NZLII Home | Databases | WorldLII | Search | Feedback

New Zealand Yearbook of New Zealand Jurisprudence

You are here:  NZLII >> Databases >> New Zealand Yearbook of New Zealand Jurisprudence >> 2000 >> [2000] NZYbkNZJur 6

Database Search | Name Search | Recent Articles | Noteup | LawCite | Download | Help

Coxhead, Craig --- "Corrobating recovered memories" [2000] NZYbkNZJur 6; (2000) 4 Yearbook of New Zealand Jurisprudence 43

Last Updated: 11 April 2015

Corroborating Recovered Memories


1. Introduction

The debate on recovered memories reached its peak in 1995.' Six years on, with an emerging consensus from the psychological community, it is timely to reflect on how (or whether) the New Zealand legal system is adapting to the task of determining sexual abuse cases based on recovered memories.

It is impossible to encapsulate, in a paper of this length, the many facets of the recovered memory debate. The subject matter raises many different psychological and legal issues that are beyond the scope of this paper. The position of the writer in attempting to resolve some of the legal issues is based upon the following summary.

The reports of several professional organisations' in this area urge caution in dealing with recovered memories and accept that it is possible for false memories to be implanted (unwittingly or otherwise). However, the consensus of these organisations is that it is also possible for some people who were sexually abused as children, to have a period of amnesia for that abuse and for those memories to be triggered at some later point. To date, there is no evidence suggesting that recovered memories are any less reliable than continuous memories. While these views are not universally held, the writer endorses them in order to place parameters around the following discussion, and this position appears to represent the "middle ground" which seems a sensible way to approach the issue. As Lindsay points out:

There is no reason to doubt that ["memory work] techniques can enable people to recover accurate but long-forgotten memories of childhood traumas. Unfortunately...there are many reasons to believe that these same techniques

BA, LLB, MJur, Lecturer in Law, University of Waikato

In 1995 the author researched this topic for her Masters thesis. See Midson, B. "Determining the Truth in Recovered Memory Cases" (unpublished thesis, University of Waikato, 1996).

  1. For example, the American Psychological Association, British Psychological Soci
    ety, and the New Zealand Psychological Society.

can also enable nonabused clients to develop compelling but illusory memories and firmly held but false beliefs about C[hildhood] S[exual] A[buse].'

However, despite the existence of a middle ground approach, the New Zealand legal system is likely to be reluctant to deal with recovered memory cases, due to past controversy surrounding the issue. Yet the principle of access to justice demands that all complainants of sexual assault should have the opportunity to have their claims heard. A development in some United States jurisdictions has been to require some form of "independently verifiable objective evidence"^ before courts will consider suspending limitation statutes in "repressed memory" cases.' This paper will thus consider how the types of corroboration accepted by United States courts can assist the fact-finder in the context of New Zealand criminal trials.

Firstly this paper will discuss dissociative amnesia as a possible mechanism for causing the phenomenon of recovered memory. Secondly, the writer will review New Zealand recovered memory cases in order to appraise the current approach of the courts to this issue. Thirdly, the United States case of Moriarty v Garden Sanctuary Church of God will be assessed for its treatment of the corroboration requirement. Fourthly, the writer will consider how the different types of corroboration identified in Moriarty might assist the fact-finder in New Zealand criminal trials. Finally, the writer concludes that New Zealand courts must negotiate their way around the extreme views associated with this issue, and adopt an approach that ensures justice in each individual case.

2. Dissociative Amnesia

Dissociative amnesia is identified in the fourth edition of the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders (DSM — IV) as a clinical diagnosis that may be responsible for causing amnesia for traumatic events. The amnesia is reversible.

  1. Lindsay, D.S. "Contextualizing and Clarifying Criticisms of Memory Work in Psychotherapy" Pezdek, K & Banks, W.P. (eds) The Recovered Memory/False Memory Debate (1996) 270.

The term used in Moriarty v Garden Sanctuary Church of God 334 S.C. 150; 511 S.E.2d 699; 1999 S.C. App. LEXIS 9.

  1. See also Olsen v Hooley 865 P.2d 1345; 1993 Utah LEXIS 141; 224 Utah Adv. Rep. 41 and Meiers-Post v Schafer 170 Mich. App. 174; 427 N.W.2d 606; 1988 Mich. App. LEXIS 413.

6 Supra at n 4.

The essential feature of Dissociative Amnesia is an inability to recall important personal information, usually of a traumatic or stressful nature, that is too extensive to be explained by normal forgetfulness.'

The American False Memory Syndrome Foundation' argues that the identification of dissociative amnesia in DSM — IV does not mean that "repression" is a real phenomenon.' The Foundation's arguments include the fact that the DSM — IV itself notes that there is considerable controversy surrounding the recovered memory phenomenon, and that dissociative amnesia cannot be distinguished from malingering. The Foundation also points out that the DSM — IV was developed by committee, rather than from scientific evidence. It is the writer's contention that these arguments lack credibility. Firstly, dissociative amnesia and repression are two distinct concepts and the existence of one does not "prove" or "disprove" the existence of the other. Secondly, while many declaimers of recovered memory theory show a distrust of clinical data, on the basis that it is subjective and anecdotal (not validated by scientific research), recovered memory theory is impossible to prove by scientific research because, as Wylie points out, traumatic events cannot be staged with human subjects to prove the impact of trauma on memory.° Thirdly, while DSM — IV does state that there are no methods for distinguishing dissociative amnesia from malingering, they also add that individuals with dissociative amnesia often score high on standard measures of hypnotisability and dissociative capacity. They further state that malingering is more common in individuals who present with elaborate symptoms in contexts that are indicative of secondary gain (although true amnesia might also occur with such stressors)."

There are several mechanisms that may be responsible for the phenomenon whereby individuals are unable to remember incidents of childhood sexual abuse. However using dissociative amnesia as an explanatory model provides a framework for the legal system to adopt in adjudicating individual cases.

American Psychiatric Association Diagnostic and Statistical Manual of Mental Disorders (4th ed 1994) 478.

  1. The False Memory Syndrome Foundation is an advocacy group formed in 1992 in Philadelphia by an accused father, Peter Freyd, and his wife, Pamela.
  2. False Memory Syndrome Foundation Newsletter, FMS Foundation, October 1998 Vol 7 No. 8.

to Wylie, "Trauma and Memory" (1993) Family Therapy Networker 42.
it DSM-IV, supra at n 7 at 479-481.

3. New Zealand Case Law

Prior to 1995, the most flexible judicial approach to the recovered memory issue was demonstrated in R v R,12 where the court admitted expert evidence on a range of topics including the concepts of repression, dissociation and "false memory syndrome". The expert's testimony was admitted because it was relevant to the central issue of the credibility and reliability of the complainants' evidence.

A search of New Zealand case law since 1995 reveals that a small number of cases dealing with "recovered memory" have arisen in the courts. However, there are difficulties in drawing any useful comparison between these cases because each case deals with different legal issues, and courts (amongst others) use the psychological terminology inconsistently.

For example, what are often classified as cases of "recovered memory" are not. They tend to be cases where complainants" have always remembered the abuse, but not appreciated the effects such abuse has had on their life, or alternatively chosen not to think about the abuse for a long period of time. To classify these as cases of"recovered memory" does complainants a disservice as there is a tendency by the courts to see evidence based on a recovered memory as unreliable.

For instance, in R v Hughes14 counsel for the accused had argued in a pre-trial hearing that the evidence of one complainant should be ruled inadmissible as "inherently unreliable",° because it was alleged to have been based on a recovered memory. In this case, however, the Judge was satisfied that this was not a case involving recovered memory.16 This result raises the question of what would have been the outcome if it were found to be an instance of recovered memory. Would counsel for the accused have been successful in having the relevant evidence deemed inadmissible, and if so, on what grounds?

12 [1994] NZHC 85; (1994) 11 CRNZ 402.

  1. Research suggests that females form the majority of sexually abused children whereas males form the majority of sexual abuse perpetrators. Therefore, throughout this paper, complainants will be referred to as "she" and perpetrators/accused will be referred to as "he". The writer acknowledges that these are generalisations, however has adopted these pronouns for the sake of simplicity. See Watkins & Bentovin, "The Sexual Abuse of Male Children and Adolescents" (1992) 3 Journal of Child Psychology and Psychiatry 1, 239; Martin et al, Otago Women's Health Survey : Results (1991) 1.

14 Unreported CA 470/96 28 May 1997 Eichelbaum CJ, Blanchard & Robertson JJ.
15 Ibid, 3.
16 Ibid, 4.

Similarly, in D v 107 the court commented as follows:

There is no evidence to show that at the time of charging the applicant there was concern known to the police about the retrieved memory syndrome...Armed with the experience gained in this case, the police may have greater future ability to assess the validity of the evidence founded on allegedly retrieved memory.'"

These comments suggest that allegations of recovered memory provide circumstances that warrant special investigation techniques in order to determine the veracity of a complainant's story before the matter goes to trial.

In H v B,'9 counsel for both parties agreed that the case was not one of "repressed" or "recovered" memory." This is despite the fact that the plaintiff deposed that the shock of dealing with her son's sexual abuse "jogged my memory about what had happened to me as a child.""

In R v Campbell," Campbell was convicted with the murder of a man whom the appellant perceived as having made homosexual advances to him. The appellant had been sexually abused by another man (V) as a child. The defence sought to have the appellant's murder conviction reduced to manslaughter, on the basis of provocation. According to the defence, when the deceased had put his hand on the appellant's thigh and smiled at him, the appellant had a flashback of V, lost control and struck the deceased with a poker, punched him, and struck him with an axe. The appellant gave evidence at trial that the resemblance of the deceased to V was so great, that at the time he thought he was striking V.

Dr. Edwards, expert for the defence, testified that when someone had "suppressed" the memory of abuse they could later manifest themselves in their consciousness under situations of stress, when a trigger caused a recollection of past events to resurface.

Dr. Edwards also testified that "repression of the intense feelings of anger, rage and humiliation, would result in the subject having less ability than the average person to control his anger.""

17 [1995] 3 NZLR 366.
'" Ibid, 370.
19 Unreported, High Court, Wellington, CP 7/97 4 September 1998, Gendall J.
20 Ibid, 17.
21 Ibid, 3.
22 (1996) 14 CRNZ 117.
23 Ibid, 121.

It is not clear from the facts of this case whether there was a point in time when the appellant did not remember the abuse, and whether this is a case of recovered memory or not. However, the case is helpful in that it is clear that the expert evidence was admissible to support the appellant's contention of loss of self-control. What can be drawn from this case, then, is that the court was prepared to accept that disturbed memory functioning is a possible outcome of sexual abuse.

In summary, the approach of New Zealand courts to the issue of recovered memory has been generally inconsistent. Felicity Goodyear-Smith, founding President of Casualties of Sexual Abuse Allegations," believes it unlikely that police would now bring charges where the complainant's testimony is based on recovered memories." The writer suspects this is an accurate assessment, based upon the controversy such memories engender. If so, complainants who have had amnesia for abusive episodes will effectively be denied the use of the legal system and serious questions must be raised about access to justice. The writer believes that it is not an option for the legal system to simply put recovered memory cases in the "too hard" basket and leave them there until further research provides more insights.

4. Corroborating Recovered Memory Claims in Criminal Trials: Using Moriarty

In New Zealand, because of the Accident Compensation Insurance regime, there are limited circumstances in which a complainant of sexual abuse may bring a civil action for damages. Thus the recovered memory issue has arisen more often in the context of criminal trials.

However, New Zealand criminal courts can gather some assistance from U.S. civil cases. The decision in Moriarty is just one example of U.S. courts requiring some form of corroborating evidence before accepting "memory repression" as a ground for suspending limitation statutes. In that case, the majority opinion held that the "objectively verified evidence" requirement may be satisfied by, for example:

  1. An admission by the abuser.
  2. A criminal conviction (for use in a civil case).
  3. COSA is a New Zealand advocacy group established to support those falsely accused of sexual assault.
  4. Goodyear-Smith, F. "Notes from New Zealand" in False Memory Syndrome Foundation Newsletter, vol 6 no. 8, September 1997 finsf97.827.htm1.
  5. A documented medical history of childhood sexual abuse.
  6. Contemporaneous records or written statements of the abuser.
  7. Photographs or recordings of the abuse.
  8. An objective eyewitness's account.
  9. Evidence the abuser had sexually abused others.
  10. Proof of a chain of facts and circumstances having sufficient

probative force to produce a reasonable and probable conclusion that sexual abuse occurred.26

Howell CJ dissented on the last type of evidence identified by the majority. He noted that this particular type of evidence is simply the definition of circumstantial evidence. It was his view that only direct evidence can satisfy the requirement of "objectively verifiable corroboration". With respect, the writer disagrees with Howell CJ's interpretation of the phrase "objectively verifiable corroboration". In the context of the case, this phrase suggests that what is required is corroboration of a type external to the complainant's testimony. If this analysis is correct, the writer respectfully submits that circumstantial evidence (or evidence from which an inference may be drawn), while not direct, may still be objectively verifiable (verified from a source external to the complainant).

There has been much discussion elsewhere on what corroboration actually means. Jackson" identifies apparently conflicting approaches. It has been said that corroboration has no special meaning, and simply means evidence that confirms, supports or strengthens other evidence.28 Another definition is that corroborative evidence must be independent testimony that connects, or tends to connect, the accused with the crime.29 This latter definition emerged from the case of R v Baskerville and is referred to as the Baskerville test.

We hold that evidence in corroboration must be independent testimony which affects the accused by connecting or tending to connect him with the crime. In

  1. It must be remembered that Moriarty v Garden Sanctuary Church of God was a civil suit and hence not subject to the criminal standard of proof.
  2. Jackson, J.D. "Credibility, Morality and the Corroboration Warning" (1988)47 CLJ 3, 428.
  3. DPP v Hester [1973] AC 296, 315; DPP v Kilbourne [1973] AC 729, 746, cited in Jackson, (1998) ibid.

29 R v Baskerville [1916] 2 KB 658.

other words, it must be evidence which implicates him, that is, which confirms in some material particular not only the evidence that the crime has been committed, but also that the prisoner committed it.'°

Jackson concludes that there are two senses in which the word "corroboration" is used. It can be used in the ordinary sense of evidence that "fits in" with other statements. Alternatively, it can be used in the sense of a legal doctrine that requires a jury to look for independent evidence in support of a complainant's testimony that implicates the accused. Jackson refers to the latter definition as "legal corroboration" as opposed to corroboration in the ordinary sense." Despite this distinction, the writer suggests that in recovered memory cases, any form of supporting evidence is unusual, thus an adherence to strict rules regarding what is, or is not, capable of being legally corroborative, should be abandoned. This is supported by the United Kingdom case of R v Hills:

Corroboration is not infrequently provided by a combination of pieces of circumstantial evidence, each innocuous on its own, which together tend to show that the accused committed the crime.32

In New Zealand, the common law corroboration rules have been abrogated by statute, meaning that corroboration is not required by law, or practice, for a conviction to be sustained. However, in sexual offence cases it may still be difficult to obtain convictions. This is particularly so in historical cases where any evidence that may have existed has long since disappeared. Even in cases that are prosecuted soon after the alleged events there is often no corroborative evidence. However, the decision in Moriarty provides some assistance in respect of areas where corroboration may be found. When viewed as a whole, such corroboration may provide compelling evidence of the truth, or otherwise, of the allegations. This is not to suggest that corroboration rules should be reinstated in recovered memory cases, requiring corroborative evidence to sustain a conviction, but rather that such evidence can be available and of assistance to the courts.

Admissions by the Abuser

While there may be a lack of evidence generally in sexual abuse cases, the accused is almost always aware of whether he committed the crime with which he is charged." Thus, any admissions made by the accused should be admissible, and

  1. Ibid, 667 per Lord Reading CJ.
    Jackson, (1988), supra at note 27.

32 R v Hills (1987) 86 Cr App R 26.

  1. Department of Justice "Adverse Inferences from Silence" (1989) 22 U Mich JL Ref 1005.

usually are unless excluded on the grounds of involuntariness or unfairness. This applies to admissions by conduct also. Because an accused's behaviour and demeanour when first questioned can provide clues as to his guilt or innocence, the writer suggests that videotaping of accused in initial interviews be adopted as a standard practice in sexual assault cases.

Because of the absence of corroborative evidence in most sexual abuse cases, the writer also submits that in such cases if an accused failed to give an explanation to the police when asked, refused to give evidence in his defence in court, or lied either to the police or in cross-examination, these factors should all be available as corroborative evidence, under the classification of "admissions by the abuser".34 In proposing such reforms it is necessary to strike a balance between the interests of society in having sexual abuse perpetrators convicted, and the rights of the accused not to be wrongfully convicted." However, it should be noted that it is already permissible for juries to consider an accused's conduct as part of the evidence against him. For example, the fact that an accused unexpectedly left town the day after the offence, or fled from a crime scene, or concealed, destroyed or fabricated evidence is usually admissible as evidence against him.36

Failure to Answer Questions or Give Evidence

In 1994 the British Parliament adopted Prime Minister John Major's proposal to significantly curtail the right to silence, by allowing judges and juries to draw adverse inferences when a suspect remains silent.37 These provisions followed a

  1. For a fuller discussion on failure to give evidence and lies by the accused, see Midson, B. "Determining the Truth in Recovered Memory Cases", supra at n 1 at 158-166.
  2. It is important to note the rights protected under the "Search, Arrest and Detention" provisions of the New Zealand Bill of Rights Act 1990, and in particular the rights afforded to persons arrested, detained or charged (ss 23, 24), and the right to be secure against unreasonable search and seizure (s 21). However, these rights are subject "only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society" (s 5).

36 Department of Justice, supra at n 33.

  1. O'Reilly, GW. "England Limits the Right to Silence and Moves Towards an Inquisitorial System of Justice" (1994) 85 J Crim L 402. Sections 34-37 of the Criminal Justice and Public Order Act 1994 allow inferences to be drawn from the silence of the accused where he fails to mention facts relevant to his defence when questioned or charged (the ambush defence); refuses to account for suspicious marks, objects or substances or his presence at the scene of a crime; and fails to give evidence at his trial. Section 38(3) of the Act provides that no person shall be convicted solely on the basis of an inference founded on the defendant's silence: Enright "Right to Silence (1995) 145 NLJ 6690, 495.

similar enactment in Northern Ireland in 1988, namely the Criminal Evidence (NI) Order 1988. Although the Order was justified on the basis of dealing with terrorist suspects and persons suspected of holding money from parliamentary activity, it actually applies to all criminal suspects in Northern Ireland."

In some Australian states the trial judge cannot comment upon an accused's decision to remain silent. In other states only the prosecution may not comment upon such silence. However, in these states the judge must not make comments that invite the jury to infer guilt directly from silence."

In the United States, adverse comment on the defendant's failure to testify is prohibited under Griffin v California" and under Carter v Kentucky" the jury must be instructed not to draw any inferences from a defendant's silence at trial." Conversely, French law provides multiple opportunities for questioning of the defendant, and freely permits adverse inferences from silence."

In New Zealand, the Court of Appeal in Trompert v Police" held that silence can be the deciding factor in arriving at a conclusion of guilt beyond reasonable doubt." However in R v Drain" the Court of Appeal stated that silence is not independent evidence of guilt and may only be used to strengthen existing evidence. In R v Gunthorp" the Court of Appeal reiterated that failure to testify

  1. Jackson, J.D. "Curtailing the Right of Silence: Lessons from Northern Ireland" (1991) Crim LR 404. Articles 3-6 of the Order allow inferences to be drawn where the defendant relies on the ambush defence; fails to testify at trial; fails to account for the presence of objects, substances or marks; or fails to account for his presence. All of the articles permit the court, on the basis of any inferences drawn from silence, to treat such silence as capable of amounting to corroboration of any evidence (in relation to which the silence is material) given against the accused. However, article 2(2) provides that no person shall be convicted of a criminal offence solely on inferences drawn from silence.

39 McGinley, G & Wave, V. Evidence Handbook (1994) 112.
40 Griffin v California [1965] USSC 127; 380 U.S. 609 (1965).
41 Carter v Kentucky [1981] USSC 46; 450 U.S. 288 (1981).
42 Department of Justice, supra at n 33.
43 Ibid.

44 Trompert v Police (1984) 1 CRNZ 324.

  1. It should also be noted that pursuant to s 24 of the New Zealand Bill of Rights Act 1990, a person charged with an offence has the right not to be compelled to be a witness or to confess guilt, and s 23 provides that persons arrested or detained shall have the right to refrain from making any statement and to be informed of that right.
  2. R v Drain, Court of Appeal. 11 October 1994 (CA 249/94). Eichelbaum CJ, Casey & Thorp JJ.
  3. R v Gunthorp, Court of Appeal. 9 June 1993 (CA 46/93). Casey & Hardie Boys JJ, & Sir Gordon Bisson.

can be enough to take a prima facie case to the point of proof beyond reasonable doubt. However, the Court denied that this amounted to drawing an inference from the accused's silence." So, while the jury must not use a failure to testify as a fact from which they might infer an accused's guilt, they may:

[D]raw inferences adverse to the accused more readily by considering that the accused, being in a position to deny, explain or answer the evidence against him, has failed to do so.49

Therefore, judges have not been permitted to invite the jury to conclude that failure to testify is itself an indication of guilt. However, they have been permitted to instruct the jury that, where the accused does not testify it means that there is no evidence from the accused to undermine, contradict, or explain the evidence put forward by the prosecution." These are seemingly contradictory propositions. As Mahoney points out, it is difficult to understand the distinction between drawing an inference of guilt from silence, and a finding that silence strengthens the evidence against the accused which may lead to a finding of guilt beyond reasonable doubt." It is submitted that the distinction should be abolished in sexual offence prosecutions. As mentioned earlier, the distinction has been abolished altogether in other jurisdictions, where a full inference of guilt may be based on an accused's silence or failure to testify.

A frequently made argument against curtailing the right of silence is that alternative explanations may be drawn from someone's silence when arrested, or failure to testify." Silence may be consistent with confusion, loyalty, or fear of a co-accused." The suspect may have a well-founded fear that the answers will not be treated in confidence; or be reluctant to admit to having done something that was discrediting or embarrassing but not illegal." However, as Robertson points out, all circumstantial evidence is capable of alternative explanations. Furthermore, the accused will have the opportunity at trial to explain the reasons for his silence."

It is submitted that the arguments for the right to silence are outweighed by the justifications for curtailing it. In particular, there are certain offences, such as

48 Mahoney, R "Evidence" (1994) New Zealand Recent Law Review 82.

  1. Weisensteiner v R (1993)178 CLR 217, 235 cited in Mahoney, R "Evidence" (1995) New Zealand Law Review 58.

50 O'Reilly, supra at n 37.
51 Mahoney (1994), supra at n 48.
52 Robertson, "The Right to Silence Ill-Considered" (1991) 21 VUWLR 139.
51 Enright, supra at n 37.
54 Murdie, "The Right to Silence" (1995) 139 SJ 6, 148.
55 Robertson, supra at n 52.

sexual abuse, in which it is almost impossible to obtain a conviction without a confession or the ability to draw adverse inferences from an accused's silence."

In New Zealand inroads have already been made into the right to silence. For example, in paternity proceedings, a court is entitled to draw any such inferences as appear proper in the circumstances from a refusal to give a blood test, subject to the right of the individual to explain their reasons for refusal."

As Spencer points out:

If it is politically possible to knock one dent in the right of silence when the interests ofjustice require it, it is possible to knock another.58 [Emphasis added].

The writer submits that because of the lasting negative psychological outcomes associated with child sexual abuse, and the likelihood of sexual abuse perpetrators reoffending,59 sexual abuse of children is a serious offence for which perpetrators must be convicted. This is so regardless of whether or not the alleged offences are historical, or whether or not the charges are based on recovered memories. In order to ensure that perpetrators are convicted, traditional principles such as the right to silence may have to be revisited.


In some circumstances, lies told by an accused can corroborate the evidence against him. In R v Collings,6° McCarthy P said:

Before instructing a jury that they may treat a lie (if such they find it to be) as corroborating either the commission of the crime or the identity of the criminal, a Judge must hold that it can reasonably be regarded as more than merely consistent with either the truth or untruth of [the complainant's] testimony.6'

56 Ibid.

57 Family Proceedings Act 1980, s 57(2).

  1. Spencer "Child Witnesses, Corroboration and Expert Evidence" (1987) Crim LR 239.
  2. See for example Salter, A Transforming Trauma (1995) 40-41; Hall & Proctor (1987) cited in Bull, R & Carson (eds) D Handbook ofPsychology in Legal Contexts (1995) 360.

60 [1976] 2 NZLR 104.
61 Ibid, 117.

The Court of Appeal later emphasised in R v Toia62 that the circumstances in which lies by an accused add to the Crown's case are rare and in R v Samuels" it was held that lies must not be used too readily by the prosecution as part of a chain of proof of the accused's guilt. In most cases lies have a bearing only on the credibility of the accused. However, Mahoney points out that while the New Zealand Court of Appeal has consistently stated that the situations in which an accused's lie can prove guilt are rare, the test formulated in R v Collings" for allowing a lie to amount to corroboration does not actually set too stringent a test.

In sexual abuse cases, if it can be established that the accused has told a lie, and that this was motivated by an awareness that the truth is inconsistent with his innocence, then his lie should be taken as an implied admission of guilt, and therefore constitute corroborative evidence against him.

Criminal Convictions

The court in Moriarty was concerned with the use of a criminal conviction for the abuse at issue, as corroboration of the complainant's testimony in a civil claim. Cases where such a conviction exists are rare. However, the writer suggests that criminal convictions of the accused in other matters may be useful in making a determination in recovered memory cases. The evidence that the accused behaved in a similar way on occasions other than that being considered by the courts is generally excluded. This is an extension of the fundamental principle in western legal systems that an accused's criminal record may not be used in evidence against him, nor can evidence be used that shows that he has a tendency to commit crimes. However, evidence of prior convictions is admissible in certain circumstances as similar fact evidence (to be discussed later) and also as a factor affecting the accused's credibility.

In terms of prior convictions and an accused's credibility, section 5(4)(b) of New Zealand's Evidence Act 1908 provides that the accused may be cross-examined like any other witness, on matters not necessarily arising out of his or her examination-in-chief However, the Court may limit cross-examination as to previous convictions or to the accused's credit "as it thinks proper". Despite this open-ended discretion, the Court of Appeal in R v Clark66 made it clear that, in general, the discretion should be exercised in accordance with the limits prescribed by

62 [1982] 1 NZLR 555.
63 R v Samuels [1985] 1 NZLR 350.
64 R v Collings, supra at n 60.
65 Mahoney (1995), supra at n 49.
66 [1953] NZCA 18; [1953] NZLR 823.

s 1(f) Criminal Evidence Act 1898 (UK). In following the English provisions, NZ courts have held that the "shield" can be thrown away by the accused if he or she gives evidence or calls witnesses to testify to his or her good character; or attacks the character of prosecution witnesses.

However, even if the accused "throws away the shield", there is a question surrounding what types of conviction the accused may be questioned on. In R v Shrimpton67 Alderson B said a rape conviction would have nothing to do with the accused's character for honesty. However, in R v Winfield" the accused was charged with indecent assault. A character witness who testified as to the accused's "good behaviour with ladies" was cross-examined about the accused's prior conviction for an offence involving dishonesty. Humphreys J approved the cross-examination saying:

There is no such thing known to our procedure as putting half a prisoner's character in issue and leaving out the other half.

This decision has been criticised, but the New Zealand Court of Appeal in R v Johnston" did not overturn a trial judge's decision to allow cross-examination as to a conviction for dishonesty in a trial for indecent assault. However, a more recent English case, SelveyvDPP,7° did not allow cross-examination on convictions unrelated to the offences charged. New Zealand courts have not overturned Johnston, although R v Anderson" adopts the Selvey approach to the extent that any previous convictions must have some "real probative value".72

The writer suggests that courts should allow a wider range of convictions to go to the issue of credibility. While a conviction for dishonesty may not be strictly relevant to an indecent assault charge, it might rebut a defence argument of the "blameless life" of the accused. In R v Accused (CA 311/88)" the accused was charged with indecent assault on his daughter. The prosecution was granted leave to cross-examine on his previous convictions for indecent assault. The accused appealed against his conviction on the ground that such leave should not have been granted. The Court of Appeal said that the trial Judge had not erred in allowing the cross-examination. The accused had set out to establish his own

67 (1851)2 Den 319.
68 (1939) 27 Cr App R 139.
69 [1956] NZPoliceLawRp 8; [1956] NZLR 516.
70 [1970] AC 304.
71 [1999] NZCA 319; [2000] 1 NZLR 667.
72 Ibid, at 677 per Elias CJ.
73 [1989] NZCA 54; [1989] 3 NZLR 21

good character and himself as a benevolent family man, and had also suggested that the complainant was a liar and a thief, thereby impugning her character. While in this case, the accused's prior conviction was for a related offence, even if it had been for dishonesty, it would have rebutted his good character and may also have shed a new light on his assessment of his daughter's credibility.

Documented History of Abuse

It is most unusual in historical sexual abuse cases, recovered memory or otherwise, to find a documented history of the abuse. It seems plausible to suggest, in relation to at least some recovered memory cases, that if help was sought at the time, and the abuse acknowledged by a third party, that the events might not have been banished from consciousness. While any documented history of the abuse would certainly fall into the category of corroborative evidence, by virtue of the scarcity of such evidence, courts are unlikely to be assisted by this type of evidence.

Contemporaneous Records of Abuser & Photographs or Recordings

These two categories of evidence are unlikely to yield much in the way of corroboration of recovered memories. It is improbable that abusers record details of abuse. However, as a matter of course, warrants should be issued for the seizure of diaries, journals, and other documentation that may contain contemporaneous records. In relation to photographs or recordings, it is possible that an abuser has collected photographs or recordings of the abuse for pornography purposes. In addition, if the accused is in possession of child pornography generally, this evidence could potentially be admissible as similar fact evidence.

Objective Eyewitness Accounts

`Objective' eyewitness accounts of the abuse may be difficult to obtain, but third parties may have evidence of behaviours exhibited by the child at the time. Suspicious activity on the part of the accused that might not necessarily amount to child abuse as such should also be admissible, for example persistent attempts to be alone with a child. In terms of 'objectivity' surely evidence from anyone other than the accused and the complainant should meet this requirement. Any issues such as collusion should go to the weight of the evidence rather than its admissibility.

Objective 'earwitness' accounts should also be admissible. The rule against hearsay poses particular problems in sexual abuse cases because of the paucity of any other corroborative evidence. However, some common law and statutory exceptions to the hearsay rule where sexual offences are involved have been developed in response to the problems of young children testifying in court. Children are easily confused by cross-examination; they are often reluctant witnesses; and they sometimes recant, disclaiming prior testimony in order to protect the accused who is often a relative or friend,74 or from fear of reprisals.

The recent complaint exception provides that the fact that the victim of a sexual assault complained of it immediately afterwards is admissible in evidence as an exception to the rule against hearsay. However, it is admitted only as evidence of the consistency of the complainant's account, and not to prove the truth of the contents of the statement.


One justification for the hearsay exceptions is that of reliability. The statement made closest in time to the alleged events is usually the most reliable." Accordingly, any statements that may have been made by the complainant at or around the time of the alleged events would provide the court with an account that is not distorted by any post-event factors bearing on the complainant. The reliability of such evidence would, of course, depend on the memory and credibility of the testifying witness.

The following hypothetical situation demonstrates how hearsay statements can be utilised in recovered memory cases." A child complains to a friend that her father has been committing indecent acts upon her. In the intervening years the complainant has no memory of either the alleged abuse or the complaint. Upon first recovering her memory of the abuse, the person to whom she complained spontaneously reveals that an early complaint was made, and the memories recovered are consistent with the early complaint. This pre-amnesia complaint could corroborate a complaint made after memory recovery. If the details of the two complaints were consistent, it would suggest that the offences did, in fact, occur. There is no opportunity for the complainant to manufacture a "memory" to

74 Notes "The Testimony of Child Victims in Sexual Abuse Prosecutions: Two Legis

lative Innovations" (1985) 98 Harvard Law Review 806.

R v T [1998] 2 NZLR 257; R v H [1997] 1 NZLR 673.

  1. Kelly, "Legislative Responses to Child Sexual Abuse Cases: The Hearsay Exception and the Videotape Deposition" (1985) 34 Cath UL Rev 1021.
  2. See Midson, B. "Determining the Truth in Recovered Memory Cases" supra at n 1, at 147.

comply with the early complaint. This is not a radical proposal considering that reliability is one justification for admitting hearsay evidence. The early complaint is likely to provide the most accurate account of what actually happened.

In many instances of recovered memories the first disclosure will be made many years after the alleged events. While courts currently are unlikely to admit such complaints under the recent complaint exception, they may provide the most reliable account of what the complainant remembers as happening. Evidence of these complaints should be admissible, as they record what the complainant remembers before the possibility that later events distort her memory.

While hearsay statements may also be admitted under the res gestae or excited utterances exception, the requirement of spontaneity precludes many statements that might otherwise be reliable. In Washington, the legislature has enacted a statutory exception to the rule against hearsay, which permits the courts to look at indicia of reliability other than that of spontaneity. That is, out of court disclosures can be admissible if "the time, content, and circumstances of the statement provide sufficient indicia of reliability"." In New Zealand, the Law Commission's proposed Evidence Code contains a similar provision relating to the admissibility of hearsay evidence in criminal proceedings. Section 19 of the proposed Code provides that hearsay is admissible if "the circumstances relating to the hearsay statement provide reasonable assurance that the statement is reliable"." If this provision is enacted as proposed, New Zealand courts could be guided in its interpretation by the U.S. courts. In Washington, in applying the statutory exception, courts have held that the following factors can be taken into account as indicia of reliability: the age of the child; their physical or mental condition; the exact circumstances of the alleged event; the language used by the child; the child's family, school and peer relationships; and the reliability of the testifying witness." In recovered memory cases, these factors could be taken into account in admitting evidence of early complaints, as well as in determining the weight or reliability of the early complaint. Evidence of an early complaint can be probative of the fact that the abuse occurred, given the unlikelihood of a complainant persisting with a false complaint over an exceptionally long period of time.

Generally, because of the huge difficulties in proving allegations of sexual abuse, any statements that may have been made to friends, siblings or other family

  1. Washington Statute cited in Yun, "A Comprehensive Approach to Child Hearsay Statements in Sex Abuse Cases" (1983) 83 Columbia Law Review 1745.
  2. Law Commission "Evidence Code and Commentary" Report 55 Volume 2 (1999) 52.

80 Yun, supra at n 78.

members should be admissible to prove the truth of the statement made. Such statements could also counter the defence that the allegations are recent fabrications or that therapists have implanted the memories. In admitting the complaint for these purposes, it should be irrelevant whether or not the complaint was made within a reasonable time, or elicited by leading questions.

Evidence Abuser had Abused Others

Evidence that an accused had abused others falls into what is referred to in New Zealand as "similar fact evidence". Evidence of similar conduct in the past is admissible if there is an identifiable pattern of behaviour, and the evidence is more probative than prejudicial."

There are compelling reasons for admitting similar fact evidence in child sexual abuse cases:

Only a person who has the lust for sex with children is likely to abuse a child sexually, this lust is unusual and the fact that the defendant has it makes it much more likely that he did indeed commit the offence."

Similarly, Spencer suggests that evidence of paedophilia ought generally to be admitted:

In the very nature of things, evidence of paedophilia in paedophile cases tells us something much more important than evidence of previous dishonesty in a trial for theft or of previous violence in a trial for common assault, and this should justify its being admitted."

As Munday points out, people who sexually abuse children tend to display set behaviour patterns, and their case histories can prove most informative." Literature on sexual offending suggests that an addiction model successfully captures many aspects of sexually deviant behaviour. Sexual aggression can be highly repetitive, it is a reliable way to alter and enhance mood, it is shrouded in secrecy and denial, it involves thinking errors that justify the behaviour, it can be compulsive and driven, and subject to frequent relapse."

See DPP v P [1991] 2 AC 447, [1991] 3 All ER 337; adopted in New Zealand in R v Accused (CA 202/91) (1991) 7 CRNZ 604.
82 Spencer (1987), supra at note 58.
83 Ibid.

84 Munday, "Sexual Abuse and the Similar Fact Rule" (1992) 51 CU 1, 41.
85 Salter, supra at n 59, at 40-41.

In recovered memory cases, as in sexual abuse cases generally, evidence of past convictions for sexual offences against children should be admissible. Further, if the accused can be shown in some other way to have sexual inclinations towards children, this evidence should be admissible." This evidence could be established by psychological assessments of the accused, although such assessments are not foolproof. Before using such assessments in criminal proceedings certain levels of reliability would need to be ensured.

Similar fact evidence is not necessarily limited to evidence of prior convictions. An accused may have been acquitted several times resulting from a lack of corroboration," or charges may have been withdrawn for other reasons. In R v Pinkerton" the Court of Appeal held that evidence related to a charge which was withdrawn by the prosecution following the accused's compliance with the Police Diversion Scheme, could still be relied upon by the prosecution as similar fact evidence when prosecuting another charge against the accused.

If evidence of past indecencies with children is not admitted, the facts of the case often appear in a false light." In recovered memory cases where allegations are made against the father, the accused may attempt to paint a picture of an ideal family life and perfect parenthood. Evidence that the accused has a history of indecencies with children would at least undermine the credibility of his account.

The obvious argument against admitting such evidence is that it is highly prejudicial to the accused. Spencer argues that the solution to this problem is to give the courts a discretionary power to reject any such evidence in inherently weak cases." It would seem that this is already the case at least in relation to sexual abuse cases. A further protection would be provided for the accused if courts were unable to convict solely on the basis of similar fact evidence. While in R v McIntosh9' the Court ofAppeal stated that a conviction may be obtained on the basis of similar fact evidence alone, the Court was more constrained in the later case ofR v F:92

86 Spencer (1987), supra at note 58.
87 Williams, G "The Corroboration Question" (1987) NLJ 131.

  1. R v Pinkerton, Court of Appeal. 23 March 1993 (CA 342/92). Cooke P, Richardson, Hardie Boys, Gault & McKay JJ, cited in Mahoney (1994) supra at n 48.

N9 Spencer (1987), supra at note 58.
90 Ibid.
91 R v McIntosh (1991) 8 CRNZ 514.
92 R v F, Court of Appeal. 28 March 1995 (CA 227/94). Cooke P, Casey & Heron JJ.

It is difficult to imagine that there could ever be a case where similar fact evidence of itself could conclusively prove the accused's guilt of the alleged crime for which he or she is being tried...similarities may [only] add to the likelihood that the accused committed the crime now charged, thus combining with the more direct evidence to establish the case beyond reasonable doubt.93

This would seem to be the more reasonable approach to adopt in relation to similar fact evidence. While similar fact evidence might be highly probative, it would be dangerous to convict solely on this basis. Courts must also have regard to any collusion or contamination issues that may arise in cases where the similar fact evidence relates to complainants who have been in contact with one another.

In summary, each of the types of evidence described above may be unreliable when standing alone, and there may be explanations for the evidence other than the guilt of the accused. However, the evidence can make a convincing whole when added together, if the chance of all being wrong is negligibly small." Such evidence should therefore be capable of amounting to corroboration in criminal recovered memory trials. While it might be argued that these types of evidence provide only circumstantial evidence against the accused, there is no rule in New Zealand that a conviction cannot be sustained on circumstantial evidence alone." In addition to the types of evidence outlined above, a consideration of other factors may also direct the court towards a reliable outcome in terms of the truth or otherwise of the complainant's allegations.

5. Other Considerations

A fundamental argument raised against admitting recovered memory evidence in court is based upon the belief that it is impossible to distinguish a false memory from an accurate one. It is said that a complainant's firmly held belief in her memory (albeit allegedly false) makes her unshakeable in cross-examination. It is the writer's contention, however, that witnesses in any case who might be mistaken as to their recollection may still prove as unshakeable, given their firmly held belief in the accuracy of their memory. In any event, it has not been proven that recovered memories are prima facie less reliable than continuous ones.

In practical terms, in every case courts must look at the reliability of the complainant's testimony. Thus, there may be factors (not necessarily limited to the issue of recovered memory), which detract from the overall credibility of the

93 lbid at 7 per Cooke P.
94 Williams, supra at note 87.
95 See R v Horry [1952] NZLR 111.

complainant's evidence. The competency of a complainant to give evidence may be challenged on a number of grounds, including age or mental capacity. What may also be relevant in recovered memory cases is an assessment of the suggestibility, or otherwise, of the complainant. Is the complainant susceptible to influence (intentional or otherwise)? Are drug and alcohol abuse involved? Does she suffer from any recognised medical disorders that might impair her ability to perceive reality?96

What the court must also assess in recovered memory cases is the credibility of the memory itself. That is, is the substance of the allegation credible? The presence of bizarre aspects to the allegations does not necessarily detract from their overall truth, however such characteristics may be suggestive of a false memory.

Another issue that should be addressed is whether the amnesia is full or partial. If recovered memories are, as alleged, less reliable than 'normal' memories, are the recovered memories consistent with the ones for which there was no amnesia? A further consideration might be how the complainant herself classifies her memories of abuse. Was it always remembered, never remembered, or never thought about? The way a complainant classifies her memory might not be in the way a clinician, researcher or lawyer would classify it, which leads to problems for the legal system in determining an appropriate legal response. This dilemma lends support to the conclusion that the focus in these cases should be on the substance and credibility of the allegations, and the presence or absence of corroborative evidence, rather than the aetiology of the allegations.

The timing of the recall is another significant factor in assessing the reliability of the memory. If the defence suggests that the memory is the result of therapist implantation, obviously memories recovered prior to therapy are not 'tainted' by any degree of suggestion from a therapist. Similarly, if the stimulus for recall is something other than therapy, such as spontaneous recall or some other significant trauma, the memories are less likely to have been manufactured. Other factors to consider are the use of hypnosis and so-called 'truth serums'. Given the risks associated with these methods of memory retrieval in terms of inadvertent provision of false information, memories that are recovered without recourse to such techniques should be considered more reliable.

The credibility of the accused should also be considered in light of any behavioural traits that may throw their reliability into doubt, and also that might be consistent

96 Vella, S. "Recovered Traumatic Memory in Historical Childhood Sexual Abuse Cases:

Credibility on Trial" 32 U.B.C.L. Rev 91. LEXIS 124.

with child sexual abuse. From a summary of comparative studies on child sex offenders," the most common characteristics exhibited by child sex offenders were a past history of sexual abuse as a child, low self-esteem and limited social skills. As research continues to develop in this area, it may be possible to identify a range of characteristics, which are so common to child sex offenders, that such a profile would be of real assistance in determining the validity of complaints in these cases.

Likewise, although there is dispute as to whether a particular range of characteristics can be said to identify sexual abuse victims, the presence of one or more factors such as eating disorders, self-abusive behaviour, intimacy difficulties, for example, may contribute to a determination that sexual abuse has occurred. Such a conclusion may more readily be drawn where there is no other reasonable explanation for the behaviour.

Finally, there may be other evidence that confirms or contradicts the testimony in the case when assessed as a whole. An example might be evidence showing the accused did not have the opportunity to sexually assault the complainant at the times alleged." While inconsistencies in times and places do not necessarily mean the central events alleged are untrue, in cases where precise times and places are facts in issue, the lack of opportunity in those times and places will tend to absolve (partially or completely) the accused of any wrongdoing. Say, for example, a daughter alleges her father abused her during a period when she was four years old, in the garden shed of the house in which they were living at the time. However, evidence shows that her father was estranged from her mother at the pertinent time, living in another part of the country, with no access, and did not at any point in time live in the house with them. Clearly, the complainant was incorrect either in the fact that her father abused her, or in the fact of where the abuse took place. Of course, she may have been abused by her father (either at an earlier or later time) and/or abused by someone else in the garden shed. But if her allegations are based solely on this particular time, place and perpetrator, the prosecution will be hard pressed to convince the judge or jury that the allegations are true.

6. Conclusion

It is timely to move on from the debate over whether it is possible to recover accurate memories of sexual abuse. To refuse to accept the possibility is to deny a number of complainants redress to legal remedies. The writer contends that it is

  1. Cossins, A. Masculinities, Sexualities & Child Sexual Abuse (2000) 255-257. Vella, supra at n 96.

within the capabilities of the criminal justice system to make a determination in any case as to the truth of the complainants allegations, based upon corroborating evidence. While traditional forms of corroboration may be lacking (for example, physical evidence and eyewitness observations), other forms of evidence such as confessions, admissions and other behaviours of the accused, prior convictions and similar fact evidence, and hearsay evidence can be of assistance. Such evidence may be unreliable when standing alone, but when viewed as a whole, can be compelling evidence that the complainant's memories are reliable. In order to facilitate a judgment, and justice in individual cases, wider scope must be given to those involved in the collection of evidence, and greater flexibility must be adopted in assessing the admissibility of such evidence.

NZLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback