Due Process and the Admission of Expert Evidence on Recovered Memory in Historic Child Sexual Abuse Cases: Lessons from America
| Author | Sinéad Ring |
| Date | 01 January 2012 |
| Published date | 01 January 2012 |
| DOI | 10.1350/ijep.2012.16.1.392 |
| Subject Matter | Article |
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
Due process and the
admission of expert
evidence on recovered
memory in historic child
sexual abuse cases:
lessons from America
By Sinéad Ring*
Lecturer in Law, National University of Ireland, Galway
Abstract This article reviews the decisions of the US state courts on the admissi-
bility of expert testimony on recovered memory in historic child sexual abuse
prosecutions. Unlike their English and Irish counterparts, most US courts
scrutinise the reliability of expert evidence on recovered memory. In examining
the US decisions the article explores the challenges posed to the criminal
process by the contested scientific status of recovered memory theory. It sets out
due process arguments why expert evidence on the topic should not be
admitted in a criminal trial.
Keywords Expert evidence; Recovered memory; Due process; Historic child
sexual abuse prosecutions; the United States; Ireland; England and Wales; Law
Commission
doi:10.1350/ijep.2012.16.1.392
66 (2012) 16 E&P 66–92 THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF
* Email: sinead.ring@nuigalway.ie. This article is based on doctoral research at University College
Cork funded by the Irish Research Council for the Humanities and the Social Sciences. The
research for the article was carried out at Harvard Law School in 2010. I am grateful to the
anonymous reviewers for their insightful comments. Thanks also to Caroline Fennell, Catherine
O’Sullivan, Noelle Higgins and Carol Steiker for their comments on previous drafts. Any errors are
my own.
he admission of expert evidence in a criminal trial allows specialised
knowledge to inform the jury’s decision, and thereby contribute to the
accuracy of the trial verdict. However, when convictions are found to
have been based on unreliable expert evidence, the legitimacy of the entire
criminal justice system is undermined. The past decade has seen a rash of
high-profile miscarriages of justice involving unreliable expert evidence tendered
by the prosecution,1sparking a debate in academic2and political circles3about
how to prevent junk science and other forms of untested assertions masquerading
as expertise from reaching the jury. The main criticism is that the current liberal
admissibility test—which merely requires the expert to be qualified and the
evidence to be relevant4—allows expert evidence of doubtful reliability ‘to be
admitted too freely with insufficient explanation of the basis for reaching specific
conclusions, be challenged too weakly by the opposing advocate and be accepted
too readily by the judge or jury at the end of the trial’.5The publication of the Law
Commission’s recent Report on the subject heralds an important move away from
the current laissez-faire approach and towards increased scrutiny of the reliability
of expert evidence.6Similar proposals have been mooted in Ireland.7This article
resonates with the reliability debate and examines the particular challenges
posed to due process by expert evidence on recovered memory in historic child
sexual abuse cases.
THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF 67
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
T
1RvDallagher [2002] EWCA Crim 1903, [2003] 1 Cr App R 12; RvClark (No. 2) [2003] EWCA Crim 1020,
1980, [2006] 1 Cr App R 5.
2 See, e.g., A. Roberts, ‘Rejecting General Acceptance, Confounding the Gate-Keeper: The Law
Commission and Expert Evidence’ [2009] Crim LR 551. Hartshorne and Miola advocate a specialist
gate-keeping panel to decide on the admissibility of expert evidence based on new theories and
techniques. J. Hartshorne and J. Miola, ‘Expert Evidence: Difficulties and Solutions in Prosecutions
for Infant Harm’ (2010) 30(2) LS 279 at 289.
3 House of Commons Science and Technology Committee, Forensic Science on Trial, Seventh Report of
the Session 2004–05, HC 96-I 2005.
4 C. Fennell, The Law of Evidence in Ireland, 3rd edn (Bloomsbury: Dublin, 2009) ch. 7; P. Roberts and A.
Zuckerman, Criminal Evidence (Oxford University Press: Oxford, 2004) chs 3 and 7. Reliability of
expert evidence is not a question of admissibility but rather a matter going to the weight of the
evidence: L. Heffernan, R. Ryan and E. J. Imwinkelried, Evidentiary Foundations (Tottel: Dublin, 2008)
109.
5 Leveson LJ, ‘Expert Evidence in Criminal Courts—The Problem’, paper presented to the Forensic
Science Society, Kings College London, 16 November 2010 at 15.
6 Law Commission, The Admissibility of Expert Evidence in Criminal Proceedings in England and Wales, Law
Com. Report No. 325 (2011), available at
expert-evidence.htm>, accessed 22 October 2011.
7 The Irish Law Reform Commission has provisionally recommended the introduction of a judicial
guidance note outlining ‘a non-exhaustive and non-binding list of factors, based on empirical
validation’ to help the court assess the reliability of the proffered evidence: Consultation Paper on
Expert Evidence (LRC CP 52-2008) para. 2.390.
In contrast to most historic child sexual abuse prosecutions (where the
complainant has always remembered the abuse but could not complain contem-
poraneously due to feelings of powerlessness),8in recovered memory cases the
delay in reporting is attributed to a complete lack of access to the memory of
abuse. These traumatic memories are said to be completely cut off from
consciousness for years, until the memory resurfaces, causing great distress. In
cases involving a complainant’s recovered memory of childhood abuse, the prose-
cution may proffer expert evidence to explain the theory of recovered memory.
Admissibility tests determine whether expert evidence on recovered memory
reaches the jury. Although it has inspired much academic commentary,9the
reliability of expert evidence on recovered memory has not been the subject of
detailed curial comment.10
Unlike their English and Irish counterparts, most US state courts scrutinise the
reliability of expert testimony on recovered memory. This article looks at the
experience of the US state courts in grappling with the evidential challenges posed
by expert evidence on recovered memory in criminal cases and considers what
lessons may be learned.11 In particular it considers how the US courts have dealt
with the challenges posed to the fact-finding process by the state of scientific
knowledge on recovered memory. The article makes no claims regarding the
68 THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
8 On the difficulties facing defendants in historic childhood sexual abuse prosecutions in Ireland,
see S. Ring, ‘Beyond the Reach of Justice? Complainant Delay in Historic Child Sexual Abuse Cases
and the Right to a Fair Trial’ (2009) 9(2) Judicial Studies Institute Journal 162.
9 The admissibility of expert evidence in delayed child sexual abuse prosecutions has been the
subject of academic comment in England and Wales: P. Lewis, Delayed Prosecution for Childhood Sexual
Abuse (Oxford University Press: Oxford, 2006); P. Lewis, ‘Expert Evidence of Delay in Complaint in
Childhood Sexual Abuse Prosecutions’ (2006) 10 E&P 157; P. Lewis and A. Mullis, ‘Supporting
Evidence and Illusory Double-counting: Recovered Memory and Beyond’ (2001) 5 E&P 111; M.
Redmayne, ‘A Corroboration Approach to Recovered Memories of Sexual Abuse: A Note of Caution’
(2000) 116 LQR 147; M. Redmayne, ‘Another Note of Caution’ (2001) 5 E&P 121.
10 In CvDPP [2009] IEHC 400 where the charges related to a period over 30 years earlier, the complaint
first emerged as a direct response to a question from a counsellor. Although the High Court
ultimately prohibited the prosecution of the charges on the grounds of incurable prejudice to the
defendant’s right to a fair trial, it found that the complainant ‘credibly [stated] that she sublimated
terrible memories of these events’ (at [30], per Hedigan J). Unfortunately the court offered no
guidance on how trial courts should approach the issue of recovered memory. In NC vDPP [2001]
IESC 54 the Supreme Court halted a prosecution based on a memory of abuse recovered by
hypnosis, where the therapist was no longer available to testify and there was considerable
uncertainty surrounding the circumstances of the recovered memory. The absence of an effective
test or control of the mechanisms of the alleged recovered memory rendered this a situation
‘fraught with the risk of unfairness’, per Hardiman J. In People (DPP) vMcKenna, unreported, 19
October 2001, Irish Court of Criminal Appeal, Geoghegan J noted that the concept of ‘restored
memory’ was ‘allegedly surrounded with major controversy’.
11 Admissibility thresholds are common to both civil and criminal cases. However, the focus of this
article is on US state court decisions in criminal cases.
validity of recovered memory theory or the veracity of accounts of abuse that are
founded on recovered memories. Instead it focuses on reliability as part of the
admissibility determination, and the arguments are firmly located within a legal
evidential framework committed to fundamental notions of due process and a fair
trial.
The recovered memory controversy
Over the last three decades a controversy so divisive that it has been dubbed
the ‘memory wars’ has emerged in the fields of psychology and psychiatry
regarding the phenomenon of recovered memories.12 In the United States this
controversy has become an established feature of the criminal justice landscape,
primarily in cases of historic child sexual abuse.13 Recovered memory cases are
those in which adults initially believe they were not sexually victimised as
children and later come to believe that they were.14 They should not be confused
with cases where people who always knew they survived childhood abuse
remember additional details or instances of abuse. Recovered memories are the
product of dissociative amnesia, ‘an inability to recall important personal infor-
mation, usually of a traumatic or stressful nature, that is too extensive to be
explained by normal forgetfulness’.15 The key concept in recovered memory
theory is trauma; the sexual abuse is said to be so traumatic that the victim’s
consciousness represses any memory of it in order to cope, leaving the person
with no recollection of the abuse until many years later, when the repressed
memory surfaces, usually in the form of ‘flashbacks’.16 There are two opposing
views held by psychologists and psychiatrists on the question of whether
THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF 69
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
12 J. Ost, ‘Review: Seeking the Middle Ground in the “Memory Wars”’ (2003) 94 Brit J Psychology 125 at
125. For a detailed examination of the controversy, see A. Piper, R. Kritzer and L. Lillevik, ‘What’s
Wrong with Believing in Repression? A Review for Legal Professionals’ (2008) 14(3) Psychology, Public
Policy, and Law 223; Lewis, above n. 9 at ch. 5.
13 In the United States the experts who have testified or submitted amicus curiae briefs to the courts on
the reliability or otherwise of recovered memories include professors of psychiatry and psychology
at Harvard (James A. Chu, Daniel Brown, Harrison G. Pope and Richard McNally), a professor of
psychology at University of California, Irvine (Elizabeth Loftus) and a professor of sociology at the
University of California, Berkeley (Richard Ofshe).
14 D. S. Lindsay and J. D. Read, ‘“Memory Work” and Recovered Memories of Childhood Sexual Abuse:
Scientific Evidence and Public, Professional, and Personal Issues’ (1995) 1 Psychology, Public Policy,
and Law 846 at 847.
15 American Psychological Association, Diagnostic Statistical Manual, Fourth Edition (DSM-IV-TR)
(American Psychological Association: Washington DC, 2000) 478. The DSM is a diagnostic
manual used by American and Canadian psychiatrists and psychologists to diagnose mental
illnesses.
16 L. Terr, Unchained Memories: True Stories of Traumatic Memories, Lost and Found (Basic Books: New York,
1994); L. Terr, ‘Childhood Traumas: An Outline and Overview’ (1991) 148 Am J Psychiatry 10.
accurate childhood memories of sexual abuse can be submerged or repressed,
only to emerge intact later in life.17 The divisions reflect a general disagreement
over the fundamental nature of memory as being primarily reproductive and
therefore generally accurate, or reconstructive and at best an inaccurate represen-
tation of the past.18
Those who argue that repression is a possible consequence of the trauma caused
by childhood sexual abuse are predominantly clinician-researchers. They rely on a
body of clinical studies demonstrating that traumatic experiences can be
forgotten.19 According to this group, repression of child sexual abuse is more likely
relative to other traumatic events, specifically for abuse committed by trusted
adults, for more traumatic forms of abuse, and for repeated abuse.20 They also
claim that repressed memories are stored faithfully and, once recovered, can
provide a reliable record of the traumatic events and that independent corrobo-
ration of recovered memories is often present.21
70 THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
17 Albert et al.,Final Conclusions of the APA Working Group on the Investigation of Memories of Childhood Abuse
(American Psychological Association: Final Report of the Working Group on the Investigation of
Memories of Childhood Abuse, 1996).
18 Ost, above n. 12 at 127. For a useful summary of the psychological literature on the recovered
memory debate, see P. Lewis, L. Alison and M. Kebbell, ‘Considerations for Experts in Assessing the
Credibility of Recovered Memories in Child Sexual Abuse: The Importance of Maintaining a
Case-Specific Focus’ (2006) 12(4) Psychology, Public Policy, and Law 419.
19 See D. H. Gleaves, S. Smith, L. Butler and D. Spiegel, ‘False and Recovered Memories in the
Laboratory and Clinic: A Review of Experimental and Clinical Evidence’ (2004) 11(1) Clinical
Psychology: Science & Practice 3; D. Brown, ‘Recovered Memories: The Current Weight of the Evidence
in Science and in the Courts’ (1999) 27(1) J Psychiatry and Law 5.
20 D. Brown, D. C. Hammond and A. W. Scheflin, Memory Trauma Treatment, and the Law (Norton: New
York, 1998); C. Dalenberg, ‘Recovered Memory and the Daubert Criteria’ (2006) 7(4) Trauma Violence
and Abuse 274; J. Freyd, Betrayal Trauma:The Logic of Forgetting Child Abuse (Harvard University Press:
Cambridge MA, 1996).
21 J. Herman, Truth and Recovery (Basic Books: New York, 1992); E. Bass and L. Davis, The Courage to Heal: A
Guide for Women Survivors of Child Sexual Abuse (HarperPerennial: New York, 1988); J. A. Chu, L. M.
Frey, B. L. Ganzel and J. A. Matthews, ‘Memories of Childhood Abuse: Dissociation, Amnesia, and
Corroboration’ (1999) 156(5) Am J Psychiatry 749; and R. Cheit, The Recovered Memory Project,
argues that victims of sexual abuse are less suggestible than average subjects because they have a
deeply rooted distrust towards authorities: F. Leavitt, ‘Suggestibility and Treatment as Key
Variables in the Recovered Memory Debate’ (1999) 17 American Journal of Forensic Psychology 5at
12–16.
Most sceptics of recovered memory theory are experimental cognitive psycholo-
gists.22 For this group, ‘[t]he notion that the mind protects itself by repressing or
dissociating memories of trauma, rendering them inaccessible to awareness, is a
piece of psychiatric folklore devoid of convincing empirical support’.23 Experi-
mental psychologists argue that the more traumatic an event was when it
happened, the more difficult it is to forget.24 While aspects or details may be
forgotten or distorted, the core of the memory—the sexual abuse—is not. Even the
meaning of the term ‘dissociation’ is contested.25 Sceptics point to the method-
ological limitations of the studies cited by clinicians, on the ground that the
practitioner’s inferential task is usually more subjective and therefore more open
to error,26 and that clinical research relies to a significant extent on the
self-reporting of the patient. They also argue that claims of repression and
accurate memory recovery are rarely independently corroborated.27 In this
context Davies’ warning that ‘therapists and the courts should view with extreme
caution experts who claim to be able to divine truth or falsity from the internal
qualities of a recovered memory narrative alone’ is crucial.28
The binary nature of the recovered memory debate is most obvious in relation
to the potential for therapy to elicit true and false memories.29 While clini-
cians claim that it is possible to revive ‘memories’ of long-forgotten events
and that such memories are generally correct, experimental psychologists claim
that most recovered memories are the result of suggestive therapeutic
THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF 71
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
22 See E. Loftus and K. Ketcham, The Myth of Repressed Memory: False Memory and Allegations of SexualAbuse
(St Martin’s Press: New York, 1994); E. Loftus and D. Davis, ‘Recovered Memories’, Annual Review of
Clinical Psychology, vol. 2 (2006) 469–98; R. McNally, Remembering Trauma (Harvard University Press:
Cambridge MA, 2003); R. Ofshe and E. Watters, Making Monsters (Scribner: New York, 1994); Piper et
al., above n. 9; H. G. Pope, P. S. Oliva and J. I. Hudson, ‘Repressed Memories: B. Scientific Status’ in
D. L. Faigman, D. H. Kaye, M. J. Saks and J. Sanders (eds), Modern Scientific Evidence: The Law and Science
of Expert Testimony, vol. 1 (Thomson West: St Paul, 2008) 154–95.
23 McNally, above n. 22 at 275.
24 D. Davis and E. Loftus, ‘The Scientific Status of “Repressed” and “Recovered” Memories of Sexual
Abuse’ in J. L. Skeem, K. S. Douglas and S. O. Lilienfeld (eds), Psychological Science in the Courtroom
(Guilford Press: New York, 2009) 55 at 56; Pope et al., above n. 22 at paras 20:8–20:15.
25 McHugh considers it to be a descriptive metaphor and not a diagnostic or explanatory category. P.
McHugh, Try to Remember: Psychiatry’s Clash over Meaning, Memory and Mind (Dana Press: New
York/Washington DC, 2008) 173–5.
26 Davis and Loftus, above n. 22 at 56.
27 For an in-depth review of the retrospective and prospective studies conducted to date on repressed
memories, see Pope et al., above n. 22. They conclude: ‘Repression in short is a testable hypothesis,
but it has not yet been appropriately tested. Pending satisfactory studies, therefore, the most
reasonable scientific position is to maintain skepticism’ (at para. 20:15).
28 G. M. Davies, ‘Is It Possible to Discriminate True from False Memories?’ in G. M. Davies and T.
Dalgliesh (eds.), Recovered Memories: Seeking the Middle Ground (Wiley: New York, 2001) 171.
29 Using techniques such as hypnosis, age regression, facilitated communication, visualisation,
guided imagery and the use of sodium amytal.
techniques.30 Researchers criticise what they see as the retreat by clinicians away
from a verifiable, ‘historical’ truth to a narrative truth, where the therapist
searches for a past that may not have occurred but that may be useful for thera-
peutic purposes.31 However, proponents of recovered memories argue that to
suggest that claims might be false merely represents another way for society to
deny the reality of childhood sexual abuse.32
There are some points of consensus, for example that most people who suffer
traumatic events have continuous memories of the trauma, and that certain
therapeutic practices, like hypnosis and guided imagery, are particularly
troubling.33 Nevertheless, the scientific status of recovered memory remains a
deeply contested terrain for psychologists. Many experts retain a firmly
entrenched perspective either in favour of full amnesia followed by gradual full
remembering (true recovered memory) or a false, iatrogenic process of recovery.34
Unfortunately, a resolution remains remote; science has yet to develop objective
measures to discriminate between false and true recovered memories.35 Therefore,
neither science nor the courts have recourse to an external marker or validation
for recovered memories. The unverifiable nature of recovered memories poses
evidential challenges for courts when these memories are proffered as providing
the basis for an allegation of historic child sexual abuse. The next section
examines various approaches taken by state courts in the United States to expert
evidence on recovered memory, treating them as illustrative of the various
approaches which may be adopted by other jurisdictions.
72 THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
30 Loftus and Davis, above n. 22 at 58–80. They claim that memories are open to many transfor-
mations at several levels—encoding, storage and retrieval—so that not only are most memories not
a direct match of the original events, but many things can be ‘recovered’ that never happened.
31 Ost, above n. 12 at 132.
32 See, e.g., J. Freyd, ‘Science in the Memory Debate’ (1998) 8 Ethics and Behavior 101.
33 D. W. Schuman and A. McCall Smith, Justice and the Prosecution of Old Crimes: Balancing Legal, Psycho-
logical and Moral Concerns (American Psychological Association: Washington DC, 2000) 95; Lindsay
and Read, above n. 14.
34 Lewis et al., above n. 18 at 419–20. Lewis etal. argue that these views reflect a lackof consideration of
the details of each individual case. They say that this can result in an oversimplified presentation
of a case in court, in which an expert fails to give due consideration to the range of factors and
subtleties that should inform the decision to admit the evidence (at 420).
35 The DSM states that ‘[t]here is currently no method for establishing with certainty the accuracy of
such retrieved memories in the absence of corroborative evidence’. DSM-IV, above n. 15 at 480–1.
The admissibility of expert evidence on recovered memory in the United
States
For most of the 20th century the standard for evaluating expert evidence was that
established in Frye vUS.36 The test asks whether a scientific theory is ‘generally
accepted’ by members of the relevant scientific community.37 Today, however, the
majority of state courts apply a formal admissibility test based on the seminal US
Supreme Court decision in Daubert vMerrell Dow Pharmaceuticals, Inc.38 that trial
judges are gatekeepers with responsibility to assess the reliability of the expert
opinion.39 Daubert was a response to the ‘austere’ general acceptance standard,
which excluded scientific theories about which there was no consensus.40 Never-
theless, Daubert imposes a rigorous standard for admissibility. It requires courts to
evaluate scientific evidence based on the methodology used and not the conclu-
sions derived by the researcher.41 Reliability is not to be bundled up in the jury’s
determination as to the weight to be given to the expert testimony, instead it is a
question for the judge in her role as gatekeeper. The proponent of the evidence
must demonstrate that it is the product of sound scientific methodology, and this
is determined according to a non-exclusive list of factors that scrutinise the
reliability of the methodology underlying the proffered evidence.42 These include
falsifiability, peer review/publication, potential/known error rate and general
acceptance. Most US states apply a Daubert or similar test or include Daubert factors
in their tests.43
In general there are two approaches taken by the US courts to cases involving
expert evidence on recovered memory: (a) in the minority of states that continue
to apply the Frye ‘general acceptance’ test, expert evidence on recovered memory is
viewed as exceptional and is per se admissible; and (b) in the majority of states,
where Daubert is used, expert evidence on recovered memory is subjected to an
THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF 73
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
36 293 F 1013 (DC Cir 1923).
37 Ibid. at 1014.
38 509 US 579 (1993).
39 The trial judge’s gatekeeper function was later extended to all expert testimony, with the Supreme
Court emphasising the trial court’s ability to become sophisticated consumers of science: Kumho
Tire Co. vCarmichael, 526 US 137 (1999).
40 The court in Daubert held that Frye was an ‘austere standard’ that would be at odds with the ‘liberal
thrust’ of the Federal Rules of Evidence (at 588). FRE r. 702 was later amended to require an
assessment of reliability.
41 The US Supreme Court admonished judges to focus ‘solely on principles and methodology, not on
the conclusions they might generate’ (at 595).
42 These factors include falsifiability, peer review/publication, potential/known error rate and
general acceptance.
43 Sixteen states apply a ‘general acceptance’ test, and four states have developed their own tests for
the admissibility of expert evidence: see A. Lustre, ‘Post-Daubert Standards for Admissibility of
Scientific and Other Expert Evidence in State Courts’ (2001) 90 ALR 5th 453.
admissibility test. In those states where Daubert is used, many criminal courts
seem to subject both expert evidence on recovered memory and complainant
testimony based on recovered memory to an admissibility test. Both forms of
evidence are generally deemed insufficiently reliable to be admitted. The notable
exception to this exclusionary trend is the recent Massachusetts case of Common-
wealth vShanley.44
Expert evidence on recovered memory is per se admissible
Under this approach the courts scrutinise testimony based on a ‘new scientific
technique’.45 However, testimony based on recovered memory is deemed not to be
based on complex science and therefore is exempted from an admissibility test.46
The exception has been justified on the basis that jurors can ‘temper their
acceptance of [the expert’s] testimony with a healthy scepticism born of their
knowledge that all human beings are fallible’.47 The best example of the problems
associated with the rule in the context of expert evidence on recovered memory
comes from an Arizonan civil case. In Logerquist vMcVey48 the Arizona Supreme
Court refused to adopt any test beyond basic relevancy for experience-based
testimony on the phenomenon of recovered memories.49 This was because
experience-based testimony was based on inductive reasoning and therefore did
not require any threshold test. This reasoning was denounced by two dissenting
judges and has been the subject of severe criticism by scholars.50
Admissibility test
As noted earlier, in evaluating expert testimony, the majority of state courts apply
a formal admissibility test, based on the decision in Daubert.51 While there does not
appear to be any case where a court has held that spontaneously recovered
74 THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
44 SJC-10382, 455 Mass 752, 919 N E 2d 1254 (2010), discussed below.
46 A recent example of this approach is Phillips vGelpke 190 NJ 580, 921 A2d 1087 (2007) Supreme Court
of New Jersey. Of course, even if the court concludes that the complainant’s testimony as well as
that of the expert on the topic should be admitted the court may still exclude it on the grounds
that it is more prejudicial than probative under FRE r. 403 or a similar state rule.
48 196 Ariz 470, 1 P 3d 113 (2000).
49 In the context of recovered memory evidence, these requirements of relevance and competency
can be crucial. In Frangipane the Supreme Judicial Court of Massachusetts reversed the defendant’s
conviction on the grounds that a Commonwealth social worker’s testimony about the loss and
recovery of a traumatic memory through dissociation was beyond her expertise and should not
2004 WL 535975 (Ky).
50 See ‘Symposium: Logerquist vMcVey’ 3 Ariz St LJ 40 (2001) and T. N. Stevens, ‘The Admissibility of
Expert Testimony on Repressed Memories of Child Sexual Abuse in Logerquist v. McVey: Reliability
takes a Backseat to Relevancy’ (2001) 46 VLLR 385.
51 509 US 579 (1993).
memories are per se inadmissible,52 many US courts have used admissibility
thresholds to scrutinise not only expert evidence on recovered memory, but also
complainant testimony based on recovered memory.
The Supreme Court of New Hampshire has led the way in the US in establishing a
rigorous approach to the admission of recovered memory testimony and expert
evidence. In State vHungerford, the Supreme Court of New Hampshire recognised
that the issue of recovered memory straddles both the admissibility of the expert
testimony and the question of the admissibility of the complainant’s recovered
memory.53 It found that a memory recovered in therapy cannot be separated from
the process, if any, which facilitated the recovery.54 Therefore, when challenged,
testimony that relies on recovered memories must satisfy a pre-trial reliability
determination.55 The court envisaged a pre-trial admissibility assessment based on
a threshold standard that scrutinises the reliability of the theory of recovered
memory. Following Daubert, the court held that in determining the reliability of a
recovered memory, the trial court should consider the following factors: (1) the
level of peer review and publication on the phenomenon of repression and
recovery of memories; (2) whether the phenomenon has been generally accepted
in the psychological community; (3) whether the phenomenon may be and has
been empirically tested; (4) the potential or known rate of false recovered
memories.56
Applying these criteria to the theory of recovered memory, the court found that
the level of peer review was high, but that the debate over methodology and the
meaning of results was such that it could not state that the theory had gained
general acceptance in the psychological community.57 Furthermore, ethical
considerations precluded psychological testing of the impact of traumatic events
on memory, i.e. whether or not trauma makes memory disappear. The court also
noted that it was difficult to estimate the rate of ‘false’ recovered memories.
However, Hungerford envisages a reliability assessment that goes beyond the
narrow focus on the methodologies underpinning recovered memory theory. In
addition to inquiring into the methodology and principles underlying recovered
THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF 75
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
52 Lewis (Oxford University Press: 2006), above n. 9 at 157.
53 State vHungerford 142 NH 110, 697 A 2d 916 (1997). Hungerford was referred to by Hardiman J in the
54 In Hungerford the trial judge held a two-week admissibility hearing and heard from seven eminent
psychologists and psychiatrists as well as the treating therapists and the two complainants.
55 ‘[T]estimony that relies on memories which previously have been partially or fully repressed must
satisfy a pretrial reliability determination’ (at 921).
56 These four factors are taken from Daubert, above n. 38.
memory theory, the court should also consider other relevant factors relating to
the complainant’s testimony. These other factors are: (5) the age of the witness at
the time the event(s) occurred; (6) the length of time between the event and the
recovery of the memory; (7) the presence or absence of objective, verifiable corrob-
orative evidence of the event; and (8) the circumstances surrounding the recovery
of the memory, i.e. whether the witness was engaged in therapy or some other
process seeking to recover memories or likely to result in recovered memories.58
The admissibility of the expert evidence on recovered memory and the recovered
memory testimony could be assessed in light of these indicia of reliability.59 The
court noted that the complainants’ ages and the length of time elapsed between
the alleged abuse and the recovery of the memory bore in favour of reliability.60
However, it deferred to the trial judge’s finding of a lack of corroborative evidence.
This was despite the fact that the therapist of one of the complainants testified
and one complainant’s sister made similar allegations of abuse against the
defendant. Indeed, while the therapist’s evaluation of the concerns of members of
staff at the therapy centre and the testimony of the defendant’s wife offered ‘some
enlightenment’, they were not sufficiently corroborative.61 The court was also
worried about the problematic nature of therapeutic techniques used to recover
memories. However, it was the nature of the unreliability of the phenomenon of
recovered memories that was the real concern:
The phenomenon of recovery of repressed memories has not yet
reached the point where we may perceive these particular recovered
memories as reliable … The indicia of reliability present in the
particular memories in these cases do not rise to such a level that they
overcome the divisive state of the scientific debate on the issue.62
Therefore the court refused to overturn the trial judge’s finding that the
complainants’ recovered memories were inadmissible. Since Hungerford the US
courts continue to highlight the importance of an admissibility threshold in
recovered memory cases.63
76 THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
58 697 A 2d 916 at 925.
59 Ibid.
60 The time elapsed between the alleged abuse and the recovery of the memory was one and a half
61 Ibid.
62 Ibid. at 930, per Brock CJ (emphasis added).
63 Smith vCommonwealth of Kentucky 2004 WL 535975 (Ky), 2007 WL 2812597 (Ky App); Commonwealth of
Kentucky vSmith 2007 WL 2812597 (Ky App).
The recent decision of the Supreme Judicial Court of Massachusetts in Common-
wealth vShanley64 is a notable exception to the general exclusionary trend in
relation to the admissibility of recovered memories in criminal cases.65 In a
judgment delivered in January 2010 the court made a number of important
findings in relation to expert testimony in cases involving alleged recovered
memories.
The prosecution’s case centred on the complainant’s recovered memories of
sexual abuse which was alleged to have taken place over a six-year period in the
1980s when he was attending religion classes at the church where the defendant
served as a priest.66 The memories did not surface until 2002, when the
complainant learned of abuse allegations against Shanley. The case and the subse-
quent appeal against conviction turned on whether the evidence based on the
recovered memory should have been admitted.67
The trial judge held an admissibility hearing regarding the expert testimony on
recovered memory. The test used was that established in the Massachusetts case of
Lanigan. Under Lanigan the core question is the reliability of the theory or process
underlying the expert’s testimony.68 At first glance, this would seem to be
identical to the Daubert test. However, unlike Daubert, the Lanigan test contains no
reference to specific factors to scrutinise the reliability of the methodology on
which the evidence is based. As will be seen, the vagueness of the Lanigan test
contributed to a confusing analysis here in Shanley.
Among the experts called by the prosecution were Daniel Brown and James A. Chu,
both leading experts in the field of clinical psychology and proponents of
THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF 77
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
65 Some US courts have not followed the Hungerford approach and have admitted the complainant’s
testimony and the expert testimony describing the theory of recovered memories. However,
these decisions are mostly lower court decisions in civil cases.Isley vCapuchin Province, 877 F Supp
1055 (ED Mich 1995); Shahzade vGregory, 923 F Supp 286 (D Mass 1996). See Stevens, above n. 50 at
399.
66 The case formed the epicentre of a scandal uncovered by the Boston Globe newspaper involving a
litany of abuse within the Boston Catholic Archdiocese. See M. A. Hamilton, ‘The “Licentiousness”
in Religious Organizations and Why It Is Not Protected under Religious Liberty Constitutional
Provisions’ (2010) Wm and M Bill of Rights J 953.
67 He also appealed on the grounds that the six-year period under the statute of limitations had
expired. While the court found that the trial judge had given an erroneous instruction to the jury
on this issue, it did not give rise to a miscarriage of justice. Shanley, above n. 44 at 1277–8, per Cordy
J.
testimony is subject to review only for abuse of discretion. Canavan’s Case, 432 Mass 304, 733 NE 2d
1042 (2000).
recovered memory theory.69 In addition, the trial judge also considered an
affidavit by Elizabeth Loftus for the defence, who testified as to the existence of
false memory syndrome.70 The trial judge concluded that the diagnosis and
theories behind recovered memory were generally accepted in the relevant scien-
tific community. This was a surprising conclusion for two reasons: first, because
the trial judge held a Lanigan admissibility hearing, and so should have been
concerned with the principles underlying the expert opinion rather than the level
of acceptance in the psychological community. Secondly, a finding of general
acceptance is extremely difficult to justify given the debate that still rages in the
psychological literature on this issue.71
The Supreme Judicial Court dismissed Shanley’s appeal against conviction. The
decision’s focus on the admissibility of the expert evidence on recovered memory
is interesting; unlike Hungerford, the admissibility of the complainant’s testimony
based on recovered memory is not mentioned. The focus is on the reliability of the
expert testimony. Indeed, the court was at pains to demonstrate its concern with
reliability—the first seven pages of the judgment rehearse the opinions offered by
the various experts regarding the existence of dissociative amnesia (repression).
The court held that expert evidence is sufficiently reliable if the underlying theory
or methodology is either (1) generally accepted in the relevant scientific
community72 or (2) it satisfies the requirements adopted in Lanigan, i.e. that the
theory or process underlying the complainant’s testimony is reliable. It held that
the trial judge did not abuse his discretion in finding that the expert testimony on
dissociative amnesia and recovered memory was sufficiently reliable to go to the
jury. The court deferred to the trial judge’s finding that Dr Brown’s testimony
constituted sufficient and reliable peer review for the purposes of general
78 THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
69 Dr Brown is assistant professor of clinical psychology at Harvard Medical School. He has testified in
judicial proceedings regarding dissociative amnesia in five states and has testified before the
International War Crimes Tribunal on memory and trauma. He co-authored a leading text in the
area Memory Treatment and the Law, above n. 18. Dr Chu is also a leading protagonist in the recovered
memory debate. He was a member of the task force set up in the 1990s to review the dissociative
disorder diagnosis for the DSM.
Professor of Psychology at University of California, Irvine. She has published 18 books and more
than 250 scientific articles and has served as an expert witness or consultant in hundreds of trials.
She is a past president of the American Psychological Society. In February 2011 Loftus was awarded
the American Association for the Advancement of Science Scientific Freedom and Responsibility
Award. For an intriguing personal account of the admissibility hearings in Hungerford and in other
recovered memory cases, see McHugh, above n. 25.
71 Davis and Loftus, above n. 24.
acceptance,73 and stressed that a ‘relevant scientific community’ must be defined
to include a sufficiently broad sample of scientists so that the possibility of
disagreement exists.74
Unfortunately, this analysis fails to encompass the nuances and complexity of the
recovered memory debate in the psychological field and sidesteps any real
scrutiny of the methodology underlying the expert testimony. Indeed, the court
seems to have confused the general acceptance test and the Daubert/Lanigan
scrutiny of methodology test. This confusion is evident in the court’s finding that
the lack of scientific testing did not make unreliable the theory that an individual
may experience dissociative amnesia.75 In one fell swoop, the court rejected the
defendant’s contention, often made by critics of recovered memory, that the
theory of recovered memory is invalid because there is no scientific method to test
for its existence in certain individuals, nor are there known error rates or stand-
ardisation. In so doing it adopted a definite ‘side’ in the recovered memory
debate—that of the proponents of the theory. By sidestepping the reliability
question, the court has effectively forced juries in future cases to choose between
options in a debate that is so deeply divided that the experts themselves cannot
agree, even about the existence of the phenomenon. Such a result is unacceptable
in a criminal justice system that claims to be based on evidence and not opinions.
Reliability and due process
The laissez-faire approach of a minority of US state courts, similar to the admissi-
bility test employed in England and Wales and in Ireland, will permit expert
evidence on recovered memory as long as it is relevant and not patently
unreliable. However, a more rigorous admissibility test than relevance is required
if the defendant is to be afforded due process. The liberal approach creates the
potential for unfounded speculations and assertions masquerading as expertise to
enter the jury’s decision-making process. The expert’s testimony may have a
certain aura of scientific objectivity or infallibility which, when combined with
the frailties of human decision-making, is likely to cause them to give too much
THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF 79
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
73 Dr Brown testified that recovered memory was generally accepted in the field. He relied on six
studies to that effect, which the court cited. ‘According to those surveys (taken collectively),
eighty-nine per cent of those surveyed accepted the validity or possible validity of dissociative
amnesia’ (at 1265, per Cordy J).
74 The court declined to rule out overturning a conviction in a future case where the only evidence is
1270 (2010).
75 Ibid. at 1266 (emphasis added). The court held that this conclusion was supported in the record,
‘not only by expert testimony but by a wide collection of clinical observations and a survey of
academic literature’.
weight to the expert’s opinion.76 The risk of unquestioning jury deference gives
rise to particular concerns about a fair trial if there are unresolved questions
about the reliability of that expert evidence.77 Furthermore, in an area such as
recovered memory, where there is fundamental disagreement among profes-
sionals about whether traumatic memories can be repressed and later successfully
recovered, a basic relevance inquiry risks an unfair conviction.
In recognition of the need to protect juries from unreliable opinions, the clear
preference of the majority of US state courts in relation to admissibility of expert
evidence on recovered memory has been to pass the expert testimony through an
additional evidentiary filter beyond relevance.
The Hungerford approach applies an admissibility threshold to both the expert
evidence on recovered memory and the complainant’s testimony. It would seem to
offer the courts a way of combining rigorous scrutiny of the methodology under-
lying the expert evidence with a contextualised assessment of the complaint. The
first four factors are taken from Daubert and require the judge to ensure that only
reliable expert evidence reaches the jury. As noted earlier, the court in Hungerford
found that the theory of recovered memory failed to satisfy this reliability test. The
final four factors, or indicia of reliability, are a laudable attempt to establish the
reliability of the complainant’s memory on grounds external to the memory itself
and to the theory of recovered memory. At first glance this approach seems
compelling; however, closer examination reveals that it promises far more than it
can deliver. Remember that in Hungerford, the indicia of reliability could not
overcome the divisive state of the scientific debate on recovered memory, and
both the expert evidence and the complainants’ testimony were deemed inadmis-
sible. Unfortunately, therefore, it appears that even though Hungerford offers a
potentially useful way to contextualise the admissibility test, it is clear that even
in the very case where the indicia of reliability were developed, they failed to
surmount the evidential problems caused by the lack of scientific agreement and
lack of a scientific method of testing the reliability of recovered memories. Indeed
the dissension within scientific circles regarding the reliability of recovered
memories has proved fatal in later cases applying Hungerford,78 even where the
80 THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
76 See L. Hand, ‘Historical and Practical Considerations Regarding Expert Testimony’ (1901) 15 Harv
LR 40; M. Redmayne, Expert Evidence and Criminal Justice (Oxford University Press: Oxford, 2001) 125.
77 Leveson LJ, above n. 5 at 6.
78 In Quattrocchi the court applied the Hungerford factors and excluded the expert evidence on
recovered memory on the grounds that the status of dissension within the scientific community
rendered potential expert testimony of little assistance to the trier of fact: State vQuattrocchi, 1999
WL 284882 (R I Super); III 2001 WL 100373 (R I Super).
interval between the alleged abuse and its recollection is relatively short.79 This
leaves courts looking for guidance in a difficult position: if the indicia of reliability
are not enough to carry the case beyond the scientific reliability concerns in the
very case in which they were formulated, when will they be sufficient to do so?
Furthermore, does this mean that complainant testimony based on recovered
memory must also be excluded? The notion of supporting evidence has been
mooted as a way of avoiding the Scylla of unfair convictions based on unreliable
expert testimony and the Charybdis of preventing all prosecutions based on
recovered memories from proceeding.
Supporting evidence
Lewis has suggested that recovered memory testimony may be admitted where
there is supporting evidence and that the jury should be warned of its potential
unreliability.80 If the evidence were deemed corroborative, it would be admitted to
the jury along with the recovered memory testimony. In the absence of such
supporting evidence the case would be withdrawn from the jury. The approach
can be compared to that adopted in England and Wales in relation to prosecutions
following a cot death.81 According to Lewis, the evidence should corroborate the
actual abuse alleged, and not just be part of the complainant’s story, or the occur-
rence of sexual abuse.82 There are two kinds of evidence that might be proffered as
corroborating a recovered memory: evidence of bad character and physical
evidence.
(i) Evidence of bad character
The general rule is that evidence of the bad character of an accused person is
inadmissible in the course of a trial. There are exceptions to this rule, most
important for present purposes being where evidence of bad character may be
introduced as so-called ‘similar fact evidence’. Any bad character evidence must
be more probative than prejudicial before it may be admitted.83 The main
objection to bad character evidence is the devastating prejudice it inevitably
wreaks for the defence given ‘the capacity [of the evidence] to unfairly predispose
the triers of fact toward a particular outcome’.84 Bad character evidence might
THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF 81
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
79 State vWalters 142 NH 239, 698 A 2d 1244 (1997) Supreme Court of New Hampshire.
80 Lewis, above n. 9 at ch. 7 and the articles by Lewis and Mullis, above n. 9. A similar approach has
been developed by the Supreme Court of Carolina in a leading civil case: Moriarty vGarden Sanctuary
Church of God 341 SC 320, 534 SE2d 672 (2000). For a critique of this approach, see the articles by
Redmayne, above n. 9.
82 See Lewis (Oxford University Press: 2006), above n. 9 at 170–1.
84 M. Damaska, Evidence Law Adrift (Yale University Press: New Haven, 1997) 15.
include a pre-trial admission, a confession by the defendant85 or similar allega-
tions against the defendant made by another witness. If, for example, another
person alleges similar offences by the same defendant, there is an argument that
this evidence is indicative of the reliability of the complainant’s recovered
memory. This argument becomes stronger in cases where the other person is also
a complainant and the behaviour alleged is strikingly similar and unusual; in
other words, that it is so unusual that it could not be the ‘stock in trade’ of child
sexual abuse.86 In such cases, the probative value of the evidence would perhaps
outweigh its prejudicial effect and might count towards the admission of
recovered memory testimony.
While the use of supporting evidence would be in line with recent changes in
England and Wales in relation to bad character evidence,87 the limits on the use of
bad character evidence as corroboration in a recovered memory case would have
to be extremely narrowly construed to avoid any infringement of the presumption
of innocence. It would need to implicate the accused and be independent of the
evidence that makes the corroboration desirable and be credible.88 When these
considerations are coupled with the prejudicial effect of a prior record of offences
a jury would likely view as abhorrent, and the court’s duty to guard against
unfounded assumptions entering the jury’s decision-making process, it is difficult
to see how a court could admit such evidence as corroborative of a complainant’s
recovered memory.89 It would seem, therefore, that the number of cases where
there is evidence capable of constituting corroboration would be extremely small,
and would probably only include cases where there are similar allegations made
by another complainant, or where there is a pre-trial admission or confession by
the defendant.
82 THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
85 Although as Lewis notes, confessions must be treated with caution because of evidence of false
memories in this context. See Lewis (E&P 2006), above n. 9 at 169.
86 In DO’S, where the assaults described by the other witnesses were unusual and contained
similarities regarding the location, timing and the manner in which they were carried out, the
probative value of the evidence was found to outweigh its prejudicial nature: People (DPP) vDO’S
[2004]IECCA 12.However, see thecriticism of this approachby Fennell, above n.4 at para. 10.74.
87 In England and Wales the rules governing the admissibility of evidence of bad character have been
extended to include previous allegations in respect of which the defendant has been tried and
acquitted, allegations which were never tried or other complaints that never reached the police
(such as accusations made by a pupil to the headmaster but never reported to the police) (Criminal
Justice Act 2003, s. 98).It also extends to evidence of the defendant’s ‘disposition towards
misconduct’. This could include evidence that a person has expressed sexual attraction towards
children.
88 People (DPP) vGilligan [2006] 1 IR 107 at 143, per Denham J.
89 The Irish Supreme Court has criticised cross-examination that made assertions as to the
disposition of the accused to sexually abuse children, in a case where counsel suggested that the
defendant’s involvement with the Boy Scout movement was part of the ‘profile’ of a sexual deviant
or pervert: People (DPP) vDO [2006] 3 IR 57.
(ii) Physical evidence
This might include medical and school records, plans of the building(s) where the
abuse is alleged to have occurred, photographs of the location of the alleged abuse
and photographs or recordings of the abuse.90 However, the likelihood of such
evidence being corroborative of the defendant having committed the abuse
alleged, as opposed to merely consistent with the complainant’s account, is very
low. In this context Redmayne’s point that corroborating evidence does not
increase the probative value of the corroborated evidence is crucial.91
Due process issues
In JL vDPP, Hardiman J of the Irish Supreme Court noted that the recovered
memory controversy affects vital procedural points such as the persuasiveness of
an application for prohibition on the grounds of delay: ‘If the memory is viewed
wholly or partly as a repressed one its validity as a factor excusing delay depends
on the view taken of the reality of repression as a phenomenon.’92 Its impact on
evidence is more obvious; an important issue, should a complainant’s testimony
be admissible in the absence of expert testimony, is the jury’s ability to assess
credibility based on demeanour. Unfortunately, where the complainant claims to
have recovered the memory of abuse, cross-examination is unlikely to be as
effective a tool as it is with other kinds of witnesses, since the witness honestly
believes she is telling the truth. Indeed, even patently false recovered memories
are characterised by the witness’s belief in the authenticity of the experience she
describes.93 To compound matters, recovered memories have a more tenuous
connection with the person’s other recollections than do ordinary memories,
meaning that gaps in recollection (which would ordinarily cast doubt on the
reliability of the memory) may be explained by saying that the recovered memory
was developed through therapy, whereas others were not.94 The capacity of the
powerful instrument of cross-examination to identify inconsistencies in a
complainant’s testimony is so limited as to be practically useless.
Furthermore, there are two sets of recovered memories of which the courts should
be especially wary. First, if the event is alleged to have occurred very early in
childhood, the court should be extremely sceptical of the testimony, because
THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF 83
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
90 Lewis (E&P 2006), above n. 9 at 171.
91 Redmayne (LQR 2001), above n. 9 at 152.
prosecutions see Ring, above n. 8.
93 Davies, above n. 28 at 167.
94 The example is taken from M. Karlsson, ‘Time Out of Mind: Sexual Abuse and the Statute of Limita-
tions’ in J. Bjarup and M. Blegvad (eds.), Time, Law, and Society (Franz Steiner Verlag: Stuttgart, 1995)
49.
scientists generally agree that people do not remember the first few years of life.95
Secondly, memories recovered in therapy pose particular difficulties, since the
therapeutic goal is not to determine the historical truth, but rather to heal the
patient.96 Along with the problems of suggestibility and the influence of the
therapist this raises serious questions about the reliability of such memories.97
The twin goals of the criminal trial are to accurately determine the accused
person’s guilt or innocence and to do so fairly.98 In seeking to achieve these goals, a
primary function of the rules of evidence and procedure is to ensure that reliable
evidence is produced which can form the basis of an effective trial.99 The impor-
tance of safeguards to protect against unreliable evidence entering the jury’s
decision-making process is founded on the law’s respect for the ‘principled
asymmetry’ of State–citizens’ interactions in the criminal context.100 Respect for
rights is a concomitant aim of the criminal process—not merely a side-constraint
on the pursuit of accuracy, but an objective to be attained while pursuing that
aim.101 This is obvious when one considers the role of admissibility tests in
reinforcing the legitimacy of the jury verdict.102 Fundamental criminal due
process rights are therefore inseparable from the pursuit of accurate jury
verdicts—values of fairness and due process require that only reliable evidence
reaches the jury. Expert evidence on recovered memory raises a number of
concerns in relation to the potential of such evidence to undermine the factual
accuracy of the jury’s decision and the law’s commitment to fairness and due
process. While some of these concerns are common to all cases involving expert
evidence, there are particular factors pertaining to recovered memory prosecu-
tions that mean the consequences for due process are much more serious.
84 THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
95 J. Briere and J. R. Conte, ‘Self-reported Amnesia for Abuse in Adults Molested as Children ’ (1993)6 J
Traumatic Stress 21.
96 M. Wesson, ‘Historical Truth, Narrative Truth and Expert Testimony’ (1995) 60 Wash LR 331.
97 G. Ernsdorff and E. Loftus, ‘Let Sleeping Memories Lie? Words of Caution about Tolling the Statute
of Limitations in Cases of Memory Repression’ (1993) 84(1) Journal of Criminal Law and Criminology
129 at 158–62.
98 A. Ashworth and M. Redmayne, The Criminal Process, 3rd edn (Oxford University Press: Oxford,
2005) 22.
99 Ibid. at 23.
100 Roberts and Zuckerman, above n. 4 at 19.
101 Ashworth and Redmayne, above n. 98 at 45.
102 See earlier discussion on the problems with not scrutinising expert evidence. The limits placed
on the admissibility of expert evidence in criminal trials are based on the so-called ‘jury control’
theory of evidence which holds that exclusion of unduly prejudicial or unreliable evidence and
formal directions on how certain kinds of admissible evidence should be used are the tools
employed to control the natural reasoning processes of the jury. J. D. Jackson, ‘The Function of
the Criminal Trial in Legal Inquiry’ in A. Duff, L. Farmer, S. Marshall and V. Tadros(eds.), The Trial
on Trial: Volume 1, Truth and Due Process (Hart: Oxford, 2004) 121 at 135–7.
First, the jury’s task in recovered memory cases is much more difficult than in
other kinds of cases involving competing experts. In most trials involving
‘opposing’ experts, the jury is expected to critically evaluate different interpreta-
tions of a particular set of data. The jury must then integrate that evaluation into
their assessment of the rest of the evidence at trial, giving the expert evidence
appropriate weight.103 Commentators have pointed out that the jury’s ability to
choose between competing experts is often extremely limited. Hence, while jurors
may accept or reject the witness’s conclusions, neither acceptance nor rejection
will be based on rational reasons.104 Instead the choice will be based on factors
such as credentials, qualifications, performance or who seemed to be favoured in
the judge’s summing up.105 This lack of epistemic competence raises real due
process concerns about the risk of an unfair conviction based on unscrutinised
expert testimony.106 These concerns are underscored by the inscrutability of the
jury verdict, and the minimal possibility of reconsidering factual issues on
appeal.107
In recovered memory cases the due process problems posed by the jury’s lack of
epistemic competence are even more serious than in other kinds of prosecutions.
The difference between complex expert evidence generally and expert evidence
on recovered memory is that the two sets of expert testimony conflict, not merely
on issues of the interpretation of data, but at a fundamental level, that is, on the
very existence of recovered memory as a scientifically valid phenomenon. This
conflict between experts is irreconcilable given the uncertain state of the
knowledge upon which the experts have based their opinions, a problem made all
too clear by the absence of a scientific method for testing the accuracy of
recovered memories.108 To require non-expert jurors to evaluate and distinguish
THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF 85
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
103 J. Fordham, ‘Illuminating or Blurring the Truth: Jurors, Juries and Expert Evidence’ in B.
Brooks-Gordon and M. Freeman (eds.), Law and Psychology, Current Legal Issues, vol. 6 (Oxford
University Press: Oxford, 2006) 338 at 339.
104 R. J. Allen, ‘Expertise and the Daubert Decision’ (1994) 84 Journal of Criminal Law and Criminology
1157.
105 P. Roberts, ‘Science, Experts, and Criminal Justice’ in M. McConville and G. Wilson (eds.), The
Handbook of the Criminal Process (Oxford University Press: Oxford, 2002) 272.
106 On the epistemic competence of juries, see S. Brewer, ‘Scientific Expert Testimony and
Intellectual Due Process’ (1998) 107 Yale LJ 1535.
107 Damaska, above n. 84 at 65.
108 See above. This is a problem that resonates with issues around the admissibility of scientific
evidence in general. In Daubert, the US Supreme Court attempted to lay down tests to determine
whether the proffered scientific evidence had been subjected to a scientific methodology.
However, those philosophers of science whom the court invoked in order to demarcate genuine
and reliable science from other forms of knowledge did not claim that their theories could
distinguish between reliable and unreliable evidence. Jackson, above n. 102 at 143.
between the competing arguments and methodologies in the recovered memory
debate is not only to over-estimate wildly their abilities as fact-finders, but also
risks an unfair conviction based on a deeply contested and ultimately unverifiable
opinion.
Secondly, expert evidence in favour of the existence of recovered memory should
be understood, to borrow Redmayne’s description of confession evidence, as
‘outcome-determinative’.109 The existence of the phenomenon is so fundamental
to the complainant’s credibility and to the whole prosecution case that if the jury
accepts the prosecution expert’s evidence, there is arguably no leap to be made to
find the complainant credible. In short, if the jury accepts the theory of recovered
memory as explained by the expert, it will almost certainly also accept the
complainant’s story as credible. This sidelining of the jury’s role in critically
assessing the complainant’s credibility undermines the juror’s role as arbiter of
the facts.110 Indeed, the Court of Appeal has called for special caution in cases
where expert opinion evidence is fundamental to a prosecution.111 Furthermore,
as Hartshorne and Miola note, there is a stronger case for caution than there is
with confession evidence, because, with expert evidence, ‘the jury is being asked
to decide upon the reliability of material that is beyond its ordinary knowledge
and experience; in contrast, in many cases involving disputed confessions, the
jury should need no assistance in determining whether the defendant was truth-
fully confessing’.112 The challenges posed by the outcome-determinative nature of
expert testimony on recovered memory are exacerbated by the uncertain status of
recovered memory as a psychological phenomenon. The issue is a clear due
process one; if verdicts in child sexual abuse cases are to be legitimate it is
essential that it is the jury and not the experts who have the ultimate responsi-
bility for the verdict.
Finally, a few points relating to the potential for juror prejudice in the assessment
of complainant credibility. The duty to minimise the potential for unfounded
assumptions entering the jury’s decision-making process is especially important
in recovered memory cases because of the lack of surrounding physical and
circumstantial detail. Vidmar has identified different types of juror prejudice, two
86 THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
109 Redmayne, above n. 76 at 122.
110 The objection to expert testimony on the grounds that it is an impermissible comment on the
credibility of a witness is the objection that is most likely to succeed for either the prosecution or
the defence: C. L. Bennett and A. C. Goldbach, Trying Sex Offense Cases in Massachusetts, 2nd edn
(Massachusetts Continuing Legal Education: 2009) ch. 13, section 13.3.2.
112 Hartshorne and Miola, above n. 2 at 288.
of which are relevant to the present discussion.113 Generic prejudice involves the
transferring of pre-existing attitudes, beliefs and stereotypes about the person to
the trial setting. It can be directed against people accused of certain crimes, such
as child sexual abuse,114 where the repulsive nature of the charges against the
accused may of itself give rise to an inference of truth.115 ‘Conformity’ prejudice
involves the juror feeling pressure to reach a verdict consistent with community
feelings. Without wishing to overstate the point, it is at least arguable that in
recovered memory cases, the jury’s assessment of the complainant’s credibility
may be tainted by these forms of prejudice. In this context it is worth noting that
Hardiman J of the Irish Supreme Court has voiced concerns in relation to jurors’
‘presumptive credence’ of complainants in historic child sexual abuse.116 Indeed,
the risk of ‘presumptive credence’ may be further heightened by the fact that the
theory of recovered memory is based on the idea of trauma,117 a culturally potent
notion that authenticates suffering by linking present suffering to past
violence.118
The problem of ‘presumptive credence’ and jury prejudice is particularly relevant
to cases where (as happened in Shanley) the defence expert introduces the notion of
false memories. In these cases jurors must not only consider whether or not the
theory of recovered memory is valid, but they must also grapple with testimony to
the effect that false memories can be created in adults. Since the choice is one
between competing experts, it is effectively a simple binary one so that the jury
must choose between either agreeing that the recovered memory is true and
credible (the prosecution case) or deciding that the recovered memory is false (the
defence case). This polarity may encourage jurors to fit the complainant’s story
within one of these two alternatives without any reference to their individual
THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF 87
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
113 The other two types are interest prejudice (where jurors have an interest in the outcome of the
trial) and specific prejudice (where jurors hold attitudes that prevent them from being
impartial): N. Vidmar, ‘Case Studies of Pre- and Midtrial Prejudice in Criminal and Civil
Litigation’ (2002) 26(1) Law and Human Behavior 73.
114 N. Vidmar, ‘Generic Prejudice and the Presumption of Guilt in Sex Abuse Trials’ (1997) 21(1) Law
and Human Behavior 5.
115 See E. Smith-Lee, ‘Recovered Memories ofChildhood Sexual Abuse: Should Long-Buried Memories
be Admissible Testimony?’ (1996) 37 Boston College Law Review 591 at 637.
the part of judges.
117 See the references above n. 20.
118 D. Fassin and R. Rechtman, The Empire of Trauma (Princeton University Press: Princeton, 2009); J.
Brunner, ‘Trauma and Justice: The Moral Grammar of Trauma Discourse from Wilhelmine
Germany to Post-Apartheid South Africa’ in A. Sarat et al. (eds.), Trauma and Memory (Stanford
University Press: Stanford, 2007); S. Cohen, States of Denial, Knowing about Atrocities and Suffering
(Polity Press: Cambridge, 2001).
credibility as a witness or other evidence, effectively sidelining the presumption of
innocence, the burden of proof and the option of an acquittal.
As questions of admissibility of expert evidence on recovered memory become
ever more pertinent we would do well to remind ourselves of the values under-
pinning the criminal trial and to exercise careful vigilance as to what assumptions
and values are being woven into the normative framework of the criminal trial. In
light of the fundamental unreliability and unverifiability of expert evidence
regarding recovered memory, unquestioning deference to experts offends
criminal due process.
The court in Hungerford held that the complainants’ recovered memory should
also be subjected to an admissibility threshold. Although the court was concerned
with memories recovered in therapy, the principles apply to all cases involving
recovered memory. This approach is remarkable, given that the sole criterion for
eyewitness evidence has traditionally been personal knowledge of the relevant
information.119 However, the court was alert to the need to ensure that only
reliable evidence reached the jury. The court’s attempts to ameliorate the
harshness of an absolute exclusionary rule by supplementing the reliability
inquiry with an assessment of other factors that might indicate reliability failed.
This fundamental problem of unreliability is at the heart of all recovered memory
cases—essentially the complainant’s testimony is based on a theory that is not
robust enough to withstand scrutiny of its methodology.
Similarly, the supporting evidence approach is also incapable of overcoming this
unreliability problem. When this is coupled with the ineffectiveness of cross-
examination in this context, it would seem that the supporting evidence approach
would be of no probative use in all but the most exceptional cases,120 and to admit
it along with the complainant testimony would be to risk an unfair trial.
Finally a caveat: none of the foregoing is to say that every case involving recovered
memory should be prevented from proceeding to trial. If the supporting evidence
on its own—that is, without the recovered memory testimony—is such that a
reasonable jury properly charged could convict, then of course that evidence
should go to the jury. For example, the Supreme Court of New Hampshire has
88 THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
119 R. Zoltek-Jick, ‘For Whom does the Bell Toll? Repressed Memory and Challenges for the
Law—Getting beyond the Statute of Limitations’ in P. S. Appelbaum, L. A. Uyehara and M. R. Elin
(eds.), Trauma and Memory: Clinical and Legal Controversies (Oxford University Press: New York, 1997)
470.
120 Such as where there is a pre-trial admission or confession by the defendant or where there are
similar allegations made by another complainant.
upheld convictions in a case involving both repressed and continuous
memories.121 Although the recovered memories were inadmissible under
Hungerford, the court was clear that the complainant could distinguish between
her recovered memories and her continuous memories of assaults. The decision is
to be welcomed for its vindication of the complainant’s ability to recount only the
admissible parts of her story. Furthermore, given the uncertainty surrounding the
meaning of repression, and given the cultural meanings attached to trauma
beyond the therapeutic and psychological fields,122 it is highly possible that some
victims who claim to have recovered memories have simply begun thinking about
events that they avoided thinking about for years. The reasons for complainant
delay in historic child sexual abuse cases range from fear arising from a
relationship of dominion to societal disbelief of victims.123 In recovered memory
cases it is entirely plausible that the complainant has rationalised her inability to
complain in terms of repression rather than avoidance.124 Indeed, the Irish case
law shows that the court125 or the expert psychologist may do the same.126 In a
further twist, a leading researcher at Harvard University has suggested that
ordinary memory mechanisms may be operating in cases of alleged recovered
memory.127 Clearly, therefore, courts must be alert to the need to identify whether
repression is in fact an issue.
Conclusion
Criminal trials of historic child sexual abuse take place in a landscape discon-
nected from the normal matrix of physical and circumstantial detail.128 In
THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF 89
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
121 State vGibson, 897 A 2d 957 (2006).
122 See above n. 118.
nine, but no action (apart from two medical examinations) was taken until some 13 years later
when the complainant reported the abuse again. The High Court granted an order prohibiting
the trial on the grounds of the risk of an unfair trial. See Ring, above n. 8.
124 Redmayne, above n. 9 (2001).
125 In PC, which involved an application by the accused for an order halting the prosecution of the
charges on the grounds of delay, Denham J in the Supreme Court described a continuously
remembered memory as repressed, on the grounds that ‘rational consideration of abusive events
is frequently suppressed for complex personal, family and social reasons. Unless there is
126 In WvDPP, unreported, 21 March 2002, High Court, which involved an application to halt the
trial, the prosecution’s psychologist conceded under cross-examination that insofar as the
complainant had engaged in a conscious mechanism to push material from his mind that this
was the act of ‘suppression’ and not ‘repression’ and that her use of the term repression to
describe the former activity was incorrect.
127 McNally, above n. 22.
recovered memory cases this difficulty is compounded by the state of scientific
knowledge on the phenomenon.
The currently liberal admissibility test applied to expert evidence in England and
Ireland creates a huge potential for unfair trials in historic child sexual abuse
prosecutions based on recovered memories of abuse. There are indications of a
trend in favour of increased scrutiny of expert evidence in criminal cases,129
including recent dicta from the Court of Appeal suggesting a willingness to ensure
that unsupported assertions about the nature of memory do not reach the jury.130
The Law Commission has recommended the introduction of a statutory admissi-
bility test for expert evidence to replace the current test. The Commission’s
proposed test requires trial judges to carry out a reliability determination in
deciding whether the expert opinion is ‘soundly based’.131 The draft Bill that
accompanies the Report sets out factors which could ground a finding of unreli-
ability, including that the expert opinion is based on a hypothesis which has not
been subjected to sufficient scrutiny or that the opinion is based on an unjusti-
fiable assumption.132 This is a welcome move towards confirming judges in their
duty to shield jurors from expert evidence lacking a reliable scientific foundation.
However, the courts must not be complacent. The decision in Shanley is a powerful
reminder of judicial willingness to defer to experts without subjecting their
testimony to rigorous scrutiny. It is to be hoped that in scrutinising the reliability
of expert evidence on recovered memory, the English and Irish courts will
have regard to the importance of ensuring due process, an emphasis which in
general133 suggests that expert evidence on recovered memory should not be
admitted in a criminal trial. The adoption of a general rule to that effect would
90 THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
129 See, e.g., RvAtkins [2009] EWCA Crim 1879.
130 In RvGJB [2011] EWCA 867, the Court of Appeal rejected as ‘amateur psychology’ the assertion
that the complainant had a good recollection of the appellant’s bedroom because he had been
abused in it at the age of five. The court noted that there was no expert evidence before the jury to
support the suggestion that an incident as distressing as a rape or buggery would enhance a
child’s memory of his surrounds when it happened.
131 Law Commission, above n. 6 at para. 5.17. The Irish Law Reform Commission has provisionally
rejected the Frye general acceptance test, in favour of a judicial guidance note outlining ‘a
non-exhaustive and non-binding list of factors, based on empirical validation’ to help the court
assess the reliability of the proffered evidence: above n. 7 at paras 2.390 and 2.394.
132 Section 4(2) of the draft Bill. The Bill also lists generic factors to be taken into account when
determining reliability, including, if the opinion relies on the use of any method, whether that
opinion takes proper account of matters, such as the degree of precision or margin of uncertainty
affecting the accuracy of the results (draft Bill, Sched. 1, para. 1).
133 Perhaps expert evidence could be admitted in exceptional cases where there is strong
corroborative evidence, such as a pre-trial admission by the defendant or similar allegations
made by other complainants. See below.
perhaps run counter to the common law rule that there are no closed categories
where expert evidence may not be placed before a jury.134 However, in an area as
controversial as this, values of fairness and due process require such an
exceptional category.
Furthermore, where the sole prosecution evidence is the complainant’s recovered
memory, the only outcome consistent with the presumption of innocence is the
exclusion of recovered memory testimony. The Supreme Court of New Hampshire
has suggested putative indicia of reliability that would enhance the admissibility
of recovered memory testimony. However, even these indicia fail to overcome the
unsatisfactory state of scientific knowledge on recovered memory. While
supporting evidence may be useful in cases where there is evidence external to
that of the complainant’s recovered memory, the number of such cases where
there is evidence capable of constituting corroboration is extremely low. In light
of the unverifiability of recovered memory testimony and the minimal probative
value of much supporting evidence, as well as the ineffectiveness of cross-exami-
nation in recovered memory cases, in general complainant testimony based on
recovered memories should not be admitted. To expect jurors to assess the
reliability of recovered memories when experts in the field cannot do so is to
establish an opinion-based system of adjudication and risks a miscarriage of
justice.
A blanket exclusionary approach would mean that the criminal justice system
would be deaf to complainants alleging historic childhood sexual abuse on the
basis of recovered memories. It may therefore follow that some child abusers
would escape conviction. However, the values of fairness and due process in
criminal trials require that only reliable evidence should reach the jury.135 Due
process protections must be jealously protected if we are to remain faithful to the
moral foundations of the criminal justice system.136 If we regard the verdict in a
criminal trial as being given on behalf of the community at large, not only the
legal basis but the evidential basis of the conviction must be presented in terms
that the lay public, and the jury as its representatives, can understand and
THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF 91
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
135 That these concerns are not unfounded is illustrated by the Irish case People (DPP) vWall and
McCabe [2005] IECCA 140, where memories recovered during counselling at a psychiatric hospital
were the basis for a rape conviction. Ms Wall was subsequently granted a certificate of a
miscarriage of justice based on a finding of prejudice arising from the evidence of the
complainant’s flashbacks and recovered memories, which was not supported by any scientific
evidence, and the fact that the defence had no prior notification of the provenance of these
memories.
136 R. Dworkin, Taking Rights Seriously (Harvard University Press: Cambridge MA, 1977) 8–13.
rationally accept.137 Unfortunately, therefore, until science can offer better
evidence of the reliability of recovered memories, there is no other option
available consistent with the defendant’s right to due process, the presumption of
innocence and the principled asymmetry at the heart of the criminal justice
system.138
92 THE INTERNATIONAL JOURNAL OF EVIDENCE & PROOF
RECOVERED MEMORY IN HISTORIC CHILD SEXUAL ABUSE CASES
137 T. Ward, ‘Experts, Juries, and Witch-Hunts: From Fitzjames Stephen to Angela Cannings’ (2004)
31(3) Journal of Law and Society 369 at 386.
138 ‘In accordance with the values on which our system of justice rests, the acquittal of the guilty is
not of the same order of injustice as the conviction of the innocent’: Fitzgerald vDPP [2003] 3 IR 247
at 258, per Keane J.
To continue reading
Request your trial