1 Garden CourtHind CourtCafcass advertimage of 4 Paper Buildings logoCoram ChambersHarcourt ChambersGarden CourtDNA Legalsite by Zehuti

ABE Interviews: Is the Child's ‘Best Evidence’ being achieved in alleged sexual abuse cases? (Part 1)

Brenda Robinson, a child forensic specialist, presents preliminary findings arising from her research into the practice and effectiveness of Achieving Best Evidence in child sexual abuse cases. This first part sets out the guidelines and components for ABE interviews. Part II, to be published in August, will assess how successfully these are applied in practice.

Brenda Robinson*1

The Criminal Justice Act of 1991 determined that the child’s evidence in chief could be presented at trial by means of a videotaped interview held by a police officer and a social worker, the conduct of which was guided firstly by The Memorandum of Good Practice, [2]   subsequently updated by Achieving Best Evidence.  In the late 1980s and early 1990s, Professor Ray Bull and his colleagues incorporated experts’ recommendations;[4]   many of the existing research findings, and the recommendations of The Cleveland Report [5]   into the guidance offered in The Memorandum. As a result, The Memorandum (and subsequently ABE) addressed a range of interview preparation issues, offered advice on basic questioning techniques, and recommended the use of a four-phased approach for the interview itself, to include rapport, free narrative account from the child, questioning and closure.

It is now some fifteen years since The Memorandum was implemented, and six since the guidance was revised by ABE. To what extent is children’s ‘best evidence’ regarding alleged sexual abuse currently being achieved during ABE interviews?

There have certainly been a number of impressive (and very extensive) developments and debates in the research field, particularly on children’s memory, suggestibility and recall abilities.[6]  At the same time, contributions from experts in the fields of attachment and abuse related trauma have given a number of fascinating insights into the effects these may have on the developing child.[7]   Both areas provide an enormously rich context within which to discuss the reliability of children’s evidence in alleged sexual abuse cases, and the knowledge and skills needed by competent interviewers. Currently however, the author suspects that there is a widening chasm between the increasingly sophisticated content of academic debate, and the reality of practice ‘on the ground.’

The article attempts to stimulate debate about children’s ‘best evidence’ by posing a number of questions. Firstly, to what extent is existing research about children’s reliability relevant to cases of alleged sexual abuse? Secondly, is the current guidance in ABE relevant to such cases? Thirdly, what does a competent ABE interview look like? Fourthly, to what extent do we see this competence in practice? Lastly, to what extent, and in what ways, might poor conduct of the interview have implications for the ways in which the reliability of the child’s testimony is perceived – particularly in cases of alleged sexual abuse? This last part will be tackled in Part II of this article

In discussing these questions the author makes reference to literature and research, as well as factors which have emerged from her analysis of 140 ABE interviews in Children Act proceedings, all of which concerned alleged sexual offences.

The Relevance of Research
Much of the research into children’s suggestibility, memory and truthfulness has been conducted with children who have not been sexually abused, largely because any such study would involve profound methodological and ethical difficulties. Many authors have therefore questioned the relevance or ‘ecological validity’ of such research.[8]   For example, a good deal of the research into children’s truthfulness and deception has been carried out with non-traumatised children under the age of 5 years, usually in neutral laboratory experiments where ‘ordinary’ children are shown slides or told stories and asked to remember them. Specific conclusions are then sometimes generalised, to present unreliable ‘research findings’. For example, research findings on pre-school children may be broadened to include those of all ages, and conclusions from experiments in neutral settings are generalised to ‘fit’ child abuse cases. As Goodman and Clarke Stewart state:[9] 

‘Most research on children as eyewitnesses has relied on situations that are very different from the personal involvement and trauma of sexual abuse.’

Some researchers have attempted to stage ‘real life’ events in experimental settings in order to test children’s eyewitness accounts. These range from witnessing staged thefts, to arguments between adults.[10]  Further studies that focus on children’s suggestibility for ‘real world’ events have also produced mixed results. Some find a resistance to suggestion [11]  whilst others conclude that young, mainly preschool, children are suggestible.[12]  Other researchers have designed their studies around naturally occurring stressful events, in order better to approximate real life abuse situations. Children’s memory for salient, meaningful events has been found to be good over extended periods of time in studies involving: a visit to the dentist;[13]  a physical examination,[14]   and having an inoculation.[15]  Further, several researchers have attempted to study the impact of ‘real life’ trauma as it relates to the reliability of children’s recall of the event. [16]  These studies were confined to ‘one-off’ traumas however.

The Relevance of ABE to Sexual Abuse Cases
As stated above, much of the existing research was incorporated into the original Memorandum guidance on interviewing children. However, the advised approach had to [17]  cover a wide range of situations, e.g. from the ‘ordinary’ child who has witnessed a physical assault, through to the suspected victim of longer-term sexual abuse by a key attachment figure. It is suggested that the guidance contained in both the Memorandum and ABE is rather ‘catch all’ in nature. Indeed, Professor Ray Bull (the author of the Memorandum) has been quite explicit in describing the guidance as being ‘on how to do the easy ones.’[18]  Neither The Memorandum nor ABE contain any specific guidance on interviewing children who allege sexual assault, and who may be traumatised. At the same time, research (and much anecdotal evidence) suggests that the majority of interviews are not carried out with the ‘easy ones’ but rather, with children who do allege sexual assault.[19]   It is suggested that interviewing the latter requires theoretical knowledge, specific preparation skills, questioning techniques and style which go far beyond the content of the current guidance.

Equally, ABE contains little advice as to the nature of the specific forensic preparation required in alleged sexual abuse cases, and the ways in which this may be translated into questions during the interview. Interviewers are exhorted to – ‘keep an open mind as to what may or may not have happened to the child’[20]   but are not given specific advice as to how this might translate into practice. Open minded hypothesis generation (during interview preparation) can be a particularly useful means of exploring all of the possibilities for the allegations, for example that the child is lying; that he or she has been coached; that the allegation concerns another person rather than the one named, or that the child is speaking truthfully. Such a wide range of hypotheses should then form the basis for questions during the relevant part of the interview, so that the child has an opportunity to respond in a ‘friendly’ environment, rather than first hearing such questions during cross examination in any subsequent criminal proceedings. Indeed, Poole and Lamb [21]   explicitly state that forensic interviewing should be hypothesis-testing rather than hypothesis-confirming, whilst at the same time retaining a child centred approach.  For example, instead of asking the rather accusatory – ‘Are you sure you’re not making this all up?’, interviewers should maintain neutrality by ‘disconnecting’ themselves from the question – ‘Do you know, some people might say that you have made this up. What would you say to them?’ Interviewers should also consider questions which relate to the standards of proof needed for any alleged offence, and the possible defences to each, in their preparation.

Overview of a Competent ABE Interview
Before the interview takes place, the ABE guidance requires interviewers to engage in multi-disciplinary planning; consider a wide range of factors when preparing for the interview; decide whether a psychological or psychiatric assessment of the child is required pre-interview, and consider how the content, structure and rules of the interview will be explained to the child.  [22]

Guidance for the interview itself includes the four phased structure and basic questioning techniques. Within this, there is a particular emphasis on the importance of eliciting the child’s free narrative account about any alleged offences.

Planning and preparation
At a very basic level, ABE requires the interviewers to meet and consider a number of factors prior to the interview with the child. These range from the broad to the specific, but include a number of factors pertaining to the individual child, his or her family, and background, for example – the child’s age, race, culture, and use of language; his or her religion; issues of gender and sexuality; any special needs; the child’s cognitive, memory and linguistic abilities; his or her current emotional state and range of behaviours; relationships with family members; the child’s sex education and sexual knowledge; family routines, the use of discipline and the presence of any recent stress.[23]   Interviewers should avoid discussing substantive issues and must not lead the child on these matters. At the same time however, ABE advises that interviewers should never stop a child who is freely recalling significant events. A full written record of the discussion must be made.[24] 

ABE also advises that an assessment by the interviewers should be considered for ‘any child’ and infers that this may be the day before the interview. It suggests that the following factors be explored – the child’s preferred mode of address; his or her willingness to talk within a formal interview; an explanation of the reason for the interview; the ground rules and an opportunity to practice answering open questions; the child’s cognitive, social and emotional development; the child’s use of language and understanding of concepts such as time and age; whether the child has any special requirements or mental health problem, and whether the child is competent to give consent to the interview and any medical examination.[25]   Further, ABE states that ‘interviewers must also take steps to prepare the child for the interview itself.’ [26] 

Following the author’s communications with Professor Graham Davies (the lead author of ABE) in 2000, interviewers are further guided to consider a number of ‘abuse trauma variables’ before the interview, which appear in the revised guidance as ‘additional factors to be considered.’[27]   These are: the detailed nature of the child’s attachment to his or her parents; the age and developmental level of the child at the onset of abuse; its duration and frequency; whether different forms of abuse co-exist; the relationship of the child to the abuser (s); the type and severity of the abusive act(s); the existence of multiple abusers; the degree of physical violence and aggression used; whether the child was coerced into reciprocating sexual acts; the existence of adult or peer supports; whether or not the child has been able to tell; the parental reaction to the disclosure or allegation, and the nature of any previous interventions.

Although not made explicit in ABE, it is suggested that a securely attached, well-parented child who is subject to a ‘one-off’ sexual assault by a stranger may well be able to speak more freely and spontaneously during the interview than the child who has been subjected to longer-term sexual abuse by a key attachment figure. Consideration of the number of ‘additional factors’ which may, or may not, be present in the case (and the ways in which they interact if present), is intended to assist the interviewers’ development of the appropriate strategies and questioning techniques in the individual case.

Taken together, these four different levels (i.e. the planning of wider issues; the interviewers’ preparation of strategies, questions and content; a general assessment of the child, and preparing the child for the interview) would appear to indicate the need for a planning and preparation period by the interviewers of several hours. Moreover, planning is explicitly stated as a necessary component of ABE interviews – ‘thorough planning is essential to a successful investigation and interview.’[28]   The extent to which this appears to occur in practice is examined below.

Phase One of the Interview: Rapport
ABE suggests that the interview should begin with rapport – i.e. some discussion of a neutral topic to relax and settle the child. A further benefit is that an interview ‘rhythm’ can be established where the child does most of the talking and the interviewer does most of the listening. Indeed, Poole and Lamb advise that long lists of questions at this initial stage may teach children that the interviewer will do most of the talking and that their ‘job’ is to provide short answers.  [29]  Further, ABE could not be more explicit about the role of the interviewer at the beginning of the interview:

‘It is important that the child be encouraged in the Rapport phase to talk freely through the extensive use of open-ended questions; a stream of questions which the child can answer with ‘yes’ or ‘no’ or some equally brief response should be avoided.[30] 

Interestingly, research by Sternberg et al [31]  has found that where initial open-ended rapport building existed, the amount and quality of the child’s responses to subsequent offence-related questions was much improved as a result.

Following the rapport discussion, the interviewer can progress conversationally to the requirements at the end of the rapport phase, i.e. an explanation of the ‘ground rules’, the ‘truth and lies’ test (or ‘ceremony’) and the reason for the interview. The ABE advises against the general ‘do you know the difference between truth and lies’[32]   and is also clear that the chosen examples must include the intent to deceive. Again, competent interviewers will tailor and connect the examples chosen from the content of the rapport discussion. Words such as ‘imagine’, ‘let’s pretend’ and ‘little story’ should be avoided as they suggest fantasy or play.[33]  The interviewer should then ‘bridge’ the Rapport Phase to the Free Narrative by stating the purpose of the interview. Words such ‘why do you think you are here today?’ may be suitable for older children, but younger ones may need a more specific ‘cueing’ (yet non-leading) statement of purpose.

Phase Two of the Interview: Free Narrative
The Memorandum and ABE describe this phase as ‘the heart’ and ‘the core’ of the interview respectively, and the child should be encouraged to provide a free account of the relevant events which is as free from interviewer influence as possible. Very careful (and previously planned) open questions are essential in order to prompt and sustain the child’s free recall. Indeed, ABE explicitly stipulates that only the most general, open-ended questions should be put by interviewers at this point:

‘During this phase, the interviewer’s role is that of a facilitator, not an interrogator…The free narrative phase should never be curtailed by jumping into questions too soon. Every effort should be made to obtain information from the child which is spontaneous and free from the interviewer’s influence’. [34]

At the risk of stating the obvious, the interviewer’s task is to elicit and sustain as much free narrative as possible if the child makes mention of an alleged sexual offence. Simple ‘prompts’ such as ‘so tell me all about that – right from the start to the finish?’ or ‘and then what happened’ can often be used to great effect. Where the child alleges that offences have happened at different times, the interviewer will need to encourage free narrative about one of these in the interests of clarity. The (seemingly common) lack of attention to such a technique may, in some cases, contribute to inconsistencies in the child’s account, and this issue is discussed in more detail below.

ABE guidance also emphasises the need to secure as full and comprehensive an account as possible in the child’s own words. The detailed content of such a free account may have important implications for any subsequent assessment of the child’s reliability. Although not stated in ABE, certain features have been associated with truthful accounts, for example: descriptions of interactions; reproduction of conversation; unexpected complications; peculiarities of content; unusual details; superfluous details; accurately reported details which are misunderstood; related external associations; subjective mental states (feelings of disgust or fear), and attribution of the perpetrator’s mental state (description of emotions, cognitions and motivations). Some authors suggest that the more such factors are interwoven in the same sentence or paragraph, the more likely the child is telling the truth.[35]  Further, ABE advises that:

‘The child should not at this stage be interrupted to ask for additional details or to clarify ambiguities: this can be done in the questioning phase.’[36] 

Phase Three of the Interview: Questioning
The questioning phase seeks to clarify what the child has said in his or her free narrative account. This is the place for questions which address any evidential matters, the contextual and peripheral detail of any alleged offence, and clarification (of body parts or times for example) Importantly however, the content of such questions should relate to the child’s age and developmental level. Children under the age of eight or nine years may have difficulty in stating times, dates, days of the week [37]  and errors may be used as grounds to challenge the overall reliability of the child’s account in Court. Any alleged criminal offences also need to be examined during this phase, and specific questions put to the child which satisfy standards of proof for the offence(s), and which address possible defences to the offence(s). Even here, the interviewer should select open-ended questions (those beginning with ‘where, what, who and when’) rather than more directive questions whenever possible.[38]  

Phase Four of the Interview: Close
A further summing up of the key evidential points made by the child is useful at this point, together with a return to more neutral topics to allow the child some ‘recovery time’. Several issues may need to be addressed which do not relate to alleged offences and interviewers should be mindful of which issues need to be covered on the videotape, and those which can be safely addressed with the child once the tape is switched off.

1* Currently undertaking doctoral research on the ABE interviewing of traumatised children at The School of Law, University of Sussex, supervised by Professor Jennifer Temkin and Jane Fortin
2  The Memorandum of Good Practice on Video Recorded Interviews with Child Witnesses for Criminal Proceedings. The Home Office and The Department of Health,1992.
3   Achieving Best Evidence in Criminal Proceedings: Guidance for Vulnerable or Intimidated Witnesses, Including Children .The Crown Prosecution Service, 2001.
4  For example those of Dr David Jones, Dr Eileen Vizard and Dr John Yuille.
5  Report of the Inquiry into Child Abuse in Cleveland. HMSO, 1987.
6  See for example H. L. Westcott, G. M. Davies and R. H. C. Bull (eds) Children’s Testimony. A Handbook of Psychological Research and Forensic Practice. Wiley, 2002.
7  For example A.N. Schore Affect Regulation and the Origin of the Self. The Neurobiology of Emotional Development Lawrence Erlbaum Associates,1994;
D. J. Siegel, The Developing Mind. How Relationships and The Brain Shape Who We Are. The Guilford Press, 1999, and
B.A. van der Kolk Psychological Trauma. American Psychiatric Press, 1987.
8  See for example, J. Doris (ed) The Suggestibility of Children’s Recollections. Implications for Eyewitness Testimony American Psychological Association Washington DC,1991.
9  G.S. Goodman and A. Clarke-Stewart, ‘Suggestibility in children’s testimony: implications for sexual abuse investigations’ ibid at p92
10 J. Brigham, M. Vanverst and R. Bothwell, ‘Accuracy of children’s eyewitness identifications in a field setting’, Basic and Applied Social Psychology 7 (1986)
M. A. King and J. C. Yuille ‘Suggestibility and the child witness’ in S. J. Ceci, M. P. Toglia and D. F. Ross (eds) Children’s Eyewitness Memory .New York: Springer Verlag,1987; 
D. Peters “The influence of stress and arousal on the child witness in J. Doris (ed) op cit at n. 8
11 J. Brigham, M. Vanverst and R. Bothwell ibid
12  H. R. Dent, ‘Experimental studies of interviewing child witnesses’, ibid
13  D. Peters, ‘The influence of stress and arousal on the child witness’ stress.’ ibid
14  P.A. Ornstein and C. Brainherd, ‘Children’s memory for witnessed events: The developmental backdrop.’ ibid
15  Op cit at n. 13
16  L. Terr, ‘What happens to early memories of trauma? A study of twenty children under age five at the time of documented traumatic events’ Journal of the American Academy of Child and Adolescent Psychiatry 27 (1988)  ;
R. S. Pynoos and S. Eth, ‘The child as witness to homicide’ Journal of Social Issues 40 (1984)
17  Professor Ray Bull advised that this was as a result of the specific political context in a meeting with the author (29.6.07)
18   R. Bull and G Davies ‘The effect of child witness research on legislation in Great Britain’ in B.L. Bottoms and G.S Goodman (eds) International Perspectives on Child Abuse and Children’s Testimony. Psychological Research and Law Sage Publications, 1996 at p 102
19 G. M. Davies and H. Westcott ‘The child witness in the courtroom: empowerment or protection?’  In M.S. Zaragoza, J.R. Graham, G.C.N. Hall, R. Hirschman and Y.S. Ben-Porath (eds) Memory and Testimony in the Child Witness. California: Sage Publications, 1995. The study of 100 cases at trial revealed that 94% concerned alleged child sexual abuse.
20  Op cit at n. 3,  at p 10.
21   D. A. Poole and M. E. Lamb. Investigative Interviews of Children. American Psychological Association,1998 at p 109.
22  Op cit at n. 3 p 26
23  Op cit at n. 3 at p 20
24  Op cit at n.3 p 22
25  ibid at p 22
26  ibid at p 26
27  ibid at p 21
28  ibid at p 9
29  Op cit at n 20, at p 123-124
30  Op cit at n. 3 at p 40
31  K.L. Sternberg, M. E. Lamb, I. Hershkowitz, L. Yudilevitch, Y. Orbach, P. W. Esplin and M. Hovav ‘Effects of introductory style on children’s abilities to describe experiences of sexual abuse’Child Abuse and Neglect, 21 (1997)
32  Op cit at n.3 at p 37
33  Op cit at n. 20 at p 121
34  Op cit at n. 3 at p 41
35  Vrij, A., ‘Criteria-based content analysis: A qualitative review of the first 37 studies’ Psychology, Public Policy, and Law, 11, (2005) 3-41.
36  Op cit at n. 3 at p 41
37  For a useful review of these issues see K. J. Saywitz ‘Improving children’s testimony. The question, the answer and the environment’ in M. Zaragoza et al, op cit at n. 19
38  Op cit at n. 20