; and Murray, Live Television Link: An evaluation of its use by child witnesses in Scottish criminal trials. The Scottish Office 1995).

There is no evidence to substantiate that lawyers are more skilled at questioning children than specialised interviewers. There is some evidence to the contrary. One of us (ED) developed a coding system with Fred Seymour for analysis of the types of questions asked of children in Court. Having checked the coding system for inter-rater reliability, we analysed transcripts of the evidence-in-chief and cross-examination of child complainants of sexual abuse, including transcripts of evidential interviews of the younger of these complainants ("Questioning child complainants of sexual abuse: Analysis of criminal Court transcripts in New Zealand", (1998) 5 Psychiatry, Psychology & Law 47). We found that evidential interviewers used significantly more careful neutral (open, non-loaded) questions than prosecutors, who used significantly more neutral questions than defence lawyers. Evidential interviewers also asked significantly more age- appropriate questions than lawyers. Children being asked questions that they do not understand is not in the interests of those seeking the truth.

Children who give evidence-in-chief via videotape are still cross-examined live in austere Court buildings, even if it is via closed-circuit television (CCTV). Hood again omits the advantages for the Court. There is some evidence that children appear to be more comprehensive, more forthcoming, louder, more resistant to leading questions and more fluent when giving evidence on closed circuit television compared to open Court (Davies & Westcott 1995, "The child witness in the Courtroom: Empowerment or protection?" In Zaragoza Graham Hall Hirschman Ben- Porath (Eds) Memory and testimony in the child witness (pp 199-213). Thousand Oaks CA: Sage). Lawyers in the UK and Australia have been observed to use more age- appropriate questions through CCTV than in open Court (see Cashmore). There is no evidence that the use of CCTV increases the chances of conviction (See Tobey et al. 1995, "Balancing the rights of children and defendants: Effects of closed-circuit television on children's accuracy and jurors' perceptions." in Zaragoza et al; and Ross et al. 1994, "The impact of protective shields and videotape testimony on conviction rates in a simulated trial of child sexual abuse", (1994) 18 Law & Human Behaviour 553). If by "unhampered cross-examination" of a child, Hood means a flurry of juxtapositions and multi-faceted questions designed to confuse, then this might be so, although this seems contrary to the interests of justice.

An abridged version of this article was published in The New Zealand Herald (11 August 2003). Hood's response (13 August 2003): "As befits supporters of recovered memory therapy (now known euphemistically as abuse- focused therapy), Jeffrey Masson and Emma Davies rely on statistics that have been either discredited or misrepresented to support their otherwise insupportable beliefs. Nonetheless, it is pleasing to see that some members of the sex-abuse industry have been reading A City Possessed."

Which statistics have been "discredited" or "misrepresented" by whom and where? The term "sex abuse industry" is a meaningless derogatory term used to denigrate those who research or work within the child protection system or with survivors of abuse. What did recovered memory have to do with it? Our critique did not mention it. Besides, one of us (JM) has actually written a book Against Therapy, in which all forms of professionalised therapy are criticised. The other (ED) has never been a therapist nor has she ever made a public statement on recovered memory.

Hood's response to our critique reinforces our view that she is untrustworthy in her analysis of the issues involved in child sexual abuse. In claiming that "recovered memory therapy" is the same as "abuse focused therapy", Hood once again misconstrues the literature (Seymour, 1996). In her book, Hood boldly states that cognitive psychologist Elizabeth Loftus wins "hands down" the scientific debate about whether recovered memory exists (p 55). Three prominent cognitive psychologists have recently reached a more measured conclusion (Sivers Schooler Freyd, "Recovered memories", In Ramachandran. (2002) 4 Encyclopedia of the Human Brain 169). Hood omits that the false memory in Loftus's infamous analogue study was not for abuse. It was about being lost in the mall when one was a small child (Loftus & Pickrell, "The formation of false memories", (1995) 25 Psychiatric Annals 720). Analogue studies like these can never replicate the conditions of abuse. Nonetheless, a considerable body of experimental research has shown that young children can provide accurate recall of salient events (Eisen, Quas & Goodman 2002, Memory and suggestibility in the forensic interview, New Jersey: LEA Publishers).

Since Hood mentions recovered memory, we will briefly clarify our understanding of the implications of research, although we are at a loss to see its relevance to young children. A well-designed prospective study by Linda Williams showed that some sexually abused adults have periods in their lives when they do not remember being abused ("Recall of childhood trauma: A prospective study of women's memories of child sexual abuse", (1994) 62 Jo of Consulting and Clinical Psychology 1167). Even Loftus herself believes "it is possible to lose contact with memories for long periods of time". She merely disputes that the correct term to describe this is "repression" (Ceci & Loftus 1994, "A royal road to false memories?" quoted in Seymour 1996).

With regard to therapy, cognitive experimental psychologists, Read and Lindsay, are quite clear in their concerns about specific techniques, for example hypnosis ("Moving toward a middle ground on the false memory debate" (1994) 8 Applied Cognitive Psychology 407). They clarify: "there is little reason to fear that a few suggestive questions will lead psychotherapy clients to conjure up vivid and compelling illusory memory of childhood sexual abuse". And "not all or even most memories recovered in therapy are false ... our comments focused exclusively on approaches that make extensive use of suggestive memory recovery techniques". These techniques are appropriately treated with caution in the guidelines of professional psychological associations. (Seymour, 1996)

Hood's supposedly balanced book has been used to steer public attention to one case, from which we are encouraged to draw all sorts of spurious generalisations. New Zealanders concerned with a fair and reasonable approach to child abuse need to maintain a broader focus. Where is the public demand for the outcomes from the work based on the Ministerial Inquiry into Child Protection in 2000? Where is the parallel debate about how to prevent child sexual abuse? How can careful interviewing practice be strengthened? How can young children's safe access to the criminal justice system be ensured?

The standard of scholarship of the first three chapters of Lynley Hood's book is poor. At the beginning of chapter 4 it gets no better: "The flatness of the city makes it easy for anyone with a bright idea to gather together enough like- minded people to turn any theory - be it dazzingly enlightened or downright flaky - into action. And for more than 100 years that is exactly what Christchurch people have been doing". As Brian Edwards asked on TV1 (15 August 2003) don't you see an element of absurdity in what you're telling me here?" Hood: "No".

Her ability to stereotype, distort and denigrate groups of people and bodies of scientific research undermines her credibility. Her support of the recent publication of carefully selected excerpts of children's testimony is consistent with her carefully selected version of reality evident in the first few chapters of her book. Her tendency to vilify those she doesn't respect as members of the "sex-abuse industry", "zealots", "lesbian radical feminists", or inhabitants of a particular city, seems to mirror the witch-hunt she appears to have sought and found. Her analysis in the foundation chapters is more akin to story telling than the presentation of evidence. We suggest that those who signed the petition on the basis of Hood's book might wish to think again


New Zealand Law Journal
November 2003.
[2003] NZLJ (395)

The Christchurch Creche case
by Ross Francis,

"Reasoned and accurate" was my first impression of Emma Davies and Jeffrey Masson's critique of Lynley Hood's A City Possessed. Hood inexplicably uses dated research and her sources are somewhat suspect. But as I continued reading their critique, especially following Hood's reply to them, I was struck by their choice of sources and their apparent bias.

For example, they say that the term "sex abuse industry" is derogatory and meaningless. In fact it is a legitimate term that applies to psychologists and counsellors who profit from allegations of sexual abuse without verifying or corroborating such allegations. It also applies to so-called victims who unearth memories of sexual abuse for which there is little or no scientific basis.

I have no statistics regarding the size of the sex abuse industry in this country. However, in 2002/03, the Accident Compensation Corporation had 4894 new sensitive claims and 7351 ongoing sensitive claims. Sensitive claims "are where the claimant suffered an injury as a result of being the victim of specific types of criminal behaviour...[t]hey relate mostly to sexual crimes...[c]onviction does not have to result" (ACC Injury Statistics 2003, para 13.1). The cost of sensitive claims in 2002/03 was $22.048 million or $1800 per claim. Interestingly, ACC says that conviction is unnecessary but that claims are the result of criminal behaviour. If there has been no conviction, how can ACC be sure that any criminal behaviour has occurred?

Davies and Masson can barely conceal their disdain for renowned memory expert and psychologist Elizabeth Loftus. Three psychologists cited by Davies and Masson are referred to as "prominent". No so Loftus. Loftus is a former member of the American Psychological Association. Interestingly, she resigned from the APA and joined the American Psychological Society - she was its President between 1998/99 - because she believed the latter valued science more highly and more consistently. She is a Distinguished Professor at the University of California, Irvine. She has published 20 books and more than 350 articles. She has received four honorary doctorates for her research. In 2002 the Review of General Psychology listed Loftus among the 100 most eminent psychologists of the 20th century and she was ranked the highest woman. She has appeared as an expert witness in many criminal trials. Given her credentials, and those of Davies and Masson, the authors' attitude towards her seems to be, understandably, one of envy.

Lynley Hood, say D-M, "omits that the false memory in Loftus's infamous analogue study was not for abuse". Hood simply states that Loftus' research shows that children can have false memories of traumatic events (assuming that being lost in a mall is a traumatic event). Why do D-M have a problem with this statement? Curiously, D-M refer to Loftus' study as "infamous". Why? It cannot be infamous on the basis of its conclusions because they have been replicated by many other studies. Maybe it is infamous because of its implications for cases of alleged child sexual abuse (see the case of "Nathan" below). If so, I think ground-breaking might be a more accurate term.

Loftus generally doubts the existence of recovered memory, especially where traumatic events are concerned. Davies and Masson argue that Sivers et al "have reached a more measured conclusion". Their more measured conclusion is simply one of hope, that in the future more can be learned about recovered memory. This ignores the fact that much is already known about recovered memory. Recovered memories of childhood sexual abuse should be able to be verified and falsified. Not to mention corroborated. How many are?

D-M then go on to quote two more psychologists, Read and Lindsay, who say that there is little reason to think that suggestive questions by mental health experts will lead to false memories of childhood sexual abuse. Has this been tested? Corroborated? Verified? What of the millions of dollars that US Courts have awarded victims of false memories resulting from lawsuits against their therapists? How have these victims acquired their false memories if not through suggestive questioning?

Recently, "Nathan", who is in his early 20s, spoke of being abused by Peter Ellis during his time at the Civic Creche. Nathan was four at the time of his alleged abuse. Nathan says he knows he was abused because he was different from other teenagers. At age 14 or 15 he would "freak out" if a girl tried to touch or kiss him. He also felt uncomfortable changing clothes in front of other boys. He discovered that he had been abused when he was 14 and told his parents when he was 16. He complained to the police last year. Why did it take him ten years to remember the abuse and what caused him to remember it?

According to his mother, Nathan saw Ellis on television after Ellis had been charged with sexual abuse at the Civic Crèche. Afterwards, she asked Nathan if anything had "happened" to him when he was at the crèche. He replied "no". His mother, however, believed that something had happened to him. Consequently, she contacted police soon after Ellis was charged in 1992. Nathan did not like the Civic Creche. Maybe his mother questioned him to see if he had been sexually abused. She possibly had concerns based on his dislike of the crèche and the fact that she believed he was there when Peter Ellis worked there. Maybe Nathan came to believe he had been abused because he wanted to please his mother. She had, after all, removed him from the Civic. Maybe he thought she would remove him from his new crèche, which he liked, if he didn't tell her what she wanted to hear. It seems as plausible an hypothesis as the one put forward by Nathan.

According to Nathan's mother, Nathan was abused in late-1985. According to Peter Ellis' supervisor, Dora Reinfeld, Ellis was not at the Civic Creche in any capacity until late-August 1986. The police have decided not to charge Ellis regarding Nathan's allegations. Given his strongly held belief that he was sexually abused, charging Nathan with laying a false complaint would seem inappropriate.

Davies and Masson say that A City Possessed "has been used to steer public attention to one case, from which we are encouraged to draw all sorts of spurious generalisations". What spurious generalisations are we encouraged to draw? That there are few, if any, specific behavioural indicators of child sexual abuse? That children can, if exposed to suggestive and leading questioning, recall events that did not occur? That multiple interviews of child witnesses can produce unreliable testimony? That experts can and do get it horribly wrong? These aren't generalisations; these are facts! D-M should be grateful that Lynley Hood has provided the public with an insight into child sexual abuse and the methods used to evaluate it.

If Davies and Masson had read the remaining thirteen chapters, and not just the first three, of A City Possessed, they would be able to comment on the guilt or innocence of Peter Ellis. They have certainly had no difficulty passing comment on Lynley Hood. "Untrustworthy", they call her. That is a term that could just as easily be applied to them. They have made no attempt to verify the allegations of child sexual abuse at the Civic Creche. Yet the authors profess their deep concern for children. "Where is the parallel debate about how to prevent child sexual abuse?" they ask rhetorically. Before preventing child sexual abuse, one must first be able to prove that it has occurred. Since Peter Ellis was first charged, nobody has been able to do


New Zealand Law Journal
October, 2003
pages 389-392

In response to Emma Davies and Jeffrey Masson
Felicity Goodyear-Smith, MBCHB FRNZCGP MGP

Felicity Goodyear-Smith, the University of Auckland responds to criticism of A City Possessed at [2003] NZLJ 315

Both Hood and the Masson and Davies article rightly agree that “Just how widespread is child sexual abuse, and just how serious are the consequences?” are crucial questions.

Hood reviews some of the studies that have assessed these questions, with wildly varying results. She concludes “All estimates of the prevalence of child sexual abuse are contaminated by unknown and unknowable levels of under-reporting and over-reporting” and “child sexual abuse happens. It may or may not cause lasting harm. The long-term effects are difficult to determine because traumatic sexual abuse usually occurs in association with other detrimental childhood factors”. Masson and Davies take issue with her conclusion that “we don’t know” the true prevalence of child sexual abuse nor the chance of long-term harm from such abuse.

They refer first to David Finkelhor’s review of prevalence studies (Finkelhor, 1994) which found prevalence rates for women between 7 per cent and 36 per cent. In fact, this variance high-lights the problem of estimating how often child sexual abuse (CSA) occurs. Comparing studies is problematic, because studies use no standard definition of ‘child’ (may be under the age of 12, 14 or 18) ‘sex’ (for example, some studies might include a hug or kiss from a relative that a child decided was ‘yukky’, even if the adult’s intent was not sexual) or ‘abuse’ (may involve only acts of physical contact, or may include non-contact abuse such as witnessing ‘flashing’ or viewing pornography) and many do not define ‘child sex abuse’ at all.

Furthermore, use of Finkelhor as an authority on prevalence is surprising. Finkelhor was lead author of a book on sexual abuse in child care centres (Finkelhor et al, Nursery Crimes, 1988). This book identifies the major part the Californian McMartin Preschool case played in ‘uncovering’ other cases of sexual abuse in day care centres throughout the 1980s, with two-thirds of victims being ritual-abuse cases. Finkelhor classifies the McMartin case as “substantiated”, involving more than 300 victimised children and multiple abusers engaged in systematic ritualistic bizarre abuse of children. In fact, exhaustive investigations revealed no evidence to substantiate any of the charges, apart from the testimony of children who had undergone extensive ‘disclosure interviews’. Analysis of the interview videotapes demonstrates that the repeated highly suggestive questioning, combined with rewards (stickers, food, praise) for ‘disclosures’ and punishment (calling them liars or preventing them leaving until they ‘told’) were major contributing factors in the false reports eventually elicited from the children (Garven et al 1998). A record-breaking $15,000,000 trial eventually resulted in all defendants acquitted.

The McMartin Preschool and the Christchurch Civic Crèche cases have remarkable parallels. Both started with a child attending the centre presenting with a problem considered by a parent to be an indicator of sexual abuse. There followed meetings of anxious parents and intense involvement of social workers, counsellors and the police. In both cases, none of the children interviewed made any disclosures of abuse. The children’s denials were not believed. It was only after weeks of repeated interviews that some of the children began to talk about abuse, with allegations that became increasingly extreme and bizarre as interviews progressed. In both cases, allegations progressed to involving the child care workers subjecting children to rape, anal, oral and group sex; making children drink urine and eat faeces; naked circle games; placing children in cages, tunnels or burying them in boxes, and making child pornography. In both cases a parent or parents believed that ritualistic cults were operating in the centre. A parent of one of the key complainants in the Christchurch case requested the police bring Pamela Hudson, an American ‘satanic ritual abuse’ expert, to New Zealand to assist in the inquiry (NZPA, Dec 1992),. She said her son had described incidents involving all sixteen of Hudson’s “indicators” of satanic abuse, including being suspended in cages, participating in sacrifices, and ‘circle’ rituals. (McLoughlin, 1996) Extensive investigations of over 12,000 reported accusations of satanic ritual abuse in the USA (Lanning, 1991), (Goodman et al, 1994) and similar investigations in Britain (La Fontaine, 1994) have found no substantiation of the occurrence of such ritualised abuse.

The difference between the McMartin and the Christchurch Crèche cases is that in the former, all the accused were eventually cleared by the courts of any abuse.

For prevalence figures, Masson and Davies also rely on Diana Russell’s findings that “28 per cent of the 940 women reported at least one experience of sexual abuse within the family before the age of 18”. They describe Russell’s results and methodology as “compelling”. However this 1978 study of 930 San Francisco women has significant flaws. Subjects were accessed using a public opinion-poll organisation providing a “probability sample of households” without the randomisation method being described in the paper. (Russell, 1983) Door-knocking interviewers recorded a 50 per cent refusal rate. Data are based on retrospective self-reporting of consenting women during lengthy interviews. Sixteen percent reported experiences of intrafamilial abuse ranging from “unwanted but non-forceful kissing” to “forcible rape” under age 18, with 3.7 per cent reporting “very serious” abuse as described by Masson and Davies. The significant study weaknesses limit the generalisability of such data to the general New Zealand public.

Data from David Fergusson’s Christchurch Health and Development study, which has followed a cohort of over 1000 children born in Christchurch in 1977, is certainly more reliable and representative information relevant to the New Zealand population. At age 18, 10.4 per cent of the 1019 subjects reported unwanted sexual experiences prior to sixteen, ranging from leering and suggestive comments to intercourse. (Fergusson et al, 1996). Family members were perpetrators in 23.5 per cent of cases (2.5 per cent of the cohort).

Masson and Davies use Fergusson’s data to support their claim that Miriam Saphira’s figures “are not so far from academic figures as Hood’s book would lead us believe”. The much-publicised Telethon claim “One in four New Zealand girls are sexually abused before they turn 18. Half of them by their own father” (Telethon advertisement, 1988) was based on figures promoted by Miriam Saphira (Saphira, 1981)

However, while Fergusson found that 23.5 per cent of cases were perpetrated by family members, overwhelming these were not natural parents, who were only 1.5 per cent of perpetrators. Two of the 1019 (<0.2 per cent) were sexually abused by a natural parent, and seven ( <0.7 per cent) by step-parents. These data suggest perhaps one in 500 children sexually abused by a natural father, not the one in eight suggested by the Telethon advertisement.  [Note A]
Masson and Davies’ analysis on the effects of CSA is similarly problematic. They claim that “Child sexual abuse has been linked to suicide, depression, psychiatric admissions, drug abuse and eating disorders”, failing to appreciate that association does not mean causation. Because two things are associated, this does not mean that one causes the other. Specifically if there are two phenomena, ‘A’ and ‘B’, they may be related in four different ways: A is not associated with B in any way; A causes B; B causes A; or both A and B are caused (or predisposed to) by some independent factor, C (Pope & Hudson, 1995)

As an example of the latter, when acquired immunodeficiency syndrome (AIDS) was identified in the 1980s, it was more likely to be found in homosexual men who used inhaled nitrates. Investigators wondered whether nitrate use caused AIDS. However subsequent research discovered that AIDS was caused by infection with the human immunodeficiency virus (HIV), and that homosexual men who used nitrates were also more likely to engaged in certain risky sexual behaviours (unprotected anal intercourse). The association between nitrite use (A) and AIDS (B) was therefore not a causal relationship but due to a confounding factor (C) - specific sexual practices (Vandenbroucke & Pardoel, 1989)

Masson and Davies correctly identify that there is a huge body of literature “on the sequelae of child sexual abuse”. However very little of this is rigorously conducted epidemiological research. Many studies have serious sampling biases and other methodological flaws. This topic has previously been addressed in the NZLJ (Goodyear-Smith, 1999)

For example, the scientific literature reveals no causal association between CSA and eating disorders. Neither controlled nor uncontrolled studies of bulimia nervosa found higher rates of CSA than those found in studies of the general population using comparable methods. Current evidence does not support the hypothesis that CSA is a risk factor for bulimia nervosa or other eating disorders (Pope, 1994); (Pope & Hudson, 1992); (Vogeltanz-Holm, 2000). A recent meta-analysis (synthesising and statistical pooling studies) concluded that CSA did not emerge as one of the possible risk factors for developing an eating disorder (Stice, 2002)

A meta-analysis of non-clinical population studies concludes that CSA has no inevitable adverse outcomes. Although most studies show that, on average, young adults with CSA are less adjusted than their peers, the statistical explanation is overwhelmingly due, not to the CSA, but to other negative aspects of their family environment (explaining nine times the amount of variance than CSA) (Rind & Tromovitch, 1997). In other words, in most cases psychological problems in adulthood can be attributed to other confounding adverse factors, not sexual abuse.

This is in line with New Zealand studies. The Otago Women’s Health Survey Child Sexual Abuse study found that in most cases, negative effects of CSA could be explained by the family and social context from which it emerged, although the most severely abused showed an increased risk of psychopathology even when they came from advantaged backgrounds. The effect of CSA does appear to be dose-dependant, with increased risk of long-term sequelae with very severe forms of sexual assault. (Mullen, 1993); (Mullen et al, 1993). Women reporting CSA did have a higher rate of psychiatric disorders, especially depression, as adults compared to a non-abused population, but the researchers warn against “overestimating the contribution of childhood abuse in general, and sexual abuse in particular, to explaining mental health and interpersonal difficulties in our community”. They concluded that the contribution of any form of childhood abuse to the variables they studied (psychopathology, sexual difficulties, decreased self- esteem or interpersonal problems) was extremely modest, only 1-5 per cent of the cases. (Mullen, 1996) Those people predominantly came from disadvantaged homes (most subjects reporting CSA who had developed problems had concomitant histories of family violence or emotional neglect). In many cases the apparent association between CSA and adult problems “was accounted for by this matrix of childhood disadvantage from which abuse so often emerged” (Mullen, 1996).

Fergusson similarly found that those reporting CSA had an increased rate of depression, anxiety, substance abuse, disturbed behaviour or attempted suicide than those not reporting sexual abuse, but after adjusting for the  confounding factor of family dysfunction, a maximum of 10-20 per cent of the risk of psychiatric disorder in young adults could be accounted for by exposure to CSA (Fergusson et al, 1996)

While Masson and Davies criticise Hood for not using the newest publications available, they appear similarly selective in the research they quote. Instead of referring to the more recent meta-analysis (Rind & Tromovitch, 1997) and findings from the Otago (Mullen, 1996) and Christchurch studies (Fergusson et al, 1996); (Fergusson et al, 1996), they rely on a 1993 review. This paper is based on clinical and legal cases with inherent sample biases and therefore cannot be assumed representative of the general population (Kendall-Tackett et al, 1993). Again, this reference is the work of Finkelhor and Williams, and includes child care cases with alleged satanic ritual abuse.

When ascribing causality, the association between the causal experience and the putative results must be obvious and large. It appears that just as a history of CSA is not necessary for the development of any particular psychiatric or adjustment disorder as an adult, only a minority of people exposed to CSA develop any diagnosable problem (Laidlaw et al, 2001). CSA is not to be condoned: it is both illegal and morally repugnant. Fortunately, however, most children are resilient, while a few are vulnerable and develop a wide range of idiosyncratic and unpredictable psychological problems. There are no reliable ‘behavioural indicators’ of CSA.

Masson and Davies berate Hood for challenging Summit’s paper on the process of disclosure of child sexual abuse, the Child Sexual Abuse Accommodation Syndrome (CSAAS) (Summit, 1983). Summit claimed that sexually abused children typically exhibited this syndrome, with the helplessness and secrecy about their abuse resulting in children ‘accommodating’ with delayed, conflicting and unconvincing disclosures and later retractions.

This led to children’s denial or recanting being considered as evidence of abuse. A ‘heads I win, tails you lose’ approach, this means that both a child’s admission and denial prove that abuse has occurred. In 1992 Summit published that CSAAS was based solely on his clinical impressions not research-based; that it should not be used in diagnosis; that it was not a syndrome but rather a pattern he had observed, and it that should not be used in court. (Summit, 1992) Despite this caveat, CSAAS was frequently used in courtrooms to explain why children believed to have been abused would deny it. This includes its implicit use in the Christchurch Crèche case. Masson and Davies appear unaware that Summit himself warned against courtroom use of CSAAS.

The ‘recovered memory’ phenomenon was based on the same premises of high prevalence. Problems in adult women (such as eating disorders, depression, relationship difficulties to psychosis) were thought likely to have been caused by CSA. In the late 1980s and 1990s it was believed many women had ‘repressed’ memories of these events and a variety of hypnosis-like techniques (for example, guided imagery) were used to help women recover their memories in therapy. A book by Masson lent weight to this theory (Masson, 1984). During the early development of his ‘talking therapy’, Freud wrote that almost all his adult female patients told him about sexual abuse by their fathers. Freud initially accepted the veracity of these stories, but later concluded that they were unconscious fantasies generated under hypnosis. Masson’s book claimed that Freud had recanted under pressure and that his initial statements regarding incest as the cause of his clients’ problems were correct. Of course, there is no way to know which of Freud's patients (none, some or all) were victims of incest or were victims of pseudo-memories generated through hypnosis. However Masson's book was used by promoters of recovered memory therapy to support their theory. The ‘bible’ of the recovered memory movement, the self-help manual ‘Courage to heal’ (Bass & Davis, 1988) used Masson’s explanation of how “Freud constructed the Oedipal theory as a cover-up for the truth of child sexual abuse” to support their views. ‘Courage to heal’ was followed by other books and workshops promulgating these beliefs.

While Davies may not have made a “public statement on recovered memory”, she, along with her partner John Read, and colleague Fred Seymour, wrote in response to a TVNZ programme that most recovered memories “do not occur within counselling” (Davies, Read & Seymour, 1997). Davies, Read and Seymour have also published that CSA “occurs with one in four girls” and give their opinion that the majority of children usually deny their abuse and will only describe it “over a number of interviews. ..full disclosure of sexual abuse is frequently a lengthy process, not a one-off event” (Davies, Read & Seymour, 1996)

In support of ‘recovered memory’ theory, Masson and Davies write “A well-designed prospective study showed that some sexually abused adults have periods in their lives when they do not remember being abused” (Williams, 1994). By Linda Williams, co-author with Finkelhor of publications quoted above, this study is commonly quoted as the best evidence supporting memory repression of CSA. Of 129 Afro-American women interviewed about CSA that had been investigated at a hospital emergency room 17 years previously, 38 per cent did not tell the interviewer about the alleged abuse. However, this study measures ‘not recalling’ rather than repression. There are many possible explanations for their non-recall other than repression: the reported abuse happened when they were too young to remember (over half were aged 0-6 years; 12 per cent under three years old); simple forgetting; choosing not to talk about the incident; or the original allegation was false (for a detailed critique see Goodyear-Smith, 1998).

While Masson and Davies appear to adhere to the concept of memory recovery and repression, there is considerable evidence that many cases emerged within the context of therapy. There is a huge body of scientific research demonstrating that false ‘memories’ can be generated through inappropriate use of hypnosis-like techniques. The dramatic reduction in ‘recovered memory’ cases from the 1990s to the 2000s supports the hypothesis that memory repression and ‘recovery’ through therapy was a phenomenon generated by these misguided beliefs and practices.

I agree with Masson and Davies that “a considerable body of experimental research has shown that young children can provide accurate recall of salient events” (Ceci & Bruck, 1993). However, there is also a considerable body of research, particularly by these authors, demonstrating that children can be led to make false reports they come to believe are true. Pre-school children are especially vulnerable to effects of suggestive and repeated questioning (Bruck & Ceci, 1999). They may internalise the false memories and resist debriefing. “Children's false statements can be convincing to professionals, who are unable to distinguish between the children's accurate and inaccurate narratives” (Ceci & Huffman, 1997).

Hood highlights the impact of the Evidence Amendment Act 1989 resulting in child complainants no longer required to appear in court in person. Masson and Davies outline their views on why videolink or screening is preferable. However facing your accuser in the courtroom has been one of the basic tenets of justice. There is no research to show that abused children are harmed by viewing the offender guarded in the dock from the safety of the witness box. Conversely, for non-abused children who are making false reports, seeing the real person may act as a reality check against the fantasised monster they have constructed in their heads.

The Christchurch Crèche case was not an isolated aberration. It happened because of the prevailing belief in a grossly inflated prevalence of CSA, coupled with a conviction that any behaviour problems in childhood, or psychological disturbances in adult women, are likely to be caused by CSA. Lists of wide-ranging ‘behavioural indicators’ of CSA (such as bed-wetting, nightmares, shyness, tantrums) were extensively promoted by New Zealand social workers, psychologists and doctors specialising in CSA in the 1980s and 1990s. Common teaching was that children are usually too scared to tell about CSA and may only disclose within the safety of an interview. Once suspicion of abuse had been aroused, children might be questioned many times, both informally by parents and formally by interviewers, before any allegations emerged. Both professionals and parents were instructed to always believe a child and praise them for telling about abuse. It was not acceptable to challenge a child on the veracity of a claim, no matter how improbable or even impossible it was that an alleged event could have occurred.

Masson and Davies are correct that estimates of the prevalence and effects of CSA are pivotal influences in the Crèche case. Hood has accurately identified that gross inflation of the extent and effects of CSA was a driving force in this case. Masson and Davies accuse Hood of minimising these figures; however the references they use to shore up this accusation are seriously flawed.

Challenging exaggerated statistics is not a denial of the reality of CSA and its potential harm. We all agree that CSA occurs; it is illegal and reprehensible, and that sufferers may develop subsequent psychological problems on occasion.

While Masson and Davies state “we cannot comment on the guilt or innocence of Peter Ellis”, Davies previously has not been so reluctant. In 1997 she stated in a press release complaining about a television programme on the Ellis case “One can only wonder what profound effect this biased programme has had on the families of the children abused by Ellis”, clearly expressing her view that Ellis is guilty (Davies, 1997)

Advocacy for the victim is another key element in promoting false reports. In the 1990s social workers, evidential interviewers, forensic doctors and the police were taught they must treat every sexual allegation as genuine. They should minimise distressing a complainant by avoiding critical examination of her testimony (Goodyear-Smith, 1996). A belief that false allegations do not occur, and that the role of a Crown expert witness is to be advocate for complainant, deemed ‘victim’ prior to the trial, is inconsistent with the impartiality demanded by the forensic role. However, the manual for doctors performing forensic CSA examinations instructed them to act as advocate for the child (Fancourt et al, 1994)

This perspective persists in the 2000s. Expert witness advocacy by Dr Zelas, key Crown psychiatric witness in the Crèche case, recently resulted in the Appeal Court quashing multiple child sex assault convictions (NZ Appeal Court, 2003).

The faulty assumptions underpinning investigation of sexual abuse cases and the systemic errors consequential to this, have resulted in cases of injustice not limited to the Christchurch Crèche case. This is why a Royal Commission of Inquiry is essential.


Bass E, Davis L. The Courage to heal: a guide for women survivors of sexual abuse. New York: Harper and Row; 1988

Bruck M, Ceci S. The suggestibility of children's memory. Annual Review of Psychology 1999;50:419-439

Ceci S, Bruck M. Suggestibility of the child witness: A historical review and synthesis. Psychological Bulletin 1993;113:403-439

Ceci S, Huffman M. How suggestible are preschool children? Cognitive and social factors. Journal of the American Academy of Child & Adolescent Psychiatry 1997;36:948-958

Davies E, Read J, Seymour, F (Psychology Department, Auckland University) Letter to the Editor, North & South, Oct 1996

Davies E, Read J, Seymour, F (Psychology Department, Auckland University) Letter to the Editor, Listener, 5 Apr 1997, p12

Davies, Emma, Researcher in Psychology and Law, University of Auckland and Spokesperson for the Children's Agenda. Press Release, Child Advocacy Trust 20 Nov 1997

Fancourt R, Shand C, Broadmore J, Milford R. The medical management of sexual abuse. 4th ed. Wellington: Doctors for Sexual Abuse Care; 1994, p82

Fergusson D, Lynskey M, Horwood L. Childhood sexual abuse and psychiatric disorder in young adulthood: I. Prevalence of sexual abuse and factors associated with sexual abuse. Journal of American Academy of Child & Adolescent Psychiatry 1996;35:1355-1364

Fergusson DM, Horwood LJ, Lynskey MT. Childhood sexual abuse and psychiatric disorder in young adulthood: II. Psychiatric outcomes of childhood sexual abuse. Journal of American Academy of Child & Adolescent Psychiatry 1996;35:1365-1374

Finkelhor D, Williams L, Burns N. Nursery crimes: sexual abuse in day care. Newbury Park: Sage Publications; 1988.

Finkelhor D. The international epidemiology of child sexual abuse. Child Abuse & Neglect. 1994;18:409-17

Garven S, Wood JM, Malpass RS, Shaw JS, 3rd. More than suggestion: the effect of interviewing techniques from the McMartin Preschool case. Journal of Applied Psychology. 1998;83:347-59

Goodman G, Qin J, Bottoms B, Shaver P. Characteristics and sources of allegations of ritualistic satanic abuse, Report to the National Center on Child Abuse and Neglect. Washington DC; 1994

Goodyear-Smith F. Victim-orientated law reforms: advantages and pitfalls. Issues in Child Abuse Accusations 1996;8:87-93

Goodyear-Smith FA, Laidlaw TM, Large RG. In praise of scholarship: A reply to McCullough. Health Care Analysis 1998; 6:223-226

Goodyear-Smith F. Childhood sexual abuse and mental health. NZ Law Journal 1999;October:386-388

Kendall-Tackett KA, Williams LM, Finkelhor D. Impact of sexual abuse on children: A review and synthesis of recent empirical studies. Psychological Bulletin 1993;113:164-180

La Fontaine J. The extent and nature of organised and ritual abuse. London: HMSO; 1994

Laidlaw T, Goodyear-Smith F, Gorman D. Sexual abuse counselling: what is the rationale?, Nuance 2001;2: 64-86

Lanning K. Ritual abuse; a law enforcement view or perspective. Child Abuse and Neglect 1991;15:171-5

Masson J. Assault on Truth: Freud's Suppression of the Seduction Theory, Penguin Books; 1984

McLoughlin, D (Aug 1996). ‘Second thoughts on the Christchurch Civic Crèche case: has justice failed Peter Ellis?’, North & South, 55-69

Mullen PE. Child sexual abuse and adult mental health: The development of disorder. Journal Interpersonal Violence 1993;8:429-432

Mullen et al. Childhood sexual abuse and mental health in adult life. British Journal of Psychiatry 1993;163:721-732

Mullen PE et al. The long-term impact of the physical, emotional, and sexual abuse of children: A community study. Child Abuse & Neglect 1996;20:7-21

NZ Appeal Court, CA 051/03, 25 Jun 2003, Anderson, Panckhurst, Patterson JJ

NZPA, Mother tells court of plea for expert’s help, NZ Herald, 4 Dec 1992

Pope HG, Jr., Hudson JI. Is childhood sexual abuse a risk factor for bulimia nervosa? American Journal of Psychiatry. 1992;149:455-63

Pope HG, Jr et al. Childhood sexual abuse and bulimia nervosa: a comparison of American, Austrian, and Brazilian women. American Journal of Psychiatry. 1994; 151:732-7

Pope HG, Hudson JI. Does childhood sexual abuse cause adult psychiatric disorders? Essentials of methodology. Journal of Psychiatry & Law 1995; 23:363-381

Rind B, Tromovitch P. A meta-analytic review of findings from national samples on psychological correlates of child sexual abuse. Journal of Sex Research 1997;34:237-255

Russell D. The incidence and prevalence of intrafamilial sexual abuse of female children. Child Abuse and Neglect 1983;7:133-176

Saphira M. The sexual abuse of children. Auckland: Papers Inc; 1981

Stice E. Risk and maintenance factors for eating pathology: a meta-analytic review. Psychological Bulletin. 2002;128:825-48

Summit R. The child sexual abuse accommodation syndrome. Child Abuse & Neglect 1983;7:177-193

Summit R. Abuse of the Child Sexual Abuse Accommodation Syndrome. Journal of Child Sexual Abuse 1992;1:153-163

Telethon advertisement, ‘It’s not the dark she’s afraid of’, Auckland Sun, 18 Jun 1988, p27

Vandenbroucke JP, Pardoel VP. An autopsy of epidemiologic methods: the case of "poppers" in the early epidemic of the acquired immunodeficiency syndrome (AIDS). American Journal of Epidemiology. 1989;129:455-7

Vogeltanz-Holm N et al. Longitudinal predictors of binge eating, intense dieting, and weight concerns in a national sample of women. Behavior Therapy 2000;31:221-235

Williams L. Recall of childhood trauma: A prospective study of women's memories of child sexual abuse. Journal of Consulting and Clinical Psychology 1994;62:1167-1176


[Note A].   21 October 2003
The NZ Law Journal paper incorrectly stated "not the one in two suggested by the Telethon advertisement"
This error has been corrected at the request of the author, Felicity Goodyear-Smith