KEYWORDS: direct examination; directed questions; evidence; power; restorative justice; therapeutic jurisprudence; trial process; witness testimony.

MODEL INTRODUCTION
Steven Lukes (1974) examined the notion of power. He identified one, two, and three-dimensional views. The one-dimensional view focuses on behaviour; it involves decisions being effected on identified issues. The Dean exercises power by getting the decision he or she wanted from the Faculty meeting. The two-dimensional view involves the organisation of issues and non-issues. The Dean exercises power by deciding what is, and is not, placed on the agenda for discussion at the Faculty meeting. The three-dimensional view focuses on how issues are perceived; on how we come to think about, write about, and judge topics. The Dean exercises power by arranging for us to understand the issues in a particular light.

No wonder so many governments and political parties put so much effort into 'spinning' issues.

DRAMATIC INTRODUCTION

When did that happen? Where did it take place? Unfortunately it could have been any one of many countries. It was 2001. The judge said that the only way to test the veracity of the eight year-old boy's evidence was through cross-examination. It was alleged that the six men had sexually abused him and an 11-year-old girl.

We will never know, by legal standards, whether those six men sexually abused that girl and boy. However, we know by the standards of simple common sense that they were abused by a legal system. The surprise is not so much that the trial was stopped but that the girl coped, being cross-examined by six lawyers, for 10 days. The surprise is that the case got to court. The surprise is that more parents and social workers have not refused to put vulnerable children into the legal system. And it should be no surprise if more abusers have learnt, from the case, how legally wise it is to target the especially vulnerable.

QUESTIONING INTRODUCTION
How should we respond to such cases? Must it be context-dependent? If it is a law journal must it be temperate and rational? Or is that giving others power in Lukes' three-dimensional sense? Of course it is easy to be emotional about such cases. It is almost as easy to respond with arguments about how important it is not to wrongly convict the innocent, about how easy it is to make accusations and difficult to disprove them. How would, and how should, Therapeutic Jurisprudence (TJ) and Restorative Justice (RJ) respond to such cases? Do such cases, and such sociological theories, highlight the limits of the TJ and RJ approaches?

Is there any problem? The law is being changed, at different paces in different places, but we know there has to be change (Spencer and Flin 1993). For example the Youth Justice and Criminal Evidence Act 1999, for England and Wales, ensures that children and other vulnerable witnesses can receive 'special measures,' which include screening from the accused, video links, and even removal of legal wigs and gowns. Other child witnesses have been allowed to answer the questions from another room, over a video link. In Scotland they are considering extending the protections to vulnerable defendants (Scottish Executive 2001). The quality of initial interviews has been improved, drawing upon developments in cognitive interviewing (Memon and Bull 1999; Milne and Bull 1999). When interviews have been video taped, they have been presented as evidence.

A MATTER OF PERSPECTIVE
But we still get cases like this. In a perverse sense, those children were lucky! Their allegations were acted upon. In many other cases the prosecutors would conclude that it was insufficiently likely that a conviction would be achieved. Alternatively the prosecutors, or their advisors, would consider it improper to put the particular children, whom they believed to be victims, through the additional terror of a court process. So those cases would never get to court. The police and other professionals could assure the children that they believed them. They could try to explain the legal principles. But the failure to prosecute would always suggest otherwise. The children, believed to be victims, will grow up knowing that the law and legal system did not believe them, or not enough. It does not just affect children. The 'justice gap,' (the difference between the number of crimes reported and the number prosecuted) at least in England and Wales, is growing. In 80 percent of crimes reported to the police (which of itself involves an under-representation of crime), the perpetrator goes unpunished (Home Office 2002). One in 13 rape victims sees her or his assailant in court (HMIC 2002).

How does, or should, TJ and RJ reply? Therapeutic jurisprudence (TJ) is the relatively loose title for an approach to studying, analysing, and practising law. It identifies, particularly by calling upon the behavioural sciences for insights and for supporting research, laws that have anti-therapeutic effects. It suggests ways in which the law as it is formulated or practised could have positively therapeutic effects. It is, essentially, about law reform, where 'law' is interpreted widely. Thus we could say that all TJ concerns law reform, but far from all law reform involves TJ. Therapeutic jurisprudence is concerned with outcomes, with effects. It seeks laws that have productive therapeutic effects, although it does not profess that such effects are necessarily better than others. It is also involved with processes, if those have anti-therapeutic effects. The TJ motivation appears to have more to do with effectiveness than with efficiency, although the latter might directly or indirectly permit more therapeutic outcomes (Wexler 2000).

Therapeutic jurisprudence was founded by, and continues to be fuelled and developed by, legal academics (Wexler 1990; Wexler and Winick 1991). Despite attempts to internationalise the approach, a disproportionate amount of the written output still comes from the United States of America (USA) (e.g. see Wexler and Winick 1996) and increasingly from Canada. North America has stronger associations with realist approaches to law than Europe. For Realists it is what happens in practice, rather than in theory, that dictates which laws are interpreted. It is how questions are asked in court that matters, not what the books of ethics might declare. TJ also gains in authority and persuasiveness through its links with the behavioural sciences. It involves a call to law academics to support their analyses and proposals with reference to quality research. It eschews the traditional approach of arguments for law reform (which it is acknowledged is the basis of this paper) that are merely persuasive reasoning from 'common sense' or implicit assumptions and values. It is firmly within the 'socio-legal' approach to understanding and teaching law. But it is more concerned with the behavioural rather than the sociological or social sciences. Is this a potential weakness? It draws upon psychology and, to a lesser extent, psychiatry with which it originated (Wexler 2000). It is focussed more upon the individual than the group or community. It is concerned with individuals' experiences of the law. It is the 'therapy' of the individual patient rather than the relationship, family, group, firm, etc. Individualism, rugged and independent, remains such a powerful value that it is difficult to raise it as an issue. Analyses or explanations in terms of the distribution of power or other resources, or in terms of ideologies and the control of meanings, are unlikely to feature in TJ papers. (For a list of most of the papers written on TJ topics see the web site kept by the International Network on Therapeutic Jurisprudence at: http://www.law.arizona.edu/depts/upr-intj/intj-welcome.html.)

Therapeutic jurisprudence is also 'grounded' through its connections with pro-active judges and developments in restorative justice (Rottman and Casey 1999; Simon 2003). Many judges, again particularly in the USA, have grown tired of merely processing people according to law. They want to make an effective difference through law. For example, so many problems of people and of courts arise from a failure to tackle drug dependency. So some judges have devised procedures designed to make a real difference, organising interventions, co-ordinating services and responses, and monitoring effectiveness. They are committed, in practice as well as principle, to achieving therapeutic outcomes. Others will look askance at this judicial activism and argue that the judges are going beyond their proper remit; that justice is a process not a product.

Restorative justice (RJ) is another loose label covering a number of developments. But it is essentially concerned with restoring the balances that crime and other forms of anti-social behaviour upset (e.g. Strang and Braithwaite 2001). It, too, is concerned with effectiveness but achieved by concentrating upon process. Like so many others, it wants to reduce crime. But it responds with measures designed, tested, and audited to be effective, rather than with mere clarion calls for more punitive policing and/or sentencing. Effectiveness is almost raised to the level of a principle (Graef 2000).

A virtue of TJ (and indeed of RJ) is that it is, at least apparently, apolitical. That makes it easier for judges to espouse. This is related to the greater focus in the behavioural sciences upon individuals such as litigants, rather than upon groups and communities, which are the interests of the more sociological sciences. TJ does not take a position on the issues that divide us into different political parties. Insofar as its raison d'être is reforming the law and helping individuals, it may be associated with liberal values. But this does not translate automatically into support for any particular partisan position. So supporting TJ does not presuppose any particular political position. Judges, without appearing inappropriately political, can adopt it. It concerns efficiency. If a law has counter-therapeutic effects, then TJ can show a way in which it can be changed to be more effective. Perhaps it is no coincidence that TJ has developed almost contemporaneously with a number of politicians espousing 'the third way,' an approach to politics which avoids the allegedly defunct and dichotomous opposition of capital and labour, manager and worker, even rights and duties.

Thus, therapeutic jurisprudence would appear to be an ideal model, or lens, for examining the problems experienced by vulnerable witnesses in courts. Actually, it might highlight TJ's problems and limits. Or such a suggestion might provoke a debate.

IS REFORM ENOUGH?
The problem experienced by and harming those two young children, and many others in similar circumstances, is our trial system. Lawyers are entitled to question vulnerable witnesses. Indeed it is not just their right but also their job. We can protect the witnesses, to an extent, with screens and video links, etc. (Spence and Flin 1993). We can ameliorate their experiences. But defence lawyers remain entitled to challenge the witnesses' versions of events. We have competing interests, at least those of the vulnerable witnesses and the defendants. The community also has interests in people not being wrongly charged or wrongly convicted, thereby having to go through all the stresses and other experiences of a trial. We all might, albeit innocent, be charged with such an offence.

Therapeutic jurisprudence is initially concerned with whether the law is unnecessarily having adverse effects. Clearly that is happening with vulnerable witnesses. But how does TJ sort out such conflicts of interest? It does not - perhaps it should - value a child's interests over an adult's, a victim's over a suspect's, or vice versa. Both Kress (1999) and LaFond (1999) have commented on the problems that arise when TJ is faced with disputes or choices over values. But it is more difficult than that here. The defendant has legal rights. These are not just formal rights; they include the tactical. The defendants are entitled, in the sense that the law permits and regularly expects it of the defendants and their lawyers, to use every lawful tactic open to them to defeat the charges. And the most effective tactics may be emotional rather than rational, may be to spoil a witnesses rather than their evidence. Lawyers are allowed to attack the singer as well as the song. It should not have to be necessary to explain that Spencer and Flin (1993) were being ironic when they wrote:

Therapeutic jurisprudence insists that legal rights trump, are regarded as more valuable than, therapeutic rights. The defendant has legal (constitutional in the USA) rights; the witness does not. Doctors and surgeons can do things to us which, but for their white coats and the clinical setting and purpose, would be crimes. Lawyers can do to witnesses things which can cause such emotional turmoil and upset that, were they not done (often with the aid of special clothes) in special settings and in the 'rational' atmosphere of a courtroom, would constitute the crime of intentionally causing grievous bodily harm, given that it can now be caused psychically. But the analogy is unfair; the patient is entitled to give or to withhold consent, the witness is subpoenaed.

Unless we are to simply allow the defendant's legal rights to trump everything, TJ needs to take an explicit value position or to try another tack. We have made great strides, at least in several countries, in changing how police officers and social workers interview children. This substantially draws upon the cognitive interview (Lamb et al. 1999). The virtue of this method is that it helps the interviewee to remember significantly more about the incident in question. True, it is gentle towards and respectful of the interviewee. But that is because that approach works better. Establishing rapport with the interviewee, adult or child, vulnerable or otherwise, helps the interviewer to obtain more correct information (Milne and Bull 1999). We are so satisfied with the cognitive interview that it has been effectively written into the mandated method for interviewing child witnesses in England and Wales (Bull 1998). It produces more and better goods. And yet lawyers and judges, whose currency is information and whose goal is accuracy, need know nothing about the cognitive interview, let alone the enhanced cognitive interview, or be skilled in its use. We insist that police officers and social workers follow good interviewing practice, but we exempt judges and lawyers.

Ah, but it will be replied that lawyers must challenge witnesses. They are entitled to, and sometimes must, do more than just imply that the witness is mistaken. They must suggest, if not demonstrate, that the witness is or may be lying. They must challenge, tackle, and confront the witnesses, even if they are vulnerable children, because of the effect that their evidence may have. But have the goals changed? Is not the objective to obtain high quality, accurate information? If 'yes,' then surely the rational response would be to use those methods of interviewing, or examining, a witness which were most likely to achieve that result. The evidence does not support aggressive cross-examination as being a superior method of fact finding. Any witness, adult or child, may change his or her evidence under cross-examination (see generally Gudjonsson 1991). Given the bizarre and artificial setting, at least from the witness' perspective, is that at all strange? That a witness changes his or her evidence may be cited as demonstrating the effectiveness of cross-examination. Yes, it is effective in getting witnesses to change their evidence; but that event cannot prove the inaccuracy of the first account or the accuracy of the second.

If we are to accept the claims made for the value of cross-examination as a tool in uncovering the truth, then we need much more than examples of changed evidence. And we cannot rely upon verdicts. A court may decide in favour of one party's evidence and against the other's. For practical purposes we accept that decision. (Indeed we call it a 'judgement' to suggest that it is something more than just a risk decision, although we recognise its fragility in ensuring provisions for appeal.) We have to have closure. One party has to win, or in criminal cases not to lose. State power will follow the court's decision; judgements will be enforced. And we thereby create and build upon a new reality. The court decided that way. So be it. We must accommodate to that. But it still does not follow that the decision was correct! It still does not follow that the judge or jury chose to favour the party with the true, or better, evidence. And we do not have any courage of any convictions that we might have in the system, for a finding of not guilty is not a finding of innocence.

The trial is not a scientific procedure (McEwan 1998). We cannot, correctly, make inferences about the quality of the process (the trial) from its product (the verdict) when we cannot know the accuracy of that product. We ought to be less referential and more willing to be critical of trial processes. Maintaining the status quo, for example the rights of alleged defendants in comparison with alleged witnesses and victims, is not being value neutral. We should not so readily accept, as some form of a moral baseline, that the adults' rights to defend themselves against the accusations of the two young children includes the right to use devices which are simply tactical or persuasive. Rather, we should insist upon procedures and processes that are as rational and consistent with current quality research as we can currently manage. It should be a mark of a 'modern society' that it is prepared to make use of that which makes it modern.

The claims made for the value of cross-examination cannot be demonstrated just by reference to trial outcomes. Perhaps we should stage events with lots of video cameras observing what 'really' happened. Then we could subject witnesses to examination and cross-examination, to discover if our trial processes reach the correct conclusions. But such research would be very difficult to conduct and would rarely be compelling, as so many variables would be involved. In the interim, at least, we should be examining the trial as a process and a system. We should be drawing upon models, insights, and evidence from studies of other processes and systems. For example we might develop ideas by analogy with safety engineering. Fault, Mode, Effects and Criticality Analysis (FMECA) involves identifying all the parts of a process (Abbott 1987). A useful example would be a factory process whereby a range of raw ingredients (evidence) comes in at one end. These are then processed in a variety of ways and at different stages. And, at the end, out comes the product. Hopefully the product will not be subject to a recall, or to an appeal. The safety engineer, using FMECA, identifies every stage at which something could go wrong. Then he or she assesses how likely it is to go wrong at each stage and how critical that would be. From this analysis the safety engineer can identify the parts of the process where there is the greatest danger. Then decisions can be made about where investment is most needed to reduce the likelihood of serious errors in the process. Assessments of low likelihood of serious damage or error can be compared with high frequency errors with minor consequences.

Do we have the quantity and quality of empirical evidence to identify those parts of our trial systems which are most likely to produce error? We might examine the frequency at which different types of evidence, such as identification, are offered. We know there are many problems with that kind of evidence (e.g. Cutler and Penrod 1995). It is possible to show that identification was erroneous, in a particular case when additional evidence becomes available, or that it is likely to have been wrong given the poor quality of the procedures involved. But it is not so easy to show that a witness did not mean to say what he or she did say in a witness box. An error may be detected by comparing an answer with other evidence. But it does not follow that the witness did not believe it to be correct.

Despite the difficulties with a FMECA analysis of the courts, we could, at the very least, seek consensus statements on such an analysis. We could, knowing what we currently know, identify areas of concern about the reliability of our trial systems. If we did, it is submitted that considerable concern would and/or should be expressed about examination and cross-examination as evidence-gathering and testing procedures. Inadequacies with identification evidence, for example, affect that part of those cases. But oral examination and cross-examination affect all parts of a case in court. There are so many opportunities where the damaging effects of examination could be critical to the outcome of the case. Is it not amazing and ironic that we debate, at length, the problem of poor scientific evidence in court and yet let the process of obtaining oral evidence, whether on scientific issues or not, pass unquestioned. Trials are so traditional and the stuff of such exciting media representations. Even in cases where evidence is not offered, or defendants opt not to give evidence themselves, the fear of cross-examination will regularly be a factor in that decision. It is not just what oral examination can do in court, but its reputation precedes it. We let tradition have power over us. We assume that the debate must be on the terms of adversarial trials (Lukes 1974).

THE 'NEED' TO BE MODERATE
Even if we cannot find a 'perfect' trial system we should, at least, examine whether lawyers could ask their questions in a manner that is 'fair' both to the witnesses and defendants. We might be able to identify a range of questions which lawyers could be prohibited from asking, without offending the existing legal rights and interests of defendants. Yes, rules of professional ethics and evidence already exist. But, clearly, they do not go far enough. Provided that these rules do not encourage false evidence, whether for or against the defendant, then they should be beyond objection. That is assuming, of course, that we are genuinely seeking the correct decision. But such rules are likely to be opposed: they involve limiting lawyers' rights to use all manner of tactics.

Witnesses should understand the questions they are asked. Or does the defendant have interests that are contrary to this? Walker (1994) has demonstrated, drawing upon developmental psychology in particular, how young children have difficulty with many adult uses of language. It may be impossible, or impractical in practice, to guard against all of those possible misunderstandings from occurring. But a few rules could be devised to help.

1. No double negatives. These are, by definition, unnecessary. They can only confuse.

2. No pronouns other than you. It is too easy for the lawyer to mean one person when saying she, but for the witness to think of, and answer about, someone else. It is perfectly possible to refer to each actor, in a story or event, by his or her name rather than refer to him, they, and so on.

3. No language, which is inappropriate to the age, gender, education, culture or experience of the witness, should be used. Of course, it will always be impossible to be sure which words particular witnesses understand. Certainly, we cannot insist upon witnesses confessing to ignorance before we try to make sure they know what others, and in due course they, are talking about. Judges would have to exercise discretion, but they should be able to rely upon highly educated lawyers being able to find alternative acceptable expressions and explanations.

It may be objected that such rules are unnecessary. If there is confusion, say with regard to which person the lawyer and witness are discussing, then that will become evident during the proceedings. But, quite simply, that does not have to occur. If a misunderstanding does become evident, it is most likely to be turned into a comment upon the competence and reliability of the witness rather than on the less nervous, more experienced, well-educated lawyer. How often do lawyers apologise for having confused a witness, even a vulnerable witness, in comparison with using the witness' confusion to his or her advantage? Even if the judge were to criticise the lawyer, the damage has been done. The witness is anxious and concerned, but now all the more so.

It is the lawyer's responsibility to ask a comprehensible question rather than the witness' responsibility to understand somehow what the lawyer wants to know. But the 'power' balance works the other way. The lawyer has much more control. It is much more his or her workplace than it is the witness'. He or she is more familiar with and experienced in the courtroom. Even novice lawyers will have had some practice in asking courtroom questions, but giving a witness some practice may be interpreted as coaching and used to criticise.

4. Judges should treat it as the lawyer's responsibility to ask easily understood questions. Blame for misunderstandings should attach to the lawyer's question unless there is evidence that the witness is deliberately being unhelpful.

5. Judges should encourage witnesses to say when they are confused. They should emphasise that it is a positive, not an embarrassing thing, to do. It should flaunt no law, ethic or right of the defendant, if a judge were to explain to witnesses that if they are confused by a question then they should say so, and that it is the lawyer's responsibility to rephrase the question. Indeed the judge could demonstrate how important - and acceptable - it is for witnesses to ask for help by asking them a question, in terms that the witness cannot understand, so that the witness can practice asking for an explanation. And if it is suspected that a witness does not understand the question, and therefore cannot be giving proper evidence, the judge could test with such devices as repeating a question, but in the negative, to see if a consistent answer is given. If judges are not happy with this suggestion, then it might be included in a booklet of advice for witnesses.

Acting on this suggestion would not be partial, as between the parties, and inaction impartial. Some witnesses will be warned, or have the good fortune to be sufficiently self-possessed, in order to deal with such questions in court. Some might argue that it is acceptable to retain the status quo where it is acceptable that one party might have the good - or bad - fortune of having a skilful witness, but it would be wrong to change the status quo. But that does not take us very far away from the irrational trial by ordeal.

Witnesses should not be confused by the questions. This goes beyond understanding the individual words. It has to do with the structure of individual questions and how they are asked.

6. Only one question should be asked at a time.

7. Questions should not contain more than 12 words. Obviously there is an element of judgement here. It could be slightly more than 12, but certainly not double. Lawyers should not complain about such a rule. Evans (1983, 1993), who has practised in both the English and Californian courts, recommends 10 words. Put another way, lawyers are encouraged to imagine that if their questions were typed out in a book, then they should take up no more than a single line. Quite simply, if you need more than 15 words to ask a question, it is already too complex. You, the highly educated, trained, and experienced person in your work environment, should be able and willing to turn your question into two or more separate questions. Yes, academic lawyers and others can and should separate their unnecessarily complex thoughts into separate and shorter sentences. However written sentences can be re-read and their structure re-examined.

8. Questions should contain no more than one pair of commas or, put another way, no more than one sub-clause. Arguably sub-clauses should be banned entirely. But till then, only one should be allowed. Each sub-clause makes the sentence and the topic much more complex. It is not just a case of understanding and remembering each clause, but understanding how they relate to each other. Indeed, work on understanding how the parts relate to the whole can only begin after the whole question has been heard, its words understood, and its form analysed.

9. Questions should not be asked merely for effect. For example, rhetorical questions (e.g. "Don't you believe in telling the truth?") are asked without an expectation that they will be answered. Rather they are designed to have an effect, to make an impression on the judge and any jury. For the witness it can be difficult to interpret, not just from the question but the manner in which it is asked, whether such questions are supposed to be answered. And yet the witness may have something to say about the topic. For example the witness may, indeed, believe it highly important to tell the truth and be trying to do so, but may be confused by the question. Given that lawyers are only supposed to be asking questions (not making speeches) during the examination and cross-examination stages, judges should first ban the question and invite witnesses to comment on rhetorical and similar questions. Lawyers have other opportunities to comment upon what the witnesses did or did not say.

HOW LAWYERS CAN GIVE THE EVIDENCE
It would be a mistake only to be concerned with individual questions. The sequencing of questions and the adoption of a number of devices can easily create confusion and error. In particular, asking a number of short questions can be done with such a speed that the witness will find it very difficult to think about the broader picture being 'painted' by the question and answer sequence. We know about the problems with leading questions, which suggest answers, and we prohibit them being asked in direct (but not cross) examination. It has been shown that leading questions are an excellent method of obtaining false information (e.g. Loftus and Palmer 1974). But we pay much less attention to directed, as opposed to open, questions. An 'open' question invites the witnesses to choose what to say; it empowers them. An example would be: "And what happened next?" But a 'directed' question seeks to restrict what the witness can say. An example would be: "Would you describe the injuries as 'serious' or 'life-threatening'?" Instead of the witnesses being allowed to choose their own words - "How would you describe the injuries?" - they are being offered a choice of two answers. The witness is entitled to insist upon using a different word or phrase entirely, but in the stressful setting of a court, is likely to adopt one of the lawyer's expressions. And if the lawyer is competent one of those expressions will be more likely to be adopted and useful to his or her side.

The reality is that with closed questions the lawyer, not the witness, gives the evidence. The lawyer chooses which pieces of the evidentiary picture will presented. So, if appearing for the defence, the lawyer may well ask questions about where the victim was not hit, and not dwell on where the victim was hit. Closed questions enable the lawyer to choose which parts of the picture will be painted. And with which 'colours.' The word 'strike' has different connotations to both 'hit' and 'touch.' The experienced lawyer can ask a sequence of short, seemingly simple, questions where the witness only has to say "Yes" or "No." In those circumstances the evidence is really in the question rather than in the answer. The lawyer is using the witness to present a picture of the evidence using his or her own language. The answer is perfectly predictable by the lawyer, which is why they have asked it! Witnesses, whether vulnerable or not, will find it difficult to understand the tactical implications of the question. Once the question is answered, in the terms suggested, it will be very difficult, without appearing to be contradictory or changing stories, to insist on an alternative expression. And it is very difficult for a witness to keep track of which parts of the evidence have and have not been covered by the questions. In the event of the witness being alive to this technique he or she should respond along the lines of: "But you have not asked me about … and it is critical you get the full picture." Such an answer is liable to earn a rebuke like: "Just wait; all in good time; I will come to that."

Witness may be sufficiently assertive to challenge the terms involved in a couple of directed questions. But if they are challenged repeatedly, then the witness will appear to be awkward and tendentious. A feature of closed questions is that they appear innocuous. Indeed they can be very appealing, because they get short answers from a witness who might otherwise provide long answers because of being loquacious, nervous, or confused. For example, the question: "How would you describe the injuries?" might evoke the innocent but unhelpful answer: "By telling someone about them." But, more seriously, such questions may obscure issues that deserve fuller explanation. For example, the following question seems very straightforward:

Either it did or it did not. But, whilst true, it did take two hours, the reasons for the delay may be important. If the witness is not allowed to explain completely then the court misses important evidence.

Many lawyers will argue that they must be allowed to challenge witnesses, so that courts can obtain an impression of their qualities. But do such questioning techniques reveal the witnesses' poor character or their innocence? All that we need to do is train witnesses to respond to such and other questions with the intervention, turning to the judge:

Judges should find it difficult to insist that evidence, which they are being told is false, should be introduced. They should ask the witness to explain, whereupon the witness should explain his or her problems with the question. Such devices could neutralise at least some lawyers' tricks or techniques. If we can accept that, then it should follow that we should not distinguish between witnesses upon the basis of their training in witness craft. Rather we should remove the need and motivation for resorting to such responses.

Directed questions need to be regulated. Ideally they should be banned. That would not prevent evidence being produced. That can be achieved by an open question. A ban might be restricted to vulnerable witnesses and/or to jury trials. Directed questions are particularly powerful, in terms of painting a verbal picture, when asked in a sequence. The witness has been inhibited, at least, from choosing his or her own words and adding any comment or explanation.

10. If allowed at all, a sequence of directed questions should be limited in number, say to five, and the judge should summarise the cumulative effect of what the witness has just said and invite the witness to comment.

Other sequences of questions can unfairly trap unwary witnesses. For example there is 'pining out.' Here the lawyer asks the witness on the other side a sequence of questions. The distinctive feature of the questions is that they allow the witness to show off, to reveal how much he or she knows. The witness is lulled into a false sense of security and the judge and/or jury become impressed with the witness. But the lawyer then asks a question that he knows the witness cannot answer. (The lawyer knows what the witness can and cannot say because of having read the witness' statement.) For example, the lawyer might ask the witness a lot of questions about an incident. The answers show that the witness knows a great deal about it. Then the lawyer asks a question which he or she knows that the witness cannot answer. The effect, and that is why it is used, is to devastate the witness' reputation. The judge and/or jury suddenly discover that the witness is not as knowledgeable, competent, complete, etc. as first appeared. And the witness is embarrassed. But, once again, there may be good reasons for the witness' sudden silence. Will the witness be competent enough to explain?

Once again, unless we act against 'pining out' questions, some lucky witnesses will come prepared to answer something like the following:

Surely it ought not to depend upon the luck, or otherwise, of the witness knowing some techniques for self-preservation.

TJ AND RJ TO THE RESCUE?
Clearly the mandated processes of asking questions of witnesses in court can have damaging, anti-therapeutic effects. Indeed, that has been recognised in several jurisdictions. Reforms have been instituted. Unfortunately they have been limited to children and 'vulnerable' people. However, the problem is not the witness but the lawyer. It is not the witness' fault that he or she is young, vulnerable, naïve, and inexperienced in law craft. But lawyers are entitled to take advantage of witnesses.

Lawyers are also entitled to take advantage of the legal rights of their clients. In virtually all debates on this topic this gives them some form of moral supremacy; they are protecting the legal rights of their clients. The rights and interests of suspects, witnesses, and victims are not even treated as approximately equal. The problem for therapeutic jurisprudence is that, although it may recognise the experiences of witnesses, it respects legal rights as prior. A distinction may, however, be drawn between the legal rights of the person perceived as experiencing anti-therapeutic effects and the legal rights of other actors. The legal rights of one group do not automatically trump or outweigh the rights of another. But we do not treat witnesses as holders of legal rights in a similar sense to how we treat defendants. And there is always the onus of proof; the defendant can only be convicted by proof beyond reasonable doubt. At the very best it might be perceived as a clash of interests and rights.

As has been seen, there may be some ways around TJ's problem, such as by requiring judges to prohibit and restrict questions that can have a harmful effect upon the evidence and witness. These, it has been argued, can be devised and enforced without harming the defendants' legitimate rights and interests. True, they would severely restrict lawyers' use of tactics and thereby adversely affect defendants. But, the core question is, should defence lawyers be allowed to use tactics whose aim or effect is to confuse and upset?

Many may prefer not to answer that question. Rather, given that they are not vulnerable witnesses, they may prefer to answer a different question. They may suggest that change could be harmful, and as it cannot be shown that it would not, there should be none. But it does not follow that the degree of possible harms resulting from changes in practice will be more serious than the perfectly predictable injuries that we know will occur from continuation of the status quo. We are gaining a quite significant knowledge base of how to question people appropriately. Lawyers should not be entitled to remain a class apart, ignorant, and unwilling to apply the advances.

There will be no substantial changes unless there is a dramatic readjustment in the balances. Victims and witnesses remain comparatively powerless. The debate is still undertaken in lawyers' terms. A more sociological perspective, which would include issues of power, cannot be ignored. Witnesses do not have power; lawyers, who need them to make a living, do. Meanwhile, witnesses are voting with their feet and are increasingly unwilling to give up their time to inform the courts. Surely therapeutic jurisprudence cannot avoid taking sides.

Slobogin, who identified five key issues for therapeutic jurisprudence, has asked: "How is this conflict between therapeutic and constitutional values to be resolved?" (1996:785). Perhaps we should return to the medical analogy. Doctors regularly discover illness, disability, and other anti-therapeutic experiences. They provide treatments, advice, potions, and pills. But they also know, and worry about to different extents, that the 'real' problems are poverty, ignorance, and inadequate services. Until those underlying causes are tackled they will continue treating symptoms. Screens and video-links are treatment for symptoms. The law is the problem, not just a solution. Witnesses, not just individuals or the vulnerable, deserve better. Reformism is not enough. Sometimes things are not just unfair and bad. We should not have to be polite about them.

REFERENCES

Abbott, Howard. 1987. Safer by Design: The Management of Product Design Risks Under Strict Liability. London: The Design Council.

Bull, Ray. 1998. "Obtaining information from child witnesses." Pp. 188-209 in Psychology and Law: Truthfulness, Accuracy and Credibility, edited by A. Memon, A. Vrij and R. Bull. London: McGraw Hill.

Crown Prosecution Service. 2002. Narrowing the Justice Gap. London: Crown Prosecution Service.

Evans, Keith. 1983. Advocacy at the Bar. London: Financial Training Publications.

Evans, Keith. 1993. The Golden Rules of Advocacy. London: Blackstone.

Graef, Roger. 2000. Why Restorative Justice? Repairing the Harm Caused by Crime. London: Calouste Gulbenkian Foundation.

Gudjonsson, Gisli. 1992. The Psychology of Interrogations, Confessions and Testimony. Chichester: Wiley.

Hart, H.L.A. 1963. "Introduction." Pp. vii-xi in The Idea of Justice and the Problem of Argument, by C. Perelman. London: Routledge.

HMIC (Her Majesty's Inspectorate of Constabulary). 2002. A Report on the Joint Inspection into the Investigation and Prosecution of Cases Involving Allegations of Rape. London: Home Office.

Kress, Ken. 1999. "Therapeutic Jurisprudence and the Resolution of Value Conflicts: What We Can Realistically Expect, in Practice, from Theory." Behavioral Sciences and the Law 17:555-588.

LaFond, John Q. 1999. "Therapeutic Jurisprudence: Can TJ Be Normatively Neutral? Sexual Predator Laws: Their Impact on Participants and Policy." Arizona Law Review 41:375-415.

Lamb, Michael E., Sternberg, Kathleen J., Orbach, Yael, Hershkowitz, Irit and Esplin, Phillip W. 1999. "Forensic interviews of Children." Pp. 253-277 in Handbook of the Psychology of Interviewing, edited by A. Memon and R. Bull. Chichester: Wiley.

Loftus, Elizabeth F. and Palmer, J.C. 1974. "Reconstruction Of Automobile Destruction: An Example Of The Interaction Between Language And Memory." Journal of Verbal Learning and Verbal Behaviour 13:585-589.

Lukes, Stephen. 1974. Power: A Radical View. London: Macmillan.

McEwan, Jenny. 1998. Evidence and the Adversarial Process; The Modern Law, 2nd edition. Oxford: Hart.

Memon, Amina and Bull, Ray. (Eds.). 1999. Handbook of the Psychology of Interviewing. Chichester: Wiley.

Milne, Rebecca and Bull, Ray. 1999. Investigative Interviewing: Psychology and Practice. Chichester: Wiley.

Rottman, David and Casey, Pamela. 1999. "Therapeutic Jurisprudence and the Emergence of Problem-Solving Courts." National Institute of Justice Journal, July, 13-19.

Scottish Executive. 2001. Vital voices: Helping vulnerable witnesses give evidence, Edinburgh: Stationery Office.

Simon, Leonore M.J. 2003. "Proactive Judges: Solving Problems and Transforming Communities." Pp. 449-472 in Handbook of Psychology in Legal Contexts, 2nd edition, edited by D. Carson & R. Bull. Chichester: Wiley.

Slobogin, Christopher. 1996. "Therapeutic Jurisprudence: Five Dilemmas to Ponder." Pp. 763-793 in Law in a Therapeutic Key: Developments in Therapeutic Jurisprudence, edited by D. B. Wexler and B. J. Winick. Durham, NC: Carolina Academic Press.

Spencer, John R. and Flin, Rhona. 1993. The Evidence of Children; The Law and the Psychology, 2nd edition. London: Blackstone.

Strang, Heather and Braithwaite, John. (Eds). 2001. Restorative Justice and Civil Society. Cambridge: CUP.

Walker, Anne G. 1994. Handbook on Questioning Children: A Linguistic Perspective. Washington, DC: American Bar Association.

Wexler, David B. (Ed). 1990. Therapeutic Jurisprudence: The Law as a Therapeutic Agent. Durham, NC: Carolina Academic Press.

Wexler, David B. 2000. "Therapeutic Jurisprudence: An Overview." Thomas M. Cooley Law Review 17:125.

Wexler, David B. and Winick, Bruce J. 1991. Essays in Therapeutic Jurisprudence. Durham, NC: Carolina Academic Press.

Wexler, David B. and Winick, Bruce J. 1996. Law in a Therapeutic Key: Developments in Therapeutic Jurisprudence. Durham, NC: Carolina Academic Press.

Contact Information
Law and Behavioural Sciences at University of Southampton, Southampton, S0171BJ, U.K. Email: d.c.carson@soton.ac.uk. Website: <http://www.soton.ac.uk/~law/staff/carson.html>. back