Volume 8, No. 1, Art. 14 – January 2007

Matters of Suggestibility, Memory and Time: Child Witnesses in Court and What Really Happened

Johanna F. Motzkau

Abstract: As a result of an increasing awareness of child abuse over the last few decades, children have been admitted as court witnesses more frequently, yet there has been persistent wariness about the reliability of their testimony. Examining the interaction of legal rationales and paradigms of developmental psychology, it would appear that children are still frequently positioned as deficient and passive witnesses. Three tropes can be distinguished: 1. Children are positioned as unreliable containers of facts. 2. Children have proved to be irritable dispensers of information. 3. Children are volatile interactants. In this paper I will examine how the English legal system employs special measures that are designed to manage children's apparent deficiencies while guaranteeing the accuracy and admissibility of their evidence. My analysis unfolds around the specific case of video recorded evidence. Using courtroom observations and data from interviews with legal professionals, I will follow the trajectory of the video from its planning and recording by the police to its presentation in court. Inspired by the work of Isabelle STENGERS and Bruno LATOUR, and drawing on discourse analytical tools, I will show that the collision of the different time zones of veridicality creates circumstances under which the video itself can become an ambiguous agent and ultimately a fanciful witness.

Key words: time zones of veridicality, suggestibility, child witnesses, memory, credibility, sexual abuse, time and discourse

Table of Contents

1. Introduction

2. Situating Suggestibility, Child Witnesses and Sexual Abuse

3. Child Witnesses and Special Measures in England

3.1 What do special measures do?!—Mediating absences and presences, matter and time

4. The Case of Video Recorded Evidence: Creating a Reliable Witness

4.1 Time zones of veridicality: absences, presences, accountability

4.2 Time zones, duration and sequentiality: what really happened …

5. Epilogue: Purification, Tape-Fetishism and What Really Really Happened

Acknowledgements

Notes

References

Author

Citation

 

1. Introduction

Judge: And if you do always tell the truth where will you go when you die?

Girl: Up to heaven sir.

Judge: And what will become of you if you tell lies?

Girl: Shall go down to the naughty place, sir.

Judge: Are you quite sure of that?

Girl: Yes, sir.

Judge: Let her be sworn, it is quite clear that she knows more than I do!

Exchange reported to have taken place when Mr Justice MAULE (1788-1858) had to examine a little girl on the nature of the oath, quote as SPENCER and FLIN (1993, p.51).

Children and the law have always been an awkward couple. Since the late 1970's, and as a result of an increasing awareness for child sexual abuse, children have been admitted as court witnesses more frequently, yet there has been persistent wariness about the reliability of children's testimony and about their capability to comply with the strict nature of evidential procedures in criminal court. In this article I will explore how the English legal system employs various artefacts and special procedures that are designed to accommodate children's assumed needs in order to enable them to give the most detailed evidence possible while guaranteeing the objective nature of the way in which the evidence is gathered and presented in court. I would first like to establish the historical and theoretical context of issues around sexual abuse, suggestibility and children as witnesses and delineate their close interrelationship. My analysis will then unfold around the specific case of video recorded police interviews with children. Using data from courtroom observations and excerpts from interviews I conducted with legal professionals, I will follow the trajectory of the video from its planning and recording at the police station, to its final presentation in court. It will become clear that, as this video is created and later on performs at the different stages of the legal procedure, it resonates with different assumptions and frameworks that structure the way in which for example material presences or absences, temporal immediacy or mediatedness, are seen to bolster or indeed discredit the credibility of a statement. I will argue that those actual particulars that constitute credibility and thus an assumption of truth, differ relative to the time and stage during the process at which they are considered. This phenomenon could be expressed as constituting different time zones of veridicality. Hence taking a close look at the concrete practice it will become clear how this video recorded evidence is created in view of-, relates to and mediates different time zones of veridicality. Inspired by the work of Isabelle STENGERS (STENGERS, 1997; CHERTOK & STENGERS, 1992) and Bruno LATOUR (1987, 1999), and drawing on discourse analytical tools, I will show how the collision of the different time zones of veridicality within the static rules for the production and performance of admissible evidence, create circumstances under which the video itself can become an agent in its own right and ultimately a fanciful witness.1) [1]

This wide reference to science studies and discourse analysis indicates the syncretic methods perspective I am adopting for this paper. While firmly rooted within the qualitative social sciences, for my specific analytic focus it proved most productive to draw on a variety of tools and perspectives derived from different methodological backgrounds. As this is not an entirely unproblematic move, I would like to offer a methodological disclaimer before I specify these methods any further and embark on my analysis. It has, rightly, been pointed out that there is an inherent danger in methodological eclecticism where discourse analysis is concerned, because some of the traditions that operate with the term discourse

"… may be incommensurable in their theoretical and methodological stances, including the conflicting roles assigned to the analyst. […] Indeed, the reason Edley and Wetherell (1997) call for a deténte is that the terms 'discourse' and discourse analysis are fiercely contested" (MC MARTIN, 1999, p.511). [2]

My proposition to additionally draw on ideas derived from science studies, arguably makes the issue even more complicated. MC MARTIN's article is a crucial point of reference for my inquiry, because she examines a problem very closely related to the one discussed in this paper. She suggests looking at children's disclosure of sexual abuse as discourse and outlines the possible contribution different discourse analytical approaches could make to understanding disclosure. I fully appreciate MC MARTIN's warning about the possible incommensurability of such approaches; after all it is the fiercely contested, if not insoluble, issue of (epistemic) constructionism versus (critical) realism that underpins this controversy for example in discursive approaches to psychology (EDWARDS, ASHMORE & POTTER, 1995; EDWARDS, 1997; PARKER et al, 1997; PARKER, 2002). Notwithstanding the importance of such debates, I would like to follow my specific analytic endeavour in a wider, more exploratory spirit. I will approach the interview data by selectively drawing on some of the analytical tools that are associated with discourse analysis and more specifically conversation analysis and ethnomethodology (GARFINKEL, 1967; SACKS, 1992; POTTER & WETHERELL, 1987). Still, the overall analysis is embedded in a critical account of the historical, political and cultural discourses that underpin and shape the acute reality of child witness practice and childhood as such, positioning my inquiry more closely to what PARKER termed "critical discursive research" (PARKER, 2002; PARKER, 1992; BURMAN, 1994). Yet, ultimately it is the theoretical and reflexive spirit of the philosophy of science and science studies (STENGERS, 1997; LATOUR, 1987) that guides my inquiry into the different time zones of veridicality, because my implicit, underlying agenda is to examine the concrete possibilities emerging from what LATOUR expresses so pointedly in his comment on the debates around discourse, relativism and realism.

"So we do not have to choose between realism and social construction because we should try to imagine a sort of mix up between the two ill-fated positions. Rather we have to decide between two philosophies: one in which construction and reality are opposite, and another in which constructing and realising are synonymous" (LATOUR, 1997, xiv). [3]

2. Situating Suggestibility, Child Witnesses and Sexual Abuse

Children have traditionally enjoyed a dubious reputation as witnesses in courts of law. In order to contextualise my analysis I would like to begin by recapturing the long-standing and intricate connection between issues of childhood sexual abuse, child witnessing and suggestibility. [4]

One of the most notorious cases that is still quoted with an astonishing frequency as a warning tale about the dangers of children giving evidence in court, are the Salem witch trials, that go back as far as the 17th century. These trials, held in Salem, Massachusetts in 1692, have since acquired the status of an emblematic (and infamous) example for the unreliability of children's evidence.2) Salem saw 19 villagers executed as a result of the evidence given by a group of children, even though the children had at later points during the trial attempted to recant their accusations. CECI and BRUCK (1995) suspect that this event is partly responsible for the long standing and deeply rooted distrust in child witnesses.3) In more recent times this distrust has caused considerable difficulties particularly for the prosecution of child sexual abuse, which is the most frequent reason for children to be called as witnesses in criminal courts. Perceived as a social phenomenon, the concept of child sexual abuse has only formed within the last few hundred years4) and, as HAAKEN (1998) puts it, has since had a long history of being forgotten and remembered. In the 18th and 19th century there are numerous documented cases and scientific references that reflect awareness for child sexual abuse all over Europe. Yet around the turn of the 20th century, and coinciding with Sigmund FREUD's famous public abandoning of the "seduction theory" in 1897,5) there was a growing suspicion about reports of child sexual abuse and over time the awareness for the problem waned completely (DALGLEISH & MORANT, 2001). This only changed in the late 1960's and 70's, when the campaigning effort of the growing feminist movements gained political momentum and managed to put issues like domestic violence, rape and child sexual abuse back into the public arena (HAAKEN, 1998). The growing awareness for child sexual abuse ultimately forced a reconsideration of the general distrust that had so far mostly excluded children from testifying as witnesses in court. The United States and England for example had strict corroboration laws that allowed children's testimony only if it was corroborated by an adult eyewitness (SPENCER & FLIN, 1993).6) Additionally in England, children's competency to take the oath had to be assessed by the judge, as the introductory quote exemplifies. The long-standing tendency to exclude children from giving evidence, or to hear them only in very exceptional cases, was echoed in frequent references to the Salem witch trials, and at the time it was generally supported by contemporary paradigms of developmental psychology and memory research.7) Yet, due to the changing climate towards violence against children, during the 80's it was now acknowledged that the prosecution of child sexual abuse was almost impossible as long as children's evidence was not admissible. This resulted in first changes in the juridical systems in Northern America and some European countries, which now allowed for example children's uncorroborated testimony in cases of alleged sexual abuse. As a result of this during the 1980s courts were faced with an increasing number of child witnesses who were predominantly testifying as the presumed victims of alleged sexual assaults. In the following years Northern America, but also Europe (e.g. England and Germany) saw a number of high profile miscarriages of justice around alleged child abuse cases. In some of the most notorious cases children were reporting the most bizarre scenarios of ritualistic abuse by multiple perpetrators. Yet, their accounts were later identified as the product of coercive and suggestive interviewing techniques used by investigators and parents. Driven by the climate of intense concern about child abuse, parents and professionals had been absolutely convinced something must have happened, and this spurred their (well intentioned) eagerness to get a disclosure.8) These cases sparked a sudden and intense research interest in children's suggestibility, a topic that had so far not been on the scientific agenda at all. So now a huge number of studies looked at the degree to which children can be swayed in their statements and can be made to believe and report events they have never experienced.9) While numerous studies underlined children's general ability to robustly and accurately testify (GOODMAN & CLARKE-STEWARD, 1991), the attention of the media was caught by research that reported that children were suggestible, and that they could indeed be made to believe and report things they had never experienced (CECI, HUFFMAN, SMITH & LOFTUS, 1994). Overall this research raised methodological as well as ethical questions and it caused heated public and academic debates (GOODMAN, QUAS & REDLICH, 1998; CECI, 1998). This publicity combined with the media coverage of the miscarriages of justice, ultimately fuelled a rather simplistic and scandalised understanding of suggestibility research and of children's vulnerability to suggestion. Unfortunately this also filtered back into legal decision-making, reinstating a general atmosphere of disbelief in children's testimony, making the prosecution of sexual abuse even more difficult. Additionally researchers warned that this atmosphere could result in a recurring fear in children or parents to report abuse because they felt that nobody would believe them (GOODMAN et al., 1998). Hence despite the initially established trust in children's ability accurately to remember events and to report them in court, there has been a persistent wariness about the reliability of children's memory and their ability to testify in court. So it is clear that issues of children's suggestibility, their credibility and the problem of preventing, detecting and prosecuting child sexual abuse are inseparably linked (see also MOTZKAU, 2005b). [5]

Given this history and context it also becomes clear that the way in which legal rationales and paradigms of developmental psychology interact around children, positions children as deficient and passive protagonists in legal procedure.10) Here three central tropes can be identified. Firstly, children are positioned as bad and unreliable containers of facts. Information is seen to degenerate quickly in their minds and thus has to be retrieved as quickly as possible. Secondly, children are seen be irritable dispensers of information/evidence, as they are prone to misunderstand questions and get confused and frightened by legal procedure. Hence they need to be treated delicately and with great care when questioned. Thirdly children appear as volatile interactants, that is, direct interaction with them bristles with reciprocal effects and hazards of suggestion. As I develop my analysis I will give some concrete examples of these tropes as they are reflected within the practices that surround and condition of child witnessing. [6]

At this point it is worth noting that on a much more abstract level the problematic position of children as witnesses is also reflected in international law. Since 1989 article 12 of the UN convention of children's rights establishes unequivocally that children need to be given a voice in legal proceedings concerning them and that their statement needs to be given due weight.

"1. States Parties shall assure to the child who is capable of forming his or her own views the right to express those views freely in all matters affecting the child, the views of the child being given due weight in accordance with the age and maturity of the child.

2. For this purposes the child shall in particular be provided the opportunity to be heard in any judicial and administrative proceedings affecting the child, either directly, or through a representative or an appropriate body, in a manner consistent with the procedural rules of national law" (General Assembly of the UN, 1989). [7]

On closer inspection however, it becomes clear that this article contains a twofold conditional clause. Firstly, the "capability to form an own opinion" and "age and maturity" have to be established with regard to each particular child, and there need to be criteria for judging this "maturity", which will have to be gained from the particular context. So firstly this is a question of developmental psychology. Secondly, the "opportunity" provided to be heard is even more open to interpretation as it can be realised following the existing rules of the respective national law. So secondly this poses a concrete legal procedural problem. LEE (2001) argues that this apparent ambivalence towards children, that seems implicitly to invalidate the very promise it is attempting to make, was an unavoidable ingredient of this law at the time of its creation. If article 12 functions, LEE states, it functions to generate and expose, rather than to resolve, the otherwise implicit childhood ambiguity and "then lays the responsibility for managing that ambiguity on the legislatures and policy-makers of the states that have ratified it" (LEE, 2001, p.96). In this sense, I would argue, that for the national law the central issue remains a problem situated somewhere between psychology and law, between psychological and legal truths; and it has to be managed locally, via the practices implemented to handle, negotiate and contain such ambiguity. [8]

Referring back to my outline of the history of children and the law, it is clear that what LEE terms childhood ambiguity is indissociably linked to societal and scientific issues around the acute and serious problem of child sexual abuse on the one hand, and on the other hand the perplexing issue of children's credibility and suggestibility. This is the backdrop against which the local practices of the national law have to operate and try to address the problem of giving children a voice in the face of childhood ambiguity. [9]

In the following I would like to examine the concrete efforts of the English legal system to substantiate the abstract demands set out by the convention of children's rights in the explicit attempt to give children access to justice and particularly to enable a more effective prosecution of child sexual abuse. [10]

3. Child Witnesses and Special Measures in England

Let me give a brief introduction to the more recent changes in the law of evidence concerning children in England and Wales (in the following "England"). This should illustrate how closely psychological research interacts with legal reforms in this area.11) [11]

In 1988 "Criminal Justice Act 1988" abolished the corroboration rule, according to which children's evidence could only be considered in court if it was corroborated by an adult eyewitness. The "Criminal Evidence Act 1991" implemented some special measures for child witnesses, including the option for children to give evidence via closed circuit television and video recording of their initial interviews with the police (implemented provisionally). In collaboration with psychological researchers guidelines were produced that instructed the police how to conduct interviews with child witnesses ("Memorandum of Good Practice", Home Office 1992). The 1991 Act thereby followed some of the recommendations given by the Pigot committee (SPENCER, 1997; SPENCER & FLIN, 1993). This committee of high ranking legal professionals had been set up in 1988 to thoroughly review child witness practice in England in response to the scandal that erupted around the Cleveland crisis.12) Subsequently a group of researchers and academics from psychological and legal backgrounds was commissioned to review and examine the changes implemented in 1991, and in 1998 the white paper "Speaking up for Justice" (1998) was published. Following recommendations made in this paper the Youth Justice and Criminal Evidence Act 199913) was passed and provided an even more dramatic revision of the law of evidence and criminal procedure with respect to young offenders and witnesses (specially in child abuse cases).14) Most crucially article 53 of the act states that: "At every stage in the criminal proceedings all persons are (whatever their age) competent to give evidence". For my trajectory the crucial aspect of the "Youth Justice and Criminal Evidence Act 1999" are the special measures that were implemented for vulnerable and intimidated witnesses (applies to all witnesses under the age of 17).

Alongside the new Youth Justice and Criminal Evidence Act 1999, psychological and legal academics developed new guidelines to replace the "Memorandum of Good Practice". The result of this collaboration was a document called "Achieving best Evidence" (2001). It provides guidelines and instructions for training and practice of police officers, and it is designed to help implementing the new rules for child witness practice. The Achieving best Evidence (in the following ABE) constitutes what is called "soft law", hence it provides principle rules the police is strongly advised, but ultimately not obliged, to follow. [13]

At this point it should be noted that parallel to these changes in the legal system, the prosecution and conviction rates for cases of child sexual abuse in England have seen a steep decline. KELLY (2005) reports that there has been a steady decline in prosecution and conviction rates for child sexual abuse cases in England since 1977, with conviction rates dropping from 34% in 1977 to 5,5% in 2002.15) However, the most dramatic drop is reported to have occurred over the past decade. While statistics should always be met with suspicion, because a multitude of factors might be responsible for the drop in recorded convictions, these figures throw a peculiar light onto ongoing efforts to reform child witness practice in England. [14]

Before I take a more detailed look at the video recorded interview, let me take a step back and establish how these special measures generally relate to the principles underlying the rules of evidence. What do they do? [15]

3.1 What do special measures do?!—Mediating absences and presences, matter and time

Roughly speaking the law of evidence is based upon the principle right of the accused to confront their accuser in person, that is, immediately. This encounter needs to take place in open court and the accused should be able to cross-examine the accuser and witnesses there and then (or have their counsel cross-examine them on their behalf), in order to challenge the accusations. Furthermore the defendant has the right for this to happen in the presence of a group of the defendant's peers who will judge the case. All those material facts that are disputed must be proven by evidence, and here immediate oral evidence has precedence over other forms of evidence (for a summary see SPENCER & FLIN, 1993). Hence presence and immediacy play a crucial role for the finding of fact. It is the immediate clash of opposing accounts in front of the jury that is ultimately seen to enable the jury to determine the fact of the case. Special measures clearly run counter to these legal principles. Still, they attempt to mediate absences and presences in a way that preserves the rights of the accused and thus guarantees a fair trail. But simultaneously these artefacts and practices aim to facilitate the process of giving evidence for children by mediating and altering their necessary presence in the courtroom in a way that is perceived to be less intimidating to children. Beyond the obvious intention to accommodate what are perceived to be children's special requirements, the paramount legal argument in favour of this "softening" of principle rules of evidence, is the aim to achieve "best evidence". These measures, it is hoped, provide conditions under which children are most likely to provide the most detailed and accurate accounts of what really happened, because they feel less intimidated or confused. One could say that minimising fear or indeed minimising the possibility of fear being suggested to the child, serves as a form of informational hygiene for the production of better evidence. Let me give an overall sketch of how the special measures achieve this. [16]

Generally the measures are set out to mediate the child's physical, spatial or even temporal relationship to the courtroom. The screen (curtain) for example intercepts the visual space between the child and accused, thereby mediating their physical presence in the courtroom. While both, the child and the accused are physically (immediately) present, the screen makes the child invisible to the accused (and vice versa), but keeps it visible to the court and the jury, and audible to everyone in the courtroom. Hence the screen can be seen to filter information as to allow only those aspects assumed to promote good evidence and fact finding. Removing the public or the defendant from the courtroom mediates, and thus enables the child's presence in the courtroom via the absence of those who the child might perceive as intimidating. It alters the immediate physical structure or scene of the courtroom. In a similar sense the removal of wigs and gowns will change the immediate setting of the court for the child by making it less imposing and thus, arguably, less intimidating. Closed circuit television goes a step further by mediating across the spatial relations. It allows the child to be absent from the courtroom, by transmitting the child's televised image from a different room (or even building) onto a television screen in the courtroom. So in a sense the television screen takes the child's physical place in the witness box, as a witness that is hoped to resonate much less irritably with the courtroom than the child itself might do, and in reverse might be less ambiguous to the court.16) So here one could say that it is an element of potential mutual irritation that is filtered out by the mediation across space. Still, this is a life transmission, so while being physically absent the child still shares the actual immediate time frame of the ongoing trial and it will be cross-examined in real time by the barristers via the CCTV link. In this sense one can say that CCTV mediates children's voice within the same time zone, but across space. [17]

Video recorded evidence of children goes a step further even. The police will video record the investigative interview they conduct with the child as soon as possible after an incident has been reported. Subsequently this video will not just be viewed by the prosecution to decide whether to prosecute or drop the case, but in case of a trial it will also be played in court. For the trial this video will fully substitute the children's "evidence in chief". ("Evidence in chief" is the initial evidence given by a witness under the guidance of the lawyer who called them. It forms the basis of the subsequent cross-examination, which for a child will be conducted via CCTV.) Hence similar to the CCTV this video poses as a proxy witness on children's behalf. While they are absent the video mediates their presence, it mediates across space representing them in court as a reliable witness, not irritable or fearful. But this video additionally mediates children's voice across time. It preserves their voice from the initial interview and transports it to the time of the trial. This is meant to ensure that children's account is collected as early as possible, in order to stabilise it by preserving it on tape and thus making sure nothing can alter it or be forgotten over time (the time between reporting and trial may well exceed 10 months). In a sense one can say that this video transmits the immediacy of the initial interview by the police into the immediate presence of the courtroom, allowing these two time zones to overlap in order to enable the earliest possible account of the child to be presented and heard in court in an unambiguous and objective fashion, that is, delivered by a video tape. [18]

Let me reverse the gaze for a moment in order to return to the three tropes I have outlined earlier. Looking at these measures as they organise the performance of children's "best evidence", we can see how these measures also reflect children as passive and delicate, if not slightly deficient vessels of information. Children are positioned as unreliable containers of information. They cannot contain evidence and it degenerates quickly in their immature minds. Hence it needs to be preserved on video as soon as possible, so the video can become the carrier of the evidence on the child's behalf. Children are positioned as irritable dispensers of evidence. The concrete presence of a defence barrister, the defendant, wigs, gowns, jury members, the physical arrangement of the room etc. is seen to potentially alter their evidence. It can hamper the clarity or consistency of its production, or it might even destroy it altogether by silencing the children (or reducing them to tears). So it is seen as beneficial to intercept their performance with a CCTV transmission. One could say that the television screens, substituting the court personnel for the child and respectively substituting the child in the courtroom, are there to help purifying the child's performance so that the evidence can emerge unhampered by adverse influences. Ultimately children are also positioned as volatile interactants. The measures reflect the reciprocity of the problem, because any direct encounter with children is implicitly seen to bear the potential danger of unwitting misunderstandings and suggestion. The number of direct encounters is thus kept to a minimum, by using videos and television screens that are hoped to represent children in court unambiguously. And, as I will illustrate in more detail in the following, those who do directly interact with the children, the police officers, will have to be trained according to, and will need to abide by strict guidelines, in order to stabilise their encounter with the child. [19]

On the one hand we can see how the historically manifest suspicion against children as witnesses is implicitly reflected within these measures. Still, on the other hand it is important to remember that these special measures are the concrete and well intended attempt to handle and contain childhood ambiguity. They are intended to preserve, transport and amplify children's voices. [20]

4. The Case of Video Recorded Evidence: Creating a Reliable Witness

While all of these different measures call for a detailed analysis, here I can only focus on one of them. I would like to follow the trajectory of video recorded evidence as it is cutting across time and across space. I will do this by exploring in detail some of the discourses around its production and performance, in order to see how it actually functions within the legal system. [21]

In my general introduction to English child witness practice I have already outlined the legislative-, research- and policy-making effort that has gone into creating and implementing these new measures. But intense effort is also going into instructing and training police officers. They are the ones who encounter child witnesses directly, and they need to produce the video recorded evidence in a manner compliant with the law of evidence as well as following the guidelines on interviewing child witnesses outlined in the ABE. Hence I will pick up the video's trace where it initiates, that is, within the comments and discussions of police officers on a training course, where they learn how to conduct a video recorded interview following the guidelines of the ABE.17) [22]

4.1 Time zones of veridicality: absences, presences, accountability

Video interviews are conducted in a special video interview room at a police station that is equipped with cameras. Two officers are involved, one officer conducting the interview and another officer monitoring the interview and the equipment from a room next door. The following discussion takes place on the third day of the training course I attended. Over the previous days the officers have rehearsed the law of evidence because the interview needs to comply with courtroom procedures to be admissible in court. They have talked about interviewing techniques appropriate to children and about the possible problems they might encounter with regard to silencing children or suggesting something to them. At this point they have spent all morning learning that it is crucial to build a good rapport with the child in order to draw out detailed information while avoiding misunderstandings and suggestion. [23]

The following discussion ensues around section 2.99 in the ABE guidelines. The section states that comfort breaks should be allowed any time, but if children need to leave the room to go to the toilet, police officers have to accompany them to the toilet and make sure they do not speak to anybody (ABE article 2.99, p.35).

1

TO:18)

[…] but why do we do it? It's this issue about whether they've been what? conduced

2

 

cajoled (1) threatened that's the issue […] but I find that (3) a little bit a bit a bit you

3

 

know (1) it it it's a:::lmost like saying like you a::::::re hhffff °taking this person into

4

 

custody° ((PO1: hmm)) I would if I wanted to go to the toilet during the course of an

5

 

interview at a police station the last thing I would expect is for the p'liceofficer to follow

6

 

me out of the interview room to the toilet=

7

PO1:

if the tape's running whilst they go to the toilet (2) you're gonna be seen on the tape

8

 

anyway (1)

9

TO:

you're gonna be seen?

10

 

(1)

11

PO2:

you're gonna ruin the rapport

12

 

(1)

13

TO:

you're gonna ruin the rapport? by going to the toilet with them?=

14

PO2:

Yeah I think so it's quite intimidatin'

15

TO:

I think it's potential to offend them

16

PO2:

°yea°

17

 

[[...]]

18

TO:

I think (.) it's it's it's I just think that tha' we are fixated with th::iss uhm this fear of of

19

 

evidence being (.)=

20

PO1:

[°tainted°]

21

TO:

[bei]ng tainted or it o:r the evidence being or as bein' accu::sed o::f cajoling and tainting

22

 

the evidence b:y not being able to account for every single mo::ment that we are with

23

 

them ummm and I think I think it's easily suggested by any defence barrister tha' I you

24

 

know a defence barrister worth their weight in gold could easily suggest at any point

25

 

during the course of any proceedings that any witness comes in and deals with you that

26

 

at any point you've suggested something to them or induced them and said something

27

 

°to them° and we wouldn't have evidence to say anything otherwise other than no I

28

 

didn't ummm s:::o I I you know it says in the::re and that's what it says it says that you

29

 

should accompany them to the toilet and ensure they don't speak to an'body else

30

 

°etceteraetcetera° I just think that's a little bit (1) asking a bit too °°much°°

31

PO1:

but effectively your integrity is supposed to be intact isn't it because if anybody else=

32

TO:

why? 'cause you'r a policeofficer?=

33

PO1:

°yes° absolutely

34

 

(2.5)

35

TO:

°yea ok° I I'm you know (1) it's horses its its what the the the=

36

PO1:

I think you can stand up in court and answer all the questions that they're asking you

37

 

and you'll say no they weren't interfered with no this didn't happen that didn't happen

38

 

((TO: yes)) [umm]

39

TO:

[all the time] they were with me

40

PO1:

yes

41

TO:

I can say that nothing untoward occurred=

42

PO1:

°your honour°

43

 

(1)

44

TO:

°yea°

45

 

(1)

46

PO3:

°we'll have cameras in the toilets then next time°

47

TO:

Sorry?

48

PO3:

next thing an' we'll have cameras in the toilets

49

TO:

(chuckles)

[Excerpt 1] [24]

Let me begin by capturing the overall scope and tone of the discussion. There is a constant, almost circular shift between colliding issues, made even more peculiar by the fact that it is the training officer who explains the guidelines while simultaneously presenting caveats against them. Additionally, it seems, the officers need not only be concerned with what is on the video, but they should be even more worried about what is not on the video. The training officer states clearly why it is they should accompany witnesses to the toilet, and he delivers a neat three part list ("conduce, cajole or threaten" lines 1-2). But then this might "ruin the rapport" (11), intimidate or "offend" (15) the witness, which the officers must avoid under all circumstances because as a result of this they might not get any information from the witness. And they also should not be too obsessed with "tainting" the evidence, as the training officer finds, while he simultaneously goes on to imply that there indeed is a need to already plan how the video presents the witness in court, and how the video itself is presented. And that indeed means to be able to "account for every single moment" (22) and to perform their task in a legitimate way. And here one of the officers reminds him that this ultimately means to keep their own "integrity intact" (31), as well as the integrity of the interview. Yet this again collides with the duty to facilitate the disclosure because it might ruin the rapport if it means going to the toilet with the witness. Following this exchange, the production of such a video seems to be a paradoxical task. And the video, instead of being a plain record, a neutral objective witness by proxy, seems a rather ambiguous protagonist that in itself needs to be accounted for. Let me take a closer look and isolate three salient aspects reflected in this excerpt. [25]

Firstly, by introducing a sense of complete, unadulterated footage, total visibility, the video simultaneously highlights the relative lack of visibility around itself. By minutely accounting for one space, it also produces a totally unaccounted, invisible space around it, and this space now produces heightened sensitivity. And while at first it looks as if the video will also be a neutral witness on behalf of the police officers, confirming their good conduct as they are "seen on the tape" (7), the video finally turns out to be a rather ambiguous witness that produces more uncertainty than it resolves. This is reflected ironically in the statement concluding this debate. Ultimately it seems, the issue can only be resolved by extending footage and visibility endlessly, by making it unlimited, which means to install "cameras in the toilets", as one officer remarks. Interestingly this also evokes a sense of the video being a form of surveillance, rather than just a neutral record. It is eyeing the officers' activity suspiciously. [26]

This links up to the second aspect. Traditionally a police officer would have interviewed a child witness, recorded their statement in writing and summarised it for the file. There may have been concerns about the officers' conduct, but by and large the practice functioned based on the fact that the officers were reliable recorders of evidence by virtue of being police officers, by virtue of being part of the investigative legal machinery. In this sense one could say the officers constituted what LATOUR refers to as a "black box" (LATOUR, 1987), making it virtually (and practically) impossible to challenge their practice.19) Yet with the video recording this has now changed, the black box of the police officer as a reliable and neutral "evidence-gathering-machine" has now been opened and we get an intimate view of what is going on inside. Above and beyond the question whether this opening of the black box is a good thing or not, it is clear that as soon as we can see what they are doing, their actual performance becomes a matter of debate and strict rules are needed to regulate their conduct. Now the officers have to actively produce their integrity. They have to produce it at the very moment of the interview, but also for the prospective immediacy of the courtroom presentation of this video. And this links up to the third aspect. [27]

Thirdly, a closer look at this excerpt shows that the impression of paradoxicality we get when following the exchange, is also a result of the simultaneous efficacy of two different time zones. The paradoxicality arises from the officers attempts to simultaneously operate within, and with regard to, two different time zones. In the immediate time of planning and conducting the interview it is paramount to maintain a good rapport, because it is in this time that the child will talk to the officer and here only a good rapport will enable a detailed and accurate account to emerge and be recorded. Hence it is in this time that going to the toilet with them appears like taking them "into custody" (3) like a suspect. Here this is obviously contraindicated because it is likely to offend them, and it is in this time that it is indeed a sign of being fixated with a fear of the "evidence being tainted" (19-21) and thus it is indeed "asking a bit too much" (30). Yet, these utterances are simultaneously interspersed by considerations and comments that operate within, and project forward into a completely different time zone. And that is the time zone of the trial, where the video will again perform immediately, where courtroom procedure applies but where truth is generated by completely different means. This intermingling of time zones is particularly obvious within the training officer's turn from line 21 to 30, because here the utterances relating to the time of the interview literally form a frame around the intermediate illustration of what is representative of trial time. Following the extreme case scenario that the training officer unfolds step by step in lines 23 to 27, this time of the trial presents itself as a time zone where truth is guaranteed by the application of intense and systematic uncertainty. The whole paragraph consists of extreme case formulations, arranged to add to an escalating sense of arbitrariness, culminating in the reference to the guidelines. Initiating the account by the adverb "easily" and following this up with five mentions of "any", this almost creates a sense of loss of control, of a space devoid of direction and reason. One has to account for "every single moment" (22); and it can be "easily suggested by any defence barrister […] at any point […] during any proceedings that any witness comes in and deals with you that at any point you have suggested something to them … and you couldn't say anything …" (23-27). One could almost say the training officer implies in ironic hyperbole that it is ultimately pointless to even bother following procedure because a defence barrister could always destabilise their account. Yet, beyond the training officer's potential irony, this account reflects the implicit paradox of time, the exasperation of facing a task that seems virtually insoluble, because it means to operate at one moment but within two time zones; the time zone in which the interview is conducted and the time zone of the trial. Yet at this point, to resolve the problem, the guidelines are mentioned (28). And again the training officer's triple repeat of the reference to the guidelines, adds a rather ambiguous, incredulous tone to the reference: "it says in there and that's what it says it says …" (28). Still, it is clear that the guidelines are supposed to address the problem and they propose that every step along the way has to be rule based so that it can be accounted for. Therefore it is fundamental that the police officers' integrity is intact. And again we can see that this "integrity" is not given by virtue of their position as police officers. But the black box has been opened, and it is precisely because they are police officers (32-33), that their integrity has to be actively produced and affirmed. And again, it has to be produced at the time of the interview, but following trial rules, because it is produced in anticipation of courtroom appearance, where it can be displayed in the way the two officers demonstrate in an almost stage like exchange (36-42). The closely interweaved dialogue and the particular way in which turn taking develops could be analysed in much more detail, but I would like to trace the broader dynamic involved. So let me move on to examine how the video performs at other stages of the process. [28]

Accounting for every step along the way means that the so important initial stage of the interview, the rapport stage, needs to be recorded on video as well. Hence this rapport stage, where the officer will conduct an informal conversation with the child in order to establish an open and friendly atmosphere, will end up being shown in court as well. And here the training officer reports what a judge has said about the effect this has on the jury.

1

TO:

The only thing that judge Smith20) bless him says is that the jury switch off during the

2

 

rapport stage so whilst this book21) concentrates on how it is so important to get that

3

 

rapport in the interview he quite clearly said that they switch off=

4

PO1:

He's right he's absolutely right

5

TO:

you know s:::o you have to weigh up well that's what's our question perhaps if we can do

6

 

our rapport pre-interview assessment then you only need a shorter rapport on your actual

7

 

video to make them comfortable and then later on at court the jury won't switch off well not

8

 

switch off quite as quickly as °perhaps°

[Excerpt 2] [29]

So after all, the rapport, which the guidelines suggest is so important for achieving best evidence, seems to be the very thing that will undermine the value of this evidence in court, because the jury switches off and might end up missing important aspects of the evidence because their attention has wandered. So here the video is a rather awkward witness on the child's behalf, it performs the evidence in an unpredictable way. And again we can see the collision of time zones, between what "the book" suggests for the interview and what the judge reports from the courtroom (1-3). But now there is an even clearer sense of these being different time zones of veridicality, because they differ with regard to the way in which the truth is established within them. The very aspects that bolster and underline children's evidence in the time zone of the interview, the rapport and trust that enable them to deliver a full narrative, will produce an awkward impression in the time zone of the trial. Here the rapport appears tedious and lengthy, and it might ultimately distract from the relevant evidence. Lines 5-8 show the precariousness of attempting to handle this collision. The officers should try and "weigh up" (5), refine their immediate actions in the interview, relative to the assumed, to be anticipated immediate impression this will make on prospective jurors and their specific attention span, while still managing to "make" children "comfortable" (7). And here tellingly, the training officer's recommendation remains suspended in an unfinished sentence, fading tentatively on the word "perhaps" (8), thus underlining the relative dubiousness of such an endeavour. Clearly the video is not just awkward in its production, but it is a shifty witness as it performs in court. Even those things that are on the video seem to change value when transferred from one time zone to the other. And this is even more apparent in the next excerpt, where an officer insists that to gather a clear account they might have to challenge the child. Yet the training officer reminds them that it is the time of the trial they are simultaneously positioned within, and that this is the time they have to operate towards.

1

TO:

Is it so urgent to challenge a witnesses evidence at that point in time?=

2

PO1:

well it could be that's the thing couldn't it [[...]]

3

TO:

[…] if it is one interview that is played before the court we've already particularly if the jury

4

 

get to see that aspect of it we have already ahm (.) we're implanting a seed in their mind that

5

 

we actually don't believe it [...] which is ammunition for the defence.

[Excerpt 3] [30]

So the video indeed is not a neutral and unambiguous witness, it does not deliver what it is entrusted with objectively to the court. When transposing the immediacy of a challenge that could, during the interview, help to clarify things and might thus strengthen the evidence, into the courtroom, it ends up undermining the evidence. In this different time zone of veridicality, in the courtroom and thus in its new immediacy, the challenge acquires a different meaning, because here a different structure applies. Here the challenge implants "a seed" (4) in the minds of the jury that not even the police believes this witness and this also opens a gap that can be utilised strategically by the defence (5). The expression "implanting a seed in their mind" (4) is interesting because it gives a sense of the prospective jury's passive and imponderable perceptiveness towards the video, that nonetheless needs to be anticipated with extreme delicacy. [31]

4.2 Time zones, duration and sequentiality: what really happened …

Ultimately, above and beyond the concerns of the police officers, in the courtroom the collision of the two time zones of veridicality culminates around issues of duration and sequentiality. I have observed this phenomenon myself during trials, but I would like to present it by quoting a legal professional and researcher I interviewed, because she sums it up neatly.

1

A:

[…] I found where (1) ahm there was evidence in chief given alive in court (1) the time it

2

 

took you know was much shorter because the prosecuting lawyer will focus on what is

3

 

going to prove you know the charge (1) and they draw it out and when they felt they got

4

 

enough they stop (1) ((hmhm)) so that might take twenty minutes whereas your pre recorded

5

 

evidence might be an hour an a half […]

[Excerpt 4] [32]

Firstly the issue of duration is crucial. Referring to the trial she points out that the evidence emerges in a concise and short period of time when the prosecutor "draws it out", because they will only draw out exactly what they think is needed. Secondly, and closely related to this, there is the issue of sequentiality. That is, the order in which this evidence emerges in a video recorded investigative interview.

1

A:

[…] the investigative interview can you know it doesn't proceed a b c d e nicely in a

2

 

straightforward fashion it can go backwards forwards […] so the way the evidence emerges

3

 

can be quite disordered and […] it could make it difficult I think for a jury to hang on to

4

 

some of that information simply because it wasn't being produced in a nice sequential way

5

 

[…] I don't know I mean I did think sometimes simply because the way evidence emerged

6

 

in an investigative interview was so non-sequential sometimes that that could reflect

7

 

negatively in a court situation […]

[Excerpt 5] [33]

So above and beyond the concerns of the police officers, in court the video is a fanciful witness. By transmitting the very nature of the investigative interview, that is, the disorderliness and duration that qualify its immediate nature, into the courtroom, the video ends up undermining children's evidence. The very things that represent veridicality within the time zone of the actual interview (i.e. immediacy, spontaneity of information that might emerge in a disordered fashion), have the contrary effect within the time zone of veridicality constituted by the trial. Here evidence needs to be immediate but highly structured. It is expected to emerge in a highly controlled and sequential fashion via the ordering activity of the barristers. This is usually most obvious for the defence barrister, but it similarly applies for the prosecution examining their witness, because they will also extract an orderly sequence of events from the witness that can prove the case and impress the jury. And here the video, which is assumed to be the most stable and predictable of witnesses, turns unpredictable, because it short-circuits time zones of veridicality, it short-circuits the investigative time of the police with the fact finding time of the court. In other words, it disturbs the sequentiality of the overall investigative process. [34]

Here my analysis links up neatly with SCHEFFER's elaborations on the role of sequentiality for the defence case (SCHEFFER in this issue). Sequentiality is not just important for the way in which a specific account emerges in court, but it is also central for the overall organisation and transition evidence undergoes from the investigative to the trial process. SCHEFFER argues that it is the defence barristers who unfold the "field of presence" (FOUCAULT, 1972), within which they contrast, confront, juxtapose and thus integrate all previous statements, skilfully carving out their particular shape and intended efficacy within the trial. As I said, the same must be true for the prosecution, but where the video is being played, they cannot unfold their field of presence, they have to stand back while the video clutters the field of presence with information from another time zone. [35]

5. Epilogue: Purification, Tape-Fetishism and What Really Really Happened

I would like take a step back again and close this analysis by looking at the structural dynamics implicit to the problem I have described. As a tentative outlook I will hint at two wider theoretico-methodological issues closely related to the problem discussed here. [36]

Looking back at the police officer's activity, their declared task can be described as an attempt to purify children's evidence, to free it from ambiguity. This is a move similar to what STENGERS (1997) outlines as the crucial change in the natural sciences during the 18th century. A move that she also finds reflected in FREUD's shaping of psychoanalysis as a science.

"Just as the eighteenth-century chemist no longer deals with the materials that he will use in the natural world, no longer studies the unpurified primary materials that the artisan transformed, but 'creates his object', the analyst institutes a state that has all the aspects of an 'artificial illness' …" (STENGERS, 1997, p.97). [37]

In a similar vein the police officers attempt to create a purified and stabilised forensic exhibit. They create the "pure object" that can be examined under the controlled conditions of the courtroom. Drawing on my above analysis, one could say that in this case purification means the extraction of time. The video recording is an attempt to extract, to remove actual time from the interviews, in order to create an artificially timeless, stable capture of an immediate instance of evidence emerging. As we could see it is this very attempt of stabilising and freezing an objective capture of the evidence that betrays the effort, because facts do not travel well on video, the video resonates awkwardly with the time zone it is inevitably implanted into later on. During presentation time seeps back into the process because any viewing produces its own duration. This relates to the second resonance I would like to point out. There is an intriguing link to what ASHMORE, MCMILLAN and BROWN (2004) call "tape fetishism", that is, "the treatment of a tape as direct evidential record of a past event, and thus a quasi-magical time machine." (ASHMORE et al., 2004, p.349). Following my analysis we could see that children's video recorded evidence has been designed to be exactly such a time machine, a fetish. Yet, the video itself is after all a witness with a character and a reputation as ASHMORE et al. outline. There is a mysticism of objectivity surrounding recorded material. And the indestructible credo of "What you see is what you get", it is what "really happened", has always been the devil in the detail for discourse analysts, legal professionals, jury members and media spectators alike. Looking at recordings in qualitative research and in legal contexts, ASHMORE et al. examine the question of how we actually hear/see what is on tape and which rhetorical processes are concomitant with our decisions. Other than my analysis, their elaborations focus on the transcripts of tape recordings, hence they are virtually inspecting the "opposite end" of the phenomenon I have examined. Still, they come to a similar conclusion.

"… the tape is not and cannot function as a time machine. It cannot function in this way because there is never an unmediated hearing. […] As an artefact this tape may well appear as a fetish, but if it does so this is because it has already been subjected to a series of transformations, or translations. Its career as a fetish object begins not at the moment when it slips into the analyst's audio deck, but further back, when it is revised and catalogued, as a piece of research material […] and potentially even further back when the conditions of its recording are negotiated […]. Attending this chain of transformations allows us to see around the inevitability of tape fetishism" (ASHMORE et al., 2004, pp.370-371). [38]

I have explored some of the transformations and translations the video recorded evidence undergoes before and while it is being produced and put forward to the court as what appears to be, or is indeed hoped to be a fetish of objectivity. It is precisely because of its assumed role as a fetish that it backfires, because this status obscures the fact that it creates a short circuit between different time zones of veridicality. [39]

Video recorded evidence was set up in the well intended aim to deal with childhood ambiguity. It was to allow for children's voice to be heard in court in detail and in an unambiguous fashion, in order for it to be given due weight (as demanded by the UN Convention of Children's Rights), while addressing the problem of children's immature memory and their potential suggestibility. Yet, by short-circuiting the two time zones of veridicality, the video ends up creating more ambiguity than it resolves, and this will ultimately be at the children's cost. Firstly the police officer's attention is directed away from the individual child, as they try to correctly handle the paradox of performing simultaneously within/towards two time zones of veridicality while maintaining their own integrity. Secondly the video's showing in court exposes the collision of time zones. And as the video betrays the courtroom's traditional expectations for emerging evidence, it becomes even more difficult for children's voice to be heard in court. So regardless of the officers' efforts, the video is not a time machine. In fact, the implicit discourses of children as unreliable containers, irritable dispenser and volatile interactants are still at work here, perpetuated by the practices surrounding the video and other special measures. [40]

Still, ASHMORE et al.'s considerations about the authenticity and productivity of recordings also throw a peculiar light on the nature of my own data and what I have let it say. Is what I heard all that can be heard? How can I be sure that what I outlined is what really happened during this police training course, or in court? And what does my analysis mean for this practice? These are important questions because they demarcate the concrete intersection of epistemological and methodological issues. My analysis has looked at the complex dynamics of a specific aspect of a practice, as it is inserted into-, operates and is expressed within the related historical, cultural and concrete momentary discourses. Obviously I cannot (and do not wish to) say that videos are "always" counterproductive or that police officers are "always" uncertain, or even to blame. And undoubtedly there are cases where a video has helped to get a conviction. Still, it has become clear that the video as a special measure carries a serious structural problem that undermines this practice above and beyond the question of how it is actually applied and whether anyone in particular should be blamed for "bad practice". Hence, having entered and traced this practice in quite some detail, I think it has also become clear that in order to make sense of what is happening here, we have to choose the second of the two philosophies LATOUR (1997) is proposing to us, the one that defines constructing and realising as synonymous. Because only then is it possible to begin to grasp the intricate connection between the historico-concrete dynamics around discourses of childhood, sexual abuse, memory, suggestibility and time, and the concrete attempts of the law to operate around childhood ambiguity. [41]

In this sense I would hope that my analysis bears enough relevance for the ongoing realisations around child witnesses, that it could become part of this process and thereby grow to be more than just a peculiar perspective onto legal practice. It could spark a dynamic that might help to engender change within the thinking around these serious problems.22) Obviously there is no way to predict the readiness of legal practice to consider time zones of veridicality, but I would suggest that matters of suggestibility and time can play a significant role in getting to grips with "what really happened". [42]

Acknowledgements

I am grateful to Tine JENSEN, Alexander KOZIN and Nick LEE who have provided valuable comments on earlier drafts of this paper. I would also like to thank the editors of this special issue for creating this platform for an intriguing interdisciplinary encounter and for their insistence on interrelating accounts.

Notes

1) Data used in this article was collected in the context of the author's PhD research project titled "Cross-Examining Suggestibility: Memory, Childhood Expertise" (conducted at Loughborough University, 2003-2006). The study explores the interdependency of psychological, juridical and public discourses around children's suggestibility by examining child witness practice in England and Germany. The interdisciplinary approach combines a genealogy of the history and theory of suggestibility research with the analysis of empirical data gathered in interviews with legal practitioners and psychological researchers/experts in England and Germany. <back>

2) A group of children aged 5 to 16 claimed to have seen the defendants (20 residents of the Salem village) flying on broomsticks and to have been haunted by them with appearances and spells. In dramatic court hearings some of the children fainted, broke into spasms or appeared to throw up nails and pins they reported had been placed in their stomachs by the defendants (cf. CECI & BRUCK, 1995). <back>

3) SPENCER and FLIN (1993) warn that references to the Salem trials are problematic because very often they provide a scandalised and partial view of the events and central contextual factors are forgotten or misrepresented. They point out that while during the 17th century similar trials are recorded all across Europe, those cases involving child witnesses represented only a small fraction of the huge number of witchcraft trials taking place everywhere. SPENCER and FLIN furthermore criticise the common practice to use this as an example for the sheer bizarreness of the stories children can come up with, because considering the common beliefs of the time (17th century), children's stories were in no way bizarre, they merely reflected what most people believed. Finally most of those who quote Salem as a warning tale, seem to forget that children very frequently were subject to accusations themselves, were coerced into making accusations against their parents, or were incarcerated or punished after having served as witnesses because they were seen to be afflicted and thus complicit with witchcraft themselves. Considering these harsh consequences there is little reason to believe children had carelessly or even deliberately invented or fantasised such accusations (SPENCER & FLIN, 1993, pp.309-312). <back>

4) The concept of "childhood" itself is a product of more recent times (cf. DE MAUSE, 1974; ARIES, 1978). <back>

5) In his "Aetiology of Hysteria" published in 1896, FREUD declared that hysteria, a disease then recognised as frequent particularly in young women, was the result of childhood trauma caused by sexual victimisation. Only a year after its publication FREUD publicly abandoned this theory, stating that he had mistakenly assumed the sexual traumatisation was real, whereas now he realised that patients' memories of trauma were the result of elaborate fantasies (cf. MASSON, 1984). <back>

6) Legal systems differ considerably all over the world, and while the distrust in children's evidence was wide spread, some countries have never had explicit corroboration laws (e.g. Germany). <back>

7) Early research into children's memory and suggestibility mostly agreed that children's memory was unreliable and particularly susceptible to forgetting and distortions. Yet while early studies reported unanimously that younger children are most susceptible to suggestion and that this susceptibility decreases with age (cf. COFFIN, 1941), more recent studies, commencing from the middle 1980's started to undermine this consensus considerably (CECI & BRUCK, 1993). <back>

8) For the USA see CECI and BRUCK (1995); CECI and HEMBROKE (1998). For England see BELL (1988); LEE (1999); BULL (1998). For Germany see for example: STELLER (2000). <back>

9) Suggestibility research itself has a complex history and researchers always struggled to define suggestibility. One definition that is used frequently states that suggestibility is "... the degree to which children's encoding, storage, retrieval, and reporting of events can be influenced by a range of social and psychological factors" (CECI & BRUCK, 1993, p.404). <back>

10) Here my analysis relies heavily on the critical perspectives on developmental psychology, and the critique of the concept of development, offered e.g. by BURMAN (1994; 1997), WALKERDINE (1984; 1993), and MORSS (1990; 1996). Following their critique one can identify developmental psychology's tendency to find children's memory systematically deficient but improving with age, to be a result of the implicit assumption of a linear, progressive and directional concept of development. For a discussion of this critical developmental work see MOTZKAU (2005a); MOTZKAU (forthcoming). <back>

11) For a detailed synopsis and discussion of these changes see SPENCER and FLIN, (1993); WESTCOTT, DAVIES and BULL (2003). <back>

12) The so called Cleveland Crisis took place around 1987 and was one of those high profile cases where a small number of over-concerned professionals (in this case paediatricians) had, within a very short period of time, caused a total of 197 children to be removed from their families for suspicion of having been sexually abused. While great distress was caused to the families and children, some of which were only returned to their families after months or even years, the allegations could not be substantiated in any of the alleged cases (cf. BELL, 1988). For a circumspect analysis see LEE (1999). <back>

13) See http://www.opsi.gov.uk/acts/acts1999/19990023.htm. <back>

14) See also the "Sexual Offences Act 2003" (http://www.cps.gov.org/). <back>

15) This leaves England way off at the bottom of the international statistics (these usually refer to Northern America and Europe), with only Ireland showing lower conviction rates than England. <back>

16) Obviously, for the child this means that the court personnel (defence barrister, prosecutor and judge) is replaced by the television screen the child is looking at, where their faces will appear or vanish depending on who is talking to the child. Instead of the courtroom the child will be in a small room dedicated for CCTV transmission. The child is sitting right in front of the television set, with a video camera focussed on their face. Only one person can be in this room with the child, a neutral supporter/observer. This will usually be the court usher or a member of the witness support service that operates in most courts in England. <back>

17) The course is called "Video Interviewing Vulnerable and Intimidated Witnesses". Courses vary hugely across English police forces. The course I attended in summer 2004 to collect the data used in this paper, ran full time over six consecutive days with around 12 participants (experienced as well as career young police officers). Officers can volunteer, but are usually assigned to participate. It is the expressed aim of the police forces to train every single police officer in video interviewing children, in order avoid delays and enable an immediate response if sexual abuse is reported. <back>

18) "TO" is the training officer and PO1/PO2/PO3 are participating police officers. For this article I have slightly edited the excerpts by omitting a few lines. Omissions are indicated by square brackets "[…]". Still, the overall character of the exchanges is not altered by the omissions. Full recordings and transcripts are in the author's possession. The transcript notations used are a simplified and modified version of the system developed by JEFFERSON (1984). For further discussion on issues of transcription see also WETHERELL and POTTER (1992) and ASHMORE et al. (2004). Explanation of transcript notations: Pauses appear in rounded brackets indicating seconds: "(1)"; Speaker emphasis is indicated by "underlining"; Overlapping turns are indicated in "[square brackets]"; Minimal acknowledgement tokens by other speakers are indicated in "((double rounded brackets))"; Silently voiced utterances are indicated by "°degree signs°"; Words drawn out by the speaker are indicated by "col::ons"; A turn interrupted by a take up of another speaker is indicated by an equals sign "="; Rising intonation is indicated by a question mark "?". <back>

19) LATOUR uses this expression to describe the way in which scientific and technological work, if it operates successfully, manages to close black boxes. That is, to establish an area where a matter of fact is settled in such a way that one need only focus on inputs and outputs, while its internal complexity is rendered opaque (see LATOUR, 1987; 1999). STENGERS sums this up neatly: "A black box establishes a relation between what enters it and what leaves it in such a way that no one has, practically, the means to contest it" (STENGERS, 1997, p.86). <back>

20) Name changed by the author. <back>

21) The training officer refers to the guidelines "Achieving best Evidence". <back>

22) For further discussion of the problem of critical agendas for social change and the dilemma of self-positioning within critical practice research see MOTZKAU (in press); ZAVOS, MOTZKAU, CLARK and BIGLIA (2005). <back>

References

Ariès, Philip (1978). Geschichte der Kindheit. München: Deutscher Taschenbuch Verlag.

Ashmore, Malcolm; MacMillan, Katie & Brown, Steve (2004). It's a Scream: Professional hearing and tape fetishism. Journal of Pragmatics, 36, 349-374.

Bell, Stuart (1988). When Salem came to the Boro. The true story of the Cleveland Child Abuse Crisis. London: Pan Books.

Bull, Ray (1998). Obtaining information from child witnesses. In Amina Memon, Aldert Vrij & Ray Bull (Eds.), Psychology and law. Truthfulness, accuracy and credibility (pp.188-209). London: McGraw-Hill Publishing Company.

Burman, Erica (1994). Deconstructing developmental psychology. London: Routledge.

Burman, Erica (1997). Telling stories. Psychologists, children and the production of "false memories". Theory and Psychology, 7(3), 291-309.

Ceci, Stephen J. (1998). On the ethics of memory implantation research. Applied Cognitive Psychology, 12, 230-240.

Ceci, Stephen J. & Bruck, Maggie (1993). The suggestibility of child witnesses: A historical review and synthesis. Psychological Bulletin, 113, 403-439.

Ceci, Stephen J. & Bruck, Maggie (1995). Jeopardy in the courtroom. Washington, DC: American Psychological Association.

Ceci, Stephen J. & Hembroke, Helen (1998). Expert witnesses in child abuse cases. Washington: American Psychological Association.

Ceci, Stephen J.; Huffmann, Mary L.C.; Smith, Elliot & Loftus, Elisabeth F. (1994). Repeatedly thinking about a non-event: Source misattributions among preschoolers. Consciousness and Cognition, 3, 388-407.

Chertok, Leon & Stengers, Isabelle (1992). A critique of psychoanalytic reason. Stanford: Stanford University Press.

Coffin, Thomas E. (1941). Some conditions of suggestion and suggestibility: A study of certain attitudinal and situational factors influencing the process of suggestion. Psychological Monographs, 4, 43.

Dalgleish, Tim & Morant, Nicola (2001). Socio-historical perspective. In Graham Michael Davies & Tim Dalgleish (Eds.), Recovered memories. Seeking the middle ground (pp.3-22). West Sussex: Wiley and Sons.

De Mause, Loyd (1977). Hört ihr die Kinder weinen. Frankfurt/M.: Suhrkamp Verlag.

Edley, Nigel & Wetherell, Margie (1997). Jockeying for positions: The construction of masculine identities. Discourse and Society, 8, 203-217.

Edwards, Derek (1997). Discourse and cognition. London: Sage.

Edwards, Derek; Ashmore, Malcolm & Potter, Jonathan (1995). Death and furniture: The rhetoric, politics and theology of bottom line arguments against relativism. History of the Human Sciences, 8(2), 25-49.

Foucault, Michel (1972). The archaeology of knowledge and the discourse of language. London: Tavistock.

Garfinkel, Harold (1967). Studies in ethnomethodology. Englewood Cliffs, N.J.: Pentice Hall.

Goodman, Gail & Clarke-Steward, Ann (1991). Suggestibility in children's testimony: Implications for child sexual abuse investigations. In John Doris (Ed.), The suggestibility of children's recollections (pp.92-105). Washington DC: American Psychological Association.

Goodman, Gail; Quas, Jody & Redlich, Alison (1998). The ethics of conducting "false memory" research with children: A reply to Herrman and Yoder. Applied Cognitive Psychology, 12, 207-217.

General Assembly of the United Nations (1989). UN convention of rights of the child. Available at: http://www.un.org/ [Date of access: 15.12.2006].

Haaken, Janice (1998). Pillar of salt. Gender, memory and the perils of looking back. London: Rutgers University Press.

Herrman, Douglas & Yoder, Carol (1998). The potential effects of the implanted memory paradigm on child subjects. Applied Cognitive Psychology, 12, 198-206.

Home Office (1998). Speaking up for justice. Report of the interdepartmental working group on the treatment of vulnerable or intimidated witnesses in the criminal justice system. London: Her Majesty's Stationary Office.

Home Office (1999). Youth justice and criminal evidence act. London: Her Majesty's Stationary Office. Available at: http://www.opsi.gov.uk/acts/acts1999/19990023.htm [Date of access: 15.12.2006].

Home Office and Department of Health (1992). Memorandum of good practice on video recorded interviews with child witnesses for criminal proceedings. London: Her Majesty's Stationary Office.

Home Office and Department of Health (2001). Achieving best evidence in criminal proceedings: Guidance for vulnerable or intimidated witnesses, including children. London: Her Majesty's Stationary Office.

Jefferson, Gail (1984). On stepwise transition from talk about a trouble to inappropriately next-positioned matters. In J. Maxwell Atkinson & John Heritage (Eds.), Structures of social action: Studies in conversation analysis (pp.191-222). Cambridge: Cambridge University Press. Also available at: http://www.liso.ucsb.edu/Jefferson/Stepwise.pdf [Date of access: 18.1.2007].

Kelly, Liz; Lovett, Jo & Regan, Linda (2005). A gap or a chasm? Attrition in reported rape cases. Home Office Research Study 293. London: Home Office Research, Development and Statistics Directorate.

Latour, Bruno (1987). Science in action. Cambridge: Harvard University Press.

Latour, Bruno (1997). Foreword. Stenger's Shibboleth. In Isabelle Stengers (Ed.), Power and invention. Situating science (pp.vii-xix). Minneapolis: University of Minnesota Press.

Latour, Bruno (1999). Pandora's hope. Essays on the reality of science studies. London: Harvard University Press.

Lee, Nick M. (1999). Distributions of childhood's ambiguity in adult institutions. Childhood. A global journal of child research, 6(4), 455-474.

Lee, Nick M. (2001). Childhood and society. Growing up in an age of uncertainty. Buckingham: Open University Press.

Masson, Jeffrey M. (1984) Was hat man dir, Du armes Kind, getan? Oder: Was Freud nicht wahrhaben wollte. Freiburg: Kore Verlag.

McMartin, Clare (1999). Disclosure as discourse. Theorizing children's reports of sexual abuse. Theory and Psychology, 9(4), 503-532.

Morss, John R. (1990). The biologising of childhood. Developmental psychology and the Darwinian myth. London: Lawrence Erlbaum Associates.

Morss, John R. (1996). Growing critical. Alternatives to developmental psychology. London: Routledge.

Motzkau, Johanna F. (2005a). Die Beschuldigungssemiotik: Überlegungen zur Sprache der Dekonstruktion. In Peter Mattes; Tamara Musfeld (Eds.), Psychologische Konstruktionen: Diskurse, Narrationen, Performanz (pp.181-199). Göttingen: Vandenhoek & Ruprecht.

Motzkau, Johanna F. (2005b). Cross-examining suggestibility: Memory, childhood, expertise. Children's testimony between psychological research and juridical practice. In Alicia Czerederecka, Teresa Jaskiewicz-Obydzinska, Ron Roesch & Jozef Wojcikiewicz (Eds.), Forensic psychology and law: Facing the challenges of a changing world (pp.201-213). Krakow: Institute of forensic research publishers.

Motzkau, Johanna F. (in press). Reconsidering the binaries of practice research, or the dilemmas of self-positioning within agendas for social change. International Journal of Critical Psychology.

Parker, Ian (1992). Discourse dynamics. Critical analysis for social and individual psychology. London: Routledge.

Parker, Ian (2002). Critical discursive psychology. New York: Palgrave.

Parker, Ian; Levett, Ann; Kottler, Ann & Burman, Erica (1997). On discourse, power, culture and change. In Ann Levett, Ann Kottler, Erica Burman & Ian Parker (Eds.), Culture, difference and power: Discourse analysis in South Africa (pp.198-204). London: Zed Books.

Potter, Jonathan & Wetherell, Margie (1987). Discourse and social behaviour. Beyond attitude and behaviour. London: Sage.

Sacks, Harvey (1992). Lectures on conversation (Vol. 1 and 2, ed. by Gail Jefferson). Oxford: Blackwell.

Scheffer, Thomas (2007). The duplicity of testimonial interviews—Unfolding and utilising multiple temporalisation in compound procedures and projects [45 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research [On-line Journal], 8(1), Art. 15. Available at: http://www.qualitative-research.net/fqs-texte/1-07/07-1-15-e.htm.

Spencer, John (1997). The memorandum: An international perspective. In Helen L. Westcott & Jocelyn Jones (Eds.), Perspectives on the memorandum (pp.95-109). Aldershot: Arena.

Spencer, John R. & Flin, Rhona (1993). The evidence of children. The law and the psychology. London: Blackstone Press.

Steller, Max (2000). Forensische Aussagepsychologie als angewandte Entwicklungs- und Kognitionspsychologie. Praxis der Rechtspsychologie, Sonderheft 1, 9-28.

Stengers, Isabelle (1997). Power and invention. Situating science. Minneapolis: University of Minnesota Press.

Walkerdine, Valerie (1984). Developmental psychology and the child centered pedagogy: The insertion of Piaget into early education. In Juan Henriques, Wendie Hollway, Cathy C.Urwin, Couze Venn & Valerie Walkerdine (Eds.), Changing the subject: Psychology, social regulation and subjectivity (pp.153-202). London: Methuen.

Walkerdine, Valerie (1993). Beyond developmentalism? Theory and Psychology, 3(4), 451-469.

Westcott, Helen L.; Davies, Graham M. & Bull, Ray H.C. (2002). Children's testimony. Handbook of psychological research and forensic practice. New York: Wiley & Sons.

Wetherell, Margie & Potter, Jonathan (1992). Mapping the language of racism: Discourse and the legitimisation of exploitation. Chichester: Columbia University Press.

Yoder, Carol & Herrman, Douglas (1998). Revisiting the ethics of implanted memory research with children. Applied Cognitive Psychology, 12, 245-249.

Zavos, Alexandra; Motzkau, Johanna; Clarke, Jude & Biglia, Barbara (2005). "Feminisms and activisms". Reflections on the politics of writing and the editorial process. Annual Review of Critical Psychology, 4, 9-26.

Author

Johanna MOTZKAU (Dipl. Psych. Diplom at Freie Universität Berlin) is Lecturer in Psychology at the Faculty of Social Sciences, The Open University, UK. She has a background in philosophy, German Kritische Psychologie, theoretical psychology, developmental psychology and forensic psychology. She is interested in epistemological and methodological issues in psychology, post-structuralist philosophy, critical developmental psychology, issues of children's rights, child sexual abuse and the way in which psychological knowledge is used by the law. Previous work has drawn on the writings of G. DELEUZE to examine the "language of deconstruction". More recent interests include memory, suggestibility and credibility. Current work has looked at the history and theory of suggestibility research in relation to child witness practice in England and Germany (PhD thesis at Loughborough University "Cross-examining Suggestibility: Memory, Childhood, Expertise", submitted 2006).

Contact:

Johanna Motzkau

Faculty of Social Sciences
Briggs Building
The Open University
Walton Hall
Milton Keynes MK7 6AA
UK

E-mail: j.f.motzkau@open.ac.uk

Citation

Motzkau, Johanna F. (2007). Matters of Suggestibility, Memory and Time: Child Witnesses in Court and What Really Happened [42 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 8(1), Art. 14, http://nbn-resolving.de/urn:nbn:de:0114-fqs0701145.

Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (FQS)

ISSN 1438-5627

Creative Common License

Creative Commons Attribution 4.0 International License