Dr Karen Müller(1)

Law Faculty
Vista University
Port Elizabeth Campus

1  Introduction

Traditional accusatorial courtroom procedures act against eliciting complete evidence from children. This has been highlighted in a number of studies which were conducted to investigate the effect on children of testifying in court. Hill and Hill (1987:814-5) conducted an experiment on the effect which the courtroom has on children's ability to recall. A group of children were allowed to view the videotape of an incident. The following day the children were interviewed about the contents of the videotape. Half of the children were interviewed in a courtroom and the other half in a private room. The results clearly indicated that the children who were interviewed in the private room related more central items in free recall, answered specific questions more often and said "I don't know' or gave no answer significantly less often than the children questioned in the courtroom.

These findings support the argument that traditional courtroom procedures act against eliciting complete evidence from children. After the children in the above experiment were interviewed, they were questioned about their feelings regarding their experiences. The majority of the children, who testified in the courtroom, said they were nervous, embarrassed or scared, with only a few saying they felt good as opposed to the children in the private room, who expressed feeling nervous and feeling good more or less equally. Most of the children in the courtroom said they would never want to testify again, although a few said they might be willing if they were older. The children who were interviewed in the private room were more willing to take part in a similar procedure again. Many instances were observed in the courtroom where the children showed signs of nervousness, for instance, twisting hair, trying to leave the witness stand or courtroom before they were finished, shaking and, in once instance, the child even started crying (Hill and Hill 1987:816).

Saywitz and Nathanson (1993:615) also conducted a similar study to explore the effect of the courtroom environment on the quality of children's evidence and to determine the level of stress experienced by these children. Two groups of children participated in a staged event involving body-touch play between an unfamiliar adult male and small groups of children. The children were questioned about the event two weeks later in a simulated trial environment (mock courtroom). The other half were interviewed at their schools. The results of the study indicated that certain characteristics of the courtroom context interfere with the ability of a child to give evidence optimally, and increase the stress experienced by children. The children who were interviewed in the mock courtroom produced less complete descriptions of past events in free recall than children of the same age who were questioned at school. They made more errors in response to direct questions and acquiesced more often to misleading questions than those who were interviewed at school (Saywitz and Nathanson 1993:619).

The findings of this study highlight a number of points which are applicable to the efficacy of the adversarial procedure. The physical context of the courtroom has for many decades been presumed to promote the truth. The above studies indicate that this may not necessarily be the case where the witness is a child. The findings indicate that a variation in the environment in which questioning takes place can affect the child's evidence. It would appear that more complete and detailed reports are to be expected in statements gathered from interviews which have been conducted in familiar, private and informal settings as opposed to those which are obtained in court. The results of these studies highlight the need to develop innovative methods for modifying standard courtroom procedures when children give evidence (Saywitz and Nathanson 1993:620-1).

2  The effect of the accusatorial system on child witnesses

From the above, it would appear that the accusatorial system of procedure gives rise to two serious problems when the witness is a child. Firstly, this system of procedure results in trauma for children and, secondly, it affects the accuracy of a child's evidence. The South African Law Commission (1989:11) accepted that "[o]ne of the great and repeated complaints against the present system is directed against the adversary system and everything it implies". For this reason, Zieff (1991:37) argues that where children are caught up in "a gladiatorial contest between parties", as in an accusatorial method of trial, the children are bound to suffer. At the same time, however, a court appearance can have a positive, therapeutic effect on children where the court process is conducted in a manner sensitive to the needs of the child (Giles 1989:5; Berliner and Barbieri 1984:135; Levett 1991:17). It would, therefore, be incorrect to argue that children should be eliminated from court proceedings completely.

The two features of the accusatorial system which create the greatest difficulties for children are encompassed in the elements of confrontation and cross-examination. Section 35(3)(e) of the Constitution Act 108 of 1996 provides that the accused has the right to be present when tried. Although the wording of the section does not specifically contain the terms 'confront' or 'face-to-face', Joubert (1995:522) has argued that this right extends to confrontation in the sense of being able to see witnesses and to observe their demeanour. Even before the introduction of the Constitution, it has always been a basic principle of the South African criminal procedure that accusations had to be made face-to-face (Zieff 1991:35). According to the South African Law Commission (1984:4) an accused has the right "to demand that accusations against him be made face to face with him". This principle is also contained in s158 of the Criminal Procedure Act 1977.

The right of an accused to confrontation has the implication that any child who testifies against such accused must do so in the presence of the accused (SA Law Commission 1989:4). This creates immense difficulties for children who will, in most cases, be forced to testify in the very presence of those persons who assaulted them. In addition to the stress induced by the physical confrontation, children will also be required to relate their evidence in a formal courtroom which will be alien to anything they have experienced thus far (Hammond and Hammond 1987:9).

The traumatic effect of courtroom confrontation on children has been accepted by a number of South African courts. In S v Basil Simons DCLD 84/88, 13 June 1988 (unreported) Wilson J commented as follows:

"I propose for a moment to digress and to state that it appears to me that it is time that urgent consideration is given to a change in the manner of conducting criminal trials arising out of the sexual abuse of young children ... it appears to me that it would be eminently desirable to evolve a system that when a child is called upon to give evidence that child is not required to do so in a large austere looking court room before judicial officers sitting on a bench above them. In other words in circumstances that are completely strange to the child, and must cause a great deal of stress and tension."

The South African Law Commission (1989:3) accepted that courtrooms are spartan and severe in appearance, thereby creating a forbidding experience for a witness. They further accepted that the presence of the presiding officer, the accused and the latter's legal representative, as well as the prosecutor, in the courtroom, clad in black robes, caused a child to become "afraid, uncertain and confused" (SA Law Commission 1989:14). Zieff (1991:21) argues that, not only do children suffer serious psychological and emotional damage as a result of abuse, but this psychological harm is often perpetuated by their having to testify in court. Being forced to confront an attacker in person can result in the child refusing to testify or not being able to give evidence effectively (Avery 1983:17).

In response to the above difficulties experienced by children, the legislature introduced s170A of the Criminal Procedure Act 51 of 1977 which enables a child to give evidence via electronic means in a place other than the courtroom where the child would experience undue mental stress or suffering to do so. The use of these facilities is discretionary and dependent on whether the presiding officer is satisfied that the child will experience undue mental stress or suffering in the event of testifying. It is, therefore, not automatically available to every child who has to testify.

Schwikkard (1996:148) argues that the discretionary nature of s170A gives rise to problems since it views those children who testify via closed-circuit television as being the exception rather than the norm. Since research has shown that in the vast majority of cases child witnesses suffer significant trauma when testifying in an adult-centred adversarial environment, Schwikkard (1996:148) argues that this section would be more effective if it required the court to make use of closed-circuit television in all cases where a child complainant has to testify. Schwikkard limits this to complainants. It is my submission that there does not appear to be a justification for limiting s170A to complainants. A child, who has viewed the murder of a family member, although not a complainant in a case, will experience the same traumatic effect of giving evidence in an adversarial environment (Müller and Hollely 2000:27).

Since research has shown that children provide more accurate testimony outside of the courtroom and consequently experience less stress, it will be in the interests both of children in general and justice in particular that children be allowed to testify outside the adversarial nature of the trial. For these reasons, it is submitted that s170A be amended to enable all children to testify from outside the courtroom. This will protect children from the stress and trauma associated with giving evidence in court, and will accord with s28 of the Constitution Act 108 of 1996 in that it will assist in protecting children from abuse and their interests will be taken into account. At the same time, it will also be in the interests of discovering truth, and therefore justice, in that research has shown that children give more accurate evidence outside of the courtroom.

The other feature of the accusatorial system which creates immeasurable difficulties for children is cross-examination and all that it encompasses. Cross-examination is seen as the most crucial of the court experiences by the child witness, especially where the witness is a victim of abuse. The purpose of cross-examination is to attack the credibility of the child witness by highlighting inconsistencies in the evidence, and to discover any evidence that might assist in proving the accused's innocence in a context within which all the details of the case are examined microscopically. In order to achieve these aims, certain language devices are used with very few modifications being made when the witness is a child. This is set out clearly by Brennan and Brennan (1988:61) as follows:

"Questions of linguistic appropriateness, comprehensibility and concern for the psychology of the child witness are peripheral at best, and totally exploited at worst. The language format of cross-examination has a linguistic life of its own, independent of the age or status of the witness."

Both the South African courts and the legislature have accepted that children experience difficulties with cross-examination in the traditional courtroom, hence the introduction of s170A of the Criminal Procedure Act 1977 which not only introduced closed-circuit television, but also created the persona of the intermediary. In Klink v Regional Court Magistrate NO and Others 1996(3) BCLR 402 (SE) Melunsky J made the following observations at 411E.

"[I]t is sufficient to say that I am quite convinced that a child witness may often find it traumatic and stressful to give evidence in the adversarial atmosphere of the courtroom and that the forceful cross-examination of a young person by skilled counsel may be more likely to obfuscate then to reveal the truth."

The intermediary was, therefore, introduced to assist the child witness by removing all hostility and aggression from a question and by changing a question, where necessary, so that it would be more understandable to the child. However, in practice, the use of an intermediary has given rise to a number of problems. The power of the intermediary is very limited, since the intermediary is perceived to be nothing more than an interpreter, and the court can at any time insist that the intermediary repeat the question exactly as it was phrased. A further disadvantage of the present system is that the intermediary does not have the authority to comment on a question and give an opinion as to whether a child understands a question or not. The intermediary is powerless to intervene and argue that questions should not be asked in a particular sequence or not phrased in a certain manner. The real problems is that the intermediary is not an expert witness, but only an interpreter. This was accepted by the court in Klink v Regional Court Magistrate NO and Others supra at 441I: "The intermediary acts, in a sense, as an interpreter."

Although s170A has provided some relief for children who are entitled to use it, it does not address the traumatic effect of the adversarial nature of the trial as a whole (Schwikkard 1996:156). In essence, s170A amounts to a plaster that is being used to cover the cracks of a system that is not capable of dealing with child witnesses. Based on the available research on child development and language acquisition, it becomes clear that children are not able to give accurate evidence when cross-examined in an adversarial environment. All accusatorial countries have accepted this fact, hence the introduction of statutory exceptions to ameliorate the position of the child witness. It is submitted that statutory exceptions created on an ad hoc basis simply give rise to further confusion and, sometimes, place the child in an even worse position. It does not make sense to modify a system to suit children when the system itself does not support children. In such a situation, it is the system itself which needs to change.

3  A change in systems

Since children are unable to testify effectively in an adversarial framework, it remains to be investigated whether they will be better able to do so in an inquisitorial environment. If proceedings were conducted as an inquiry into the truth rather than a contest, Zieff (1991:37) argues that children would not find giving evidence in court as traumatic, and suggests that many problems would be eliminated if an experienced judge conducted an inquiry in an informal setting with the object of trying to discover the truth.

In order to investigate the efficacy of adopting an inquisitorial-based system of giving evidence in the case of child witnesses, the various elements of the present system will have to be evaluated, namely confrontation and cross-examination, and what effect any recommendations might have on the rights afforded to the accused by the Constitution. However, it is important to note at the outset that, although a move from an accusatorial to an inquisitorial form of procedure in certain cases would be a dramatic change, this by itself does not mean that such a change should be avoided. Both the accusatorial and the inquisitorial systems are the consequences of historical growth and political developments. They have not developed as a result of scientific inquiry into which of the two models is better equipped for accurate fact-finding. Rather, each system is based on popular conviction and speculation rather than on empirical research (Herrmann 1978:12). It would, in addition, be an over-simplification to talk about these systems as though they were mutually exclusive. All systems have characteristics drawn from both models (Spencer and Flin 1993:80).

The basis of recommending any change to a system is that it will result in the discovery of the truth. The South African Law Commission (1989:28) accepted that if such a drastic change was needed, then the mere fact that the change would be of such a drastic nature would not stand in the way of reform:

"If in the interests of the entire community - including those of the accused and the child - were to demand such a solution, then the mere fact that the said solution is of a drastic nature would not stand in the way of law reform."

This approach was endorsed by the Canadian court in Regina v Toten (1993) 16 CRR (2nd) 49 (Ontario C.A.) At 58:

"The public adversarial process is, however, a means to an end - the ascertainment of truth - and has virtue only to the extent that it serves that end. Where the established process hinders the search for truth, it should be modified unless the process or resource-based considerations preclude such modification."

The responsibility, thus, is to create an environment that will maximise the accuracy of a child's evidence and minimise the stress experienced by that child while at the same time protecting the rights of the accused (Saywitz and Nathanson 1993:621).

4  A tentative proposal

It is accepted that accusatorial cross-examination does not enable children to give evidence accurately and effectively. For this reason it is proposed that children be questioned within an inquisitorial framework. The interviewing of children, especially young children, is a highly specialised task. Interviewers require thorough training in developmental psychology, language acquisition and communication with children. This knowledge must be converted into practical application so that interviewers have the necessary skills to perform these tasks.

The necessity for integrated knowledge will be highlighted in the following illustrations. Children under the age of seven are said to be in the egocentric stage of development (Bukatko and Daehler 1992:318-325). This has the implication that children of this age are unable to understand the viewpoint of others and will not take the listener's needs into account. They will assume that the listener knows what they are thinking. Children of this age are able to distinguish what they think from what they have experienced, but will assume that the listener knows the difference (Mitnick 1993:4). A professional who has been trained to interview children will be aware of the above developmental limitations and will incorporate techniques into the style of questioning that will address these issues. Children in this stage are also very concrete and unable to think in an abstract, hypothetical manner. Children will interpret questions literally, so questions must be specific in order to obtain the necessary information. This means that children will have difficulty explaining their own thinking and will also not be able to explain how they came to a conclusion (Mitnick 1998:4). Specialised interviewers will be aware of these difficulties experienced by children and will formulate their questions in a literal, concrete fashion. An extract from the transcript of S v Cilliers 1991, SH2/334/91 (E) (unreported) highlights how developmentally inappropriate questioning (by the prosecutor) can compromise a child's credibility:

"Prosecutor : Tell me, X. How did it feel when the uncle jiggled your penis?
Child : It was nice but it was also not nice.
Prosecutor : What do you mean?
Child : I don't know.
Prosecutor : Was it a little bit nice?
Child : Yes.
Prosecutor : Is that why you didn't run away?
Child : I don't know."

In the above extract, the child, an eight year old boy, is asked to explain his own thinking (What do you mean?") He is also asked a hypothetical question ("Is that why you didn't run away?") which he is also not capable of answering.

Children also have a very limited ability to understand the concept of time and it is only by the age of twelve that children are beginning to develop a sense of time. They will have an idea of how long an hour or a day or a week is, but they are not very accurate at estimating time (Mitnick 1998:5). A further extract from S v Cilliers supra is reproduced here to illustrate once again how developmentally inappropriate questioning techniques compromise a child witness' credibility:

"Prosecutor : X, it is now 1991 and you are in standard 1?
Child : Yes.
Prosecutor : Last year you were in Sub B, it was 1990?
Child : Yes.
Prosecutor : And the year before that you were in Sub A, right?
Child : Yes.
Prosecutor : That was in 1989. Can you remember when this incident took place? Do you know what year it was?
Child : I think it was in 1985 or 1986.
Prosecutor : You say you think it's 1985 or 1986. So if you say you think, are you not sure?
Child : Yes, I'm not sure."

Here the eight year old witness does not yet have the ability to understand the questions which have been put to him, and, therefore, when asked for a reply ("Do you know what year it was?"), he simply hazards a guess ("I think it was in 1985 or 1986"). The incident actually took place in 1988, but the child would not be able to understand that concept due to developmental limitations. The prosecutor, in order to save his case and the witness, then attempts to clarify to the court that the child is not exactly sure about the dates ("So if you say you think, are you not sure?"). This confusion could have been circumvented by the use of developmentally appropriate interviewing techniques.

From the above, it is very clear that the questioning of young children (under the age of twelve) is a very specialised field, and must be conducted by professionals who are experienced and skilled in the area. The proposal made earlier by Zieff that judges be allowed to interview children does not seem feasible in this context. Legal practitioners have no training in child development and no experience of communicating with children in such a specialised context. It also does not make sense to train magistrates, judges and prosecutors to this level of competence since child witnesses only form a small proportion of their duties. In addition, it would not be possible to insist that defence counsel be trained, and it is the defence counsel who are responsible for conducting cross-examination.

Article 6 of the European Convention on Human Rights undertakes to guarantee a party a fair trial in civil and criminal cases, which is a basic element of the rule of law. Subsection 3(d) provides that everyone charged with a criminal offence has a right "to examine or have examined witnesses against him and to obtain the attendance and examination of witnesses on his behalf under the same conditions as witnesses against him". This must be seen in the context of both the accusatorial and inquisitorial models. In the accusatorial system witnesses are examined and cross-examined by the parties or their representatives while in the inquisitorial system witnesses are examined by the court (Jacobs and White 1996:158). The emphasis here is on the right to challenge the evidence against the accused rather than give the accused himself the right to cross-examine the witness. This is supported by the decision in Delta v France, Judgement of 19 December 1990, Series A, No. 191-A (1993) 16 EHRR 574 where the accused was convicted of robbery on the basis of identifications which were made at the police station by the sixteen year old victim and her friend, neither of whom gave evidence at the trial. This was held to be a violation of the Convention, although the European Court of Human Rights added that the Convention does not insist on a witness facing the accused in court, all that was necessary was that the defendant be given "an adequate and proper opportunity to challenge and question [the] witness against him, either at the time the witness made his statement or at some stage of the proceedings".

In an inquisitorial environment the focus is on the evidence being examined. This does not mean that the accused or his representative is entitled to examine the witness in person. In most inquisitorial countries this is done by an examining magistrate or, in the case of children, by specially trained police officers. This is considered to be adequate protection of the accused's right "to examine or have examined witnesses against him".

In view of the above, it is proposed that a forensic interview of a child witness, conducted by a skilled interviewer in accordance with legal requirements, replace the child's evidence-in-chief and cross-examination. At the trial, this interviewer will sit with the child in another room which has been linked to the court via closed-circuit television in exactly the same manner as the intermediary would, as envisaged by s170 at supra. The interviewer will then conduct a forensic interview under the watchful eye of the court. The purpose of this interview will be to place the child's evidence before the court in as accurate a format as possible, using knowledge of cognitive development and language acquisition to achieve this aim. Any discrepancies or confusions that may arise from the evidence will be examined. This interview will be viewed by the accused, his representative and the court, thus introducing an element of the accusatorial system. This will enable the court to watch the interview and object if the interviewer should perhaps ask an unacceptably leading question or act in an inappropriate manner. In this way the accusatorial 'day in court' is still preserved and the child is not deprived of making a contribution at the trial while at the same time the accused's rights are also protected in that he can watch the demeanour of the child while the latter is testifying, and can hear the way in which the child is being interviewed. The interviewer would, in much the same way as s170A, be linked to the courtroom by means of earphones in the event that the court may wish to communicate with the child or the interviewer.

When the interviewer has placed the child's evidence before the court, the defence and the court will be afforded an opportunity to raise any matters that have not been covered in the evidence or which may have appeared inconsistent. This must not be viewed as an opportunity to cross-examine the child, and the court will have to guard against this danger. It will only be an opportunity to raise any matters that have not been covered or which have created confusion. The interviewer will then examine the child further on these aspects. By the term 'examine' is meant the placing of evidence before the court and questioning the child in detail about any aspects of the case which appear inconsistent. If, at the end of the interview, the child is unable to provide an account of events or provides an inconsistent account, then the court will take this into account when weighing up the evidence. By implication, the defence will not be able to put their version to the child, but it has already been shown that children do not understand the purpose of this technique and young children especially are not able to place themselves in a position where they can view matters from another perspective.

This proposal of interviewing the child combines aspects of both the accusatorial and the inquisitorial model. The examining of the evidence is done by one person, in much the same way as is done by the examining magistrate in inquisitorial systems. However, aspects of the accusatorial system have been assimilated, namely affording the defence an opportunity to view the child and hear while the child is giving evidence, and to put further matters to the interviewer. In addition, this proposal requires a professionally trained person to interview the child, since this was one of the problems experienced in inquisitorial systems supra, namely that examining magistrates were not trained in the intricacies of child interviewing.

This proposal has certain implications. It is essential that the interviewer be professionally trained to conduct an investigate interview with children. This proposal requires the creation of a specialist who will interface between law and psychology and form a link between the two disciplines. This training will include knowledge of developmental psychology so that the interviewer will be able to assess the developmental level of the child being interviewed, and interview accordingly. In addition, the training must also focus on the psychology of abuse so that the interviewer has a knowledge of how victims of abuse react i.e. non-disclosure, recantation etc. Thorough training will be required with respect to the acquisition of language by children so that age-appropriate questioning techniques can be employed. An essential component of this training would include aspects of procedure in court. The interviewer would require a basic knowledge of criminal law and procedure so that there is an understanding of what information needs to be elicited from the child i.e. that in the crime of rape, for instance, it is necessary to prove that penetration took place. The interviewer will need to know this so that when conducting an interview, important legal details are not omitted. Rules of procedure are also important since these will have application to the methods used in conducting a forensic interview. Practical training will be essential and the interviewer will have to have experience in interviewing children. Harvey and Daun (1993:9) suggest that in order to conduct an interview with a child, interviewers should have a knowledge of child development, children and communication, child memory and suggestibility, children and sexuality and the psychology of victimisation. They would need a thorough knowledge of investigative interviewing procedures as well as the law relating to child witnesses and certain aspects of substantive law, such as the elements of offences, for instance.

The creation of such a trained professional will eliminate the stress created by cross-examination for children. It will also reduce the confusion that arises from developmentally inappropriate interviewing practices and improve the accuracy of information provided. These advantages were described by Benedek and Schetky (1986:1228) as follows:

"In addition, a pool of professionals trained in the relevant psychological and legal technical skills might avoid the confusion created by the improper use of leading questions, language that the child cannot understand, and techniques designed to harass and intimidate a young child."

To accommodate the above proposal s170A will have to be amended to provide that the examination and cross-examination of the child witness must be conducted by the above-described interviewer, that the court may intervene where necessary, and that the defence may have an opportunity to raise any issues with the court that may appear to be unclear from the evidence. Although the term 'interviewer' has been used thus for in the proposal. It is submitted that the term 'child expert' be preferred, since the status of this person should be elevated from that of an interpreter to more of an expert who can assist the court by eliciting an accurate account from a child witness using scientifically approved methods. To ensure this, the amended s170A should also set out the training and or qualifications that a person would have to possess in order to qualify as a child expert. This qualification can be standardised by the Minister of Justice in liaison with universities in much the same way as has happened with regard to interpreters. The Institute of Interpreters offers a qualification through various participating universities to interpreters.

It is envisioned that the child expert be attached to the Ministry of Justice on a full-time permanent capacity. By creating such an office, the Ministry will enable the child expert to become experienced at the task to be performed. The creation of an office of the child expert would promote specialisation. Oates (1990:133) argues that this specialisation has become necessary:

"It may be that the time has come for the development of a speciality in child law where those undertaking this speciality have some training in developmental psychology, child development, and experience in interviewing children."

5  Conclusion

Research relating to child witnesses has focused a great deal of emphasis on the rights of the accused in the criminal justice process. With the introduction of s28 of the Constitution Act 108 of 1996 it is necessary also to take cognisance of a child's rights as set out in the section. It is, therefore, necessary to balance these rights when evaluating cross-examination. The accused's rights to employ techniques of cross-examination must be weighed up against the linguistic rights of the child witness. A mismatch between the language of the court and the language understandable to the child does, according to Brennan and Brennan (1988:61), very little to preserve the rights of children or to arrive at the ultimate truth. It is essential therefore, to modify the system to assist the child in providing the court with an accurate account as well as to ensure that the accused is accorded a fair trial. Modifications need to be based on scientific research, and it is for this reason that it is proposed that a professional interviewer conduct the examination of the child at the trial while the defence is afforded an opportunity to view the interview and object, where necessary.

In evaluating any possible changes to a legal system, the following statement by Saywitz and Nathanson (1993:621) should be borne in mind:

"As a society, we have a responsibility to create an environment that maximises the completeness and accuracy of children's testimony and minimizes the stress placed on children in the process."


Avery, D. 1983. The child abuse witness: potential for secondary victimization. Criminal Justice Journal. 7:47.

Berliner, L. and Barbieri, M.K. 1984. The testimony of the child victim of sexual assault. Journal of Social Issues. 40(2):125.

Brennan, M. and Brennan, R. 1988. Strange Language - Child Victims under Cross-examination. Wagga Wagga.

Bukatko, D and Daehler, M.W. 1992. Child Development: A Topical Approach. Houghton Mifflin Co: Boston.

Giles, C. 1989. The child as witness. Die Kinderversorger. 7(5):5.

Hammond, J.C. and Hammond, E.J. 1987. Justice and the child witness. SACC/SASK. 11:3-20.

Harvey, W. and Daun, P.E. 1993. Sexual Offences Against Children and the Criminal Process. Butterworths: Canada.

Herrmann, J. 1978. Various models of criminal proceedings. SACC/SASK 2:3.

Hill, P.E. and Hill, S.M. 1987. Videotaping children's testimony: An empirical view. Michigan Law Review. 85:809-833.

Jacobs, F.G. and White, R.C.A. 1996. The European Convention on Human Rights. 2nd edition. Clarendon Press: Oxford.

Joubert, J.J. 1995. Die Legaliteitsbeginsel in Strafprosesreg.

Levett, A. 1991. Contradictions and confusions in child sexual abuse. South African Journal of Criminal Justice. 4(1):9.

Mitnick, M. 1998. The use of language in interviewing children in American Prosecutors Research Institute: Finding Words. Conference Proceedings. Savannah, USA.

Müller, K.D. and Hollely, K.A. 2000. Introducing the Child Witness. Printrite: Port Elizabeth

Oates, R.K. 1990. Children as witnesses. The Australian Law Journal. March 64(3):129.

Saywitz, K.J. and Nathanson, R. 1993. Children's testimony and their perceptions of stress in and out of the courtroom. Child Abuse and Neglect. 17:613.

Schwikkard, P.J. 1996. The abused child: a few rules of evidence considered. Acta Juridica. 148.

South African Law Commission. 1989. Working Paper 28, Project 71. Protection of the Child Witness.

Spencer, J.R. and Flin, R.H. 1993. The evidence of children: The law and the psychology. Blackstone Press: London.

Zieff, P. 1991. The child victim as witness in sexual abuse cases - A comparative analysis of the law of evidence and procedure. SA Journal of Criminal Justice. 4(1):21-43.

1. BA (Hon) LLB PhD