Summing up in criminal trials - a new direction? 


Conference on jury research, policy and practice, Sydney, NSW, 11 December 2007


James Wood, AO QC
Chairman, NSW Law Reform Commission


The purpose of a summing-up in a criminal trial is to provide the jury with the assistance which it needs to perform its task. That task, simply stated, is to determine whether, upon the evidence presented, the prosecution has proved the guilt of the accused. Historically, the role of the judge in this respect had not been seen to be unduly difficult. Summings-up were short and to the point, and not overburdened by complex analyses of the law, or by commentary on the way in which the jury should approach the evidence.

By and large, they complied with the counsel of Sir Leo Cussen, authoritatively cited in Alford v Magee: 1

[T]he late Sir Leo Cussen [a former Chief Justice of Victoria] insisted always most strongly that it was of little use to explain the law to the jury in general terms and then leave it to them to apply the law to the case before them. He held that the law should be given to the jury not merely with reference to the facts of the particular case but with an explanation of how it applied to the facts of the particular case. He held that the only law which it was necessary for them to know was so much as must guide them to a decision on the real issue or issues in the case, and that the judge was charged with, and bound to accept, the responsibility (1) of deciding what are the real issues in the particular case, and (2) of telling the jury, in the light of the law, what those issues are.

Unless a summing-up is focused in relation to the essential issues in the case, and delivered in a form which is comprehensible to a jury made up of lay persons, it will fail its purpose. Whether current practice and expectations concerning the content and depth of a summing-up are frustrating this objective has become a cause for concern in this country and overseas. 2 The New South Wales Law Reform Commission has a current reference in relation to the requirements for a summing-up; and a similar inquiry is being undertaken by a working party led by Lord Justice Latham in England. In the United States, much work has been undertaken in rewriting model jury instructions using plain English techniques, while in other Australian jurisdictions, work is underway in preparing or reformulating model directions. Concerns in relation to the current approaches to summing up have been expressed by senior members of the judiciary and by jury research. 3

There are a number of reasons for the concerns which have been expressed, including:

  • the complexity of the criminal law, arising in part from the legislative formulation of the several offences and defences which may require consideration;
  • the extent to which it has been seen necessary for the judge to intrude into credibility issues, by way of warnings and cautions, and commentary on the evidence;
  • the use of legalese and unnecessarily complex formulations of issues in ways that pay little regard to the requirements of plain English;
  • the ever-increasing length of criminal trials, which tends to see the best-educated and qualified members of the community avoiding jury duty, either because they are entitled to an exemption as of right, or are excused on hardship grounds;
  • the tendency of judges to venture into unnecessary detail concerning the law, and to give unduly comprehensive directions in relation to the several areas of evidence that might call for a caution;
  • the willingness of appellate courts to subject the transcript of a summing-up to a minute analysis, and to entertain objections to the directions that were not taken at trial;
  • the failure of the system to respond to the need to move from a tradition of orality, with which past jurors were comfortable, to the ways in which the current generation of jurors are accustomed to receiving and processing information;
  • the emergence of ever more sophisticated forensic investigation technologies, the understanding of which may depend upon the capacity of the jurors to absorb and apply complex expert evidence;
  • the limited extent to which issues can be defined, and defence disclosure required, under existing procedural legislation, prior to the commencement of the trial.

These are some of the matters which the current inquiries will need to confront. Several of these are addressed, in a preliminary way, in this Paper, starting with the increased appellate involvement in the criminal law.


There has been a significant increase in the involvement of intermediate appellate courts, and of the High Court, in criminal matters in recent decades. This has had the effect of increasing the complexity of trials, and of the number of directions that are now required to be given to juries. The reasons for this are several. In relation to the High Court, they were identified by Justice Michael Kirby as including:

  • the introduction of leave requirements, and of criteria relevant to the concept of special leave, that allowed the Court to devote greater attention to the criminal law;
  • a departure from the disinclination of earlier members of the Court to interfere in the administration of criminal justice;
  • the provision of legal aid, and the development of a specialised criminal bar, which have led to counsel becoming more imaginative in exploring new issues and in testing earlier decisions;
  • the establishment of intermediate courts of criminal appeal whose decisions identify issues of importance for legal doctrine, and hence for review by the High Court. 4

Some of these reasons also help to explain the growth of appeals in the intermediate courts. One result has been a focus on a more strict adherence to legal doctrine, and to the observance of procedural rules. Another result has been the emergence of two separate criminal bars; a trial bar and an appellate bar, whose skills and knowledge do not always coincide. That this is so can be seen from the large number of appeals that are conducted, and succeed, on grounds that were not taken at the trial.

Just as trial counsel have difficulty in keeping abreast of the emerging developments in the criminal law, so too do first-instance judges, who are also disadvantaged by the lack of assistance which they can expect to receive during the trial.

The increase in the number and complexity of the directions which are now required, as a result of these decisions, has inevitably multiplied the opportunity for first-instance error. At the same time, the more that judges try to avoid error by framing their directions with an eye to satisfying the appeal courts, the less comprehensible such directions are likely to be to lay jurors.

Although Rule 4 of the Criminal Appeal Rules requires leave to be granted before a ground of appeal will be allowed in relation to any direction, or omission to direct, which was not the subject of an objection at trial, the practice continues unabated of such points being raised on appeal. Almost inevitably, the consideration of the leave question becomes caught up in the more critical consideration of whether the error, if there was one, is such that the case is suitable for an application of the proviso under s 6(1) of the Criminal Appeal Act. That in turn depends upon the court deciding for itself whether a substantial miscarriage of justice actually occurred. 5

The court has recognised that no simple universally-applicable description can be given of the expression 'no substantial miscarriage of justice', and that there may be cases where it would still be proper to allow an appeal and order a new trial, even where the appeal court was persuaded to the requisite degree of the appellant's guilt. Such cases may include those where there had been a 'significant denial of procedural unfairness at trial'; or where there had been a sufficiently 'serious breach of the presuppositions of the trial', such that the trial had so far miscarried as hardly to be a trial at all. 6

Questions arise, accordingly, as to whether anything can be done to simplify, or to restructure, the way in which summings-up are delivered, so as to make them more relevant and comprehensible, while avoiding the kind of errors that lead to successful appeals.

Questions also arise as to whether there should be some different approach taken by the appellate courts, either in the way in which they deal with these appeals, or in the way that they might provide some greater guidance to trial counsel and judges, in order to avoid, or at least to minimise, the risk of error.


Most jurisdictions have adopted model or standard directions in relation to the elements of various offences and defences, and in relation to the several cautions which may need to be given concerning particular categories of evidence. While they can provide an exceedingly useful starting point, they tend to suffer from the difficulties that:

  • they are usually expressed in the terms that were used in appellate decisions or in legislation 7 and, as a result, commonly employ multiple subordinate clauses, double negatives, and complex sentence structures;
  • judges very often recite them in a ritualistic way, with every possible qualification or variation which they might encompass, whether or not they were appropriate to the issues in the case at hand;
  • sometimes they are bundled up in a series of preliminary directions that risk overwhelming the jury, rather than being placed in context when dealing with the issues or respective cases.

Mindful of this, the NSWLRC has commissioned the Plain English Foundation to review a sample of summings-up delivered in the Supreme and District Courts which have employed the Bench Book directions, to identify any areas where they are unlikely to have reflected a level of communication that would be appropriate for the usual jury mix.

The challenge in this respect is to identify a suitable level of communication and structure for a summing-up that will convey the essential explanation of the relevant law and issues while remaining comprehensible to the average lay juror. 8

Examples are legion of concepts that are customarily employed in summings-up that are not used in everyday speech, or that are readily understandable. To mention but a few, directions given in circumstantial evidence cases 9 in terms of a 'rational inference or hypothesis consistent with innocence'or one which is 'inconsistent with guilt', and concepts such as 'malice' 10 and 'proximate cause'are likely to be foreign to most jurors. Similarly, a direction to 'scrutinise the evidence with great care'is likely to raise a question as to whether it adds anything to the standard direction not to convict an accused unless satisfied of his or her guilt beyond reasonable doubt.

Hopefully, this review will identify the danger areas in the way that jurors have traditionally been instructed, and identify a structure that is more relevant for the current generation of jurors. If so, it might be that judges will be able to avoid the experience that so many have encountered of the eyes of the jury glazing over when given an instruction such as that seen in one recent case:

If you are satisfied that one or more of the accused killed [the victim] but you are not satisfied that the Crown has not proved beyond reasonable doubt that the accused did not act in self defence [he] is not guilty of either murder or manslaughter. 11

Of equal importance, however, for the formulation of standard directions in plain English is the need to encourage judges to be selective in their use, and to confine the summing-up to that which is relevant to the issues in the trial, so as to conform more closely with the Cussen formula, 12 that is, unless an alternative approach that would be more relevant to modern-day conditions can be developed.


It is obviously an important part of the task of the jury to reach conclusions in relation to the credibility and reliability of the evidence of individual witnesses, so as to come to a conclusion upon the facts of the case.

The extent to which the jury members should receive assistance from the trial judge in relation to this aspect of the trial, which is within their exclusive province, is however open to debate.


The courts have taken a traditional stance based on the common law that judges should give juries assistance by way of a caution in relation to the way that they approach the evidence of witnesses:

  • concerning identification; 13
  • in sexual assault cases where there is no corroboration, particularly where there is a delayed complaint; 14
  • where they are prison informers; 15
  • where they are accomplices (or persons who might reasonably be supposed to have been criminally concerned in the events giving rise to the proceedings), 16 particularly where, as a result of assistance provided to law enforcement, they have received an immunity from prosecution or a reduced sentence 17 , with the consequent risk of the benefit being lost if they fail to give evidence;
  • where character is raised;
  • where the accused does not give evidence. 18

The Evidence Act also makes provision for the giving of a caution as to the possible unreliability of the evidence in these and in certain other situations, including evidence the reliability of which may be affected by age, ill health or injury; hearsay evidence; and evidence of official questioning that has been recorded in writing, but not signed or otherwise acknowledged in writing by the defendant. 19

Further provision is made with respect 'vulnerable people'for the giving of a warning in relation to:

  • the reception of the evidence of a vulnerable person of a previous representation in the form of a recording made by an investigating official; 20
  • a vulnerable person giving evidence by means of closed-circuit television or similar technology. 21

In some instances, these directions have become lengthy dissertations which border on giving judicial evidence, in so far as they involve a reference to the experience of trial judges. Commonly, they are supplemented by an observation to the effect that they are routinely given in any case where such evidence is led, and that the jury should not take the direction as conveying any personal conclusion that the judge has reached in relation to the credibility of the relevant witness. Quite what the average juror makes of this observation, when given a stern warning that is expected to carry the judge's imprimatur, is another matter. 22

The justification for warnings (in relation to matters of which it is assumed jurors are unaware) and for comments (in relation to matters which it is assumed they may have overlooked) depends on the assumption that trial judges, by their special experience in the criminal law, possess greater knowledge and comprehension in relation to the inherent dangers associated with these forms of evidence. 23 Whether that assumption is well-grounded is, at least in some respects, questionable.

Moreover, careful reflection is often required in determining whether some matter that might potentially affect the reliability of a particular piece of evidence is such as to require a warning, or whether it can be left to a comment, or left simply as a matter for the jurors to weigh using their own judgment. 24 The decision which the trial judge makes at this point in the context of the trial can be critical since, if it is found to be an erroneous one, it is likely to lead to a successful appeal. 25

Perhaps the time is ripe to reconsider whether the assumptions are soundly based for all or some of these warnings or comments and whether, in fact, jurors do lack the fairness, underlying knowledge, experience of life, and common sense which has underpinned their use.

To some extent, the difficulty which arises when these warnings are given, particularly in terms instructing the jury that it would be 'dangerous to convict'on the particular body of evidence unless, having scrutinised it with great care, they are satisfied of its truth, has been moderated by legislative amendment. 26 However, there will remain occasions where directions will need to be given, particularly in sexual assault cases, in similar terms to those which have been productive of an inordinate number of appeals. 27

Consideration accordingly does need to be given to this aspect of a summing-up, and as to the desirability of the use, in any instance, of the expression 'dangerous to convict', with its implicit encouragement to acquit.


Similar questions arise in relation to some of the other directions that regularly arise for consideration on appeal. There is, for example, the vexed question of post-offence conduct, such as flight and lies told by an accused, which calls for somewhat sophisticated distinctions to be drawn and respected between such conduct demonstrating an admission of guilt or a consciousness of guilt, and those going only to credibility. 28 As Justice Kirby pointed out in Zoneff, there is an inevitable difficulty for jurors in understanding and applying subtle distinctions between the use of evidence for one purpose and not another, and a risk that a warning not to use it for a particular purpose may be counterproductive in exciting the very reasoning which is forbidden, but which might otherwise not have occurred to them.

A real question arises as to whether, with legislative authority, the lies direction could be substantially shortened and reduced to a bare reminder to the jury members to take into account, as they see fit, any evidence showing that the accused has lied, bearing in mind that there may be reasons other than an acceptance of guilt for having done so. Similar formulations are regarded as acceptable in other jurisdictions, for example, in California.


It has been conventional practice for jurors to be informed that it is appropriate for them to observe the demeanour of witnesses when giving their evidence, and to take it into account as part of the fact-finding process. Appellate courts have confirmed the permissible use of demeanour by judges for this purpose, 29 even though they have recognised the existence of a significant body of scientific research that casts doubt on the ability of judges or anyone else to tell truth from falsehood on the basis of appearances alone. 30 The general conclusion of these studies is that the prospects of determining whether a witness is telling the truth from mere visual appearance, dependent on facial expressions, bodily movements, manner of speech and so on, is no better than chance. A real question arises accordingly of whether it is appropriate to continue to give jurors a demeanour direction, or whether it is possibly misleading to do so.

Judicial comment

While it is permissible for trial judges to express an opinion on, or comment about, the facts, provided that it is made clear that the ultimate decision on the facts is for the jury, 31 a question remains as to whether this is a helpful practice today. Commonly, where the trial judge goes too far, or offers some comment or argument that was not identified by counsel, this will lead to an unnecessary appeal. 32 Moreover, as has been pointed out, there is a tension between a judge commenting on the facts and then cautioning the jury that any opinion apparent from this comment should not influence them, such that it is questionable whether there is any point in making the comment. 33

Has the time come when modern jurors can be trusted to deal with the evidence upon the basis of their own experience and common sense, and with the assistance of the arguments of counsel, without the need for any additional input from the judge? If taken too far, or made too obvious, such comment may even be counterproductive, as Sergeant Sullivan suggested in the 1930s, when he said foremen of juries at the Old Bailey should be asked whether 'they found for his Lordship or against him?'

In Taleb v R,34 the court suggested that the width of a trial judge's discretion to comment upon the evidence was narrower than had been permitted in the past, and that greater restraint was now to be expected. The risk with this form of comment is that it will be seen by the jury as a binding direction of law, 35 or even as a direction to convict. 36

A change of direction?

The descent of judges into the arena by giving directions or by making comments on the significance that the jury should, or should not, attach to the evidence has not gone uncriticised. 37 There is a concern that the giving of a judicial imprimatur to these directions or observations, particularly in relation to matters that would appear to be matters of common sense, or within the common experience of life, risks giving them an undue prominence. Moreover, they often serve to add considerably to the prolixity of the summing- up, and invite the glazed-eye phenomenon among jurors who are unable to place the guidance they are receiving into context.

In summary then, should there be a retreat from the extent to which judges intrude into the fact-finding function of juries by way of comment or warnings, leaving it to jurors' own common sense and experience? After all, it was by reason of their supposed qualities, as representatives of the community in these areas of reasoning, that the fact-finding process was assigned exclusively to jurors.

At the least, would it be appropriate for judges to confine their warnings and comments on the evidence to cases where there is a particular basis for concern as to the witness's credibility or reliability, or where it can be clearly established that the evidence does fall into an area outside the ken of lay persons, beyond the mere fact that the evidence falls into one of the identified categories of evidence that can be unreliable?

Alternatively, would it be more appropriate to:

  • leave it to counsel to identify the possible difficulties with the evidence through cross examination and addresses;
  • allow expert evidence to be led, for example, in relation to the significance of delayed complaint in sexual assault cases;
  • eliminate those directions, by statute, where it is accepted that modern jurors do have sufficient understanding and experience to appreciate the potential problems with certain categories of evidence; or
  • at the least, reduce the content and breadth of some of the warnings or comments that have become the norm to a more general and brief observation?

This seems to be an area where jury research could usefully establish:

  • whether or not lay persons do lack the knowledge or understanding on which the justification for these directions rests; and
  • what such persons do in fact make of the directions that are given.


The requirement for proof of the guilt of the accused 'beyond reasonable doubt', and the current inability of a trial judge to be drawn into any explanation of it and, in particular, to avoid using the 'domestic'analogy, has become an entrenched principle of the criminal law and of trial practice. Although the courts have observed, on several occasions, that this formula is understood well enough by the average person in the community, and does not require or permit of any elaboration, 38 it is a regular experience of trial judges that jurors do ask for some explanation of what it involves. 39

The standard explanation then given, that a 'reasonable doubt'is 'one which the jury regards as reasonable'or that 'the words mean what they say', does little to assist. It has also been held that it is undesirable to observe that the words are 'ordinary everyday words'. 40

Interestingly, the New Zealand Jury Project conducted for the NZ Law Commission suggests that, unaided by any greater explanation of the concept, jurors tend to reduce the test to one involving a percentage certainty of somewhere between 50% and 100%.

The way in which judges should sum up to juries on the concept of proof beyond reasonable doubt has been considered by a Court of Appeal in New Zealand constituted by five justices. 41 After an extensive review of the provenance of the direction, of the current authority in NZ, of the approach taken in other jurisdictions, and of several jury studies, Justices William Young, Chambers and Robertson observed that they were inclined to the view that judges should explain the concept as follows:

The starting point is the presumption of innocence. You must treat the accused as innocent until the Crown has proved his or her guilt. The presumption of innocence means that the accused does not have to give or call any evidence and does not have to establish his or her innocence.

The Crown must prove that the accused is guilty beyond reasonable doubt. Proof beyond reasonable doubt is a very high standard of proof which the Crown will have met only if, at the end of the case, you are sure that the accused is guilty.

It is not enough for the Crown to persuade you that the accused is probably guilty or even that he or she is very likely guilty. On the other hand, it is virtually impossible to prove anything to an absolute certainty when dealing with the reconstruction of past events and the Crown does not have to do so.

What then is reasonable doubt? A reasonable doubt is an honest and reasonable uncertainty left in your mind about the guilt of the accused after you have given careful and impartial consideration to all of the evidence.

In summary, if, after careful and impartial consideration of the evidence, you are sure that the accused is guilty you must find him or her guilty. On the other hand, if you are not sure that the accused is guilty, you must find him or her not guilty. 42

The current position is that in England and Wales, 43 Canada, 44 and New Zealand, the concept has been equated to one of the jury being or feeling 'sure'of the guilt of the accused. In each jurisdiction, some further explanation has been regarded as acceptable, although the trend, particularly in England, is to avoid the use of the expression 'beyond reasonable doubt'in case that causes some confusion as to the meaning of the expression 'sure'. Anecdotally, it has been reported that juries rarely, if ever, seek any further explanation of it, although some studies suggest that it is not without its problems, 45 and that a direction framed in simple terms requiring the jury to 'be sure'is less stringent than one expressed in terms of being 'satisfied so that you are sure'.

The question arises whether, having regard to the problems which jurors do appear to have with the succinct direction which is used in this country, and its inherent imprecision, there is now occasion, at appellate level, or though legislation, to revisit this fundamental requirement for proof of criminal guilt.


A question arises as to whether it should continue to be regarded as either appropriate or necessary for judges, in an adversarial system of justice, to direct the jury on the possible availability of defences or verdicts for lesser offences where they have not been raised by counsel.

While the authority for doing so continues to be Pemble v The Queen, 46 it seems that very often trial judges will be anxious to include such directions, for example, in relation to the various circumstances and partial defences that might give rise to a manslaughter verdict in the case of an accused charged with murder, on the most tenuous of bases, in order to appeal-proof the summing-up. In some cases, these directions are given, and indeed the authorities suggest that they need to be given, even though the defence expressly objects to that occurring. 47

The question which arises is whether it should continue to be necessary for trial judges to give these directions, particularly in cases where the defence prefers to go to the jury on the substantive count alone.

This is not an insignificant consideration, since it adds to the complexity of the trial in circumstances where counsel in the closing addresses have given no assistance to the jury on the alternative. Certainly, trial judges should be astute not to venture into this area where to do so is to advance some unreal or fanciful possibility or to raise a speculative hypothesis. 48 Perhaps greater attention in this respect should be given to the adversarial context in which criminal trials are conducted, 49 leaving it to the parties to settle the issues for determination.


In some areas, most noticeably defences, it is the reasonable or ordinary person through whose eyes the jury is required to assess the particular facts of the case. It is curious why, when the jury is expected to apply such a test, they are routinely instructed not to perceive themselves as such a person, but to assess the matter through someone else who they consider to have the qualities required. Quite how they select such a person, and what attributes they are expected to assign to him or her, remains unexplained, and very much a matter for conjecture.

Since they are selected as representatives of the community, what is wrong in accepting the response of the ordinary or reasonable person to be that which represents their collective reaction? In fact, do they, in reality, search for some hypothetical person who has some view or reaction that differs from theirs? If not, why should the law add this complexity, and possibly confusing element, to the mix?


A real question arises as to whether the kind of lengthy summary of the evidence that commonly occurs achieves anything in circumstances where the jury have heard the evidence themselves, have received the benefit of addresses from counsel, and potentially have access to the transcript or sound or video recording of the evidence if they wish to check some aspect of it.

There is a risk that undue emphasis on this part of the trial will swamp the more helpful areas of guidance which could be contained in a short and well-focused summing-up or, alternatively, will cause a bored and tired jury to switch off.

Certainly, the tedious and wholly unnecessary exercise sometimes encountered that involves a summary of the evidence of each witness in turn, without any attempt to marshal it in relation to the critical issues, is to be avoided and discouraged by appeal courts at all costs. 50

In some jurisdictions in the United States, judges do not touch upon evidence and, in some States, they are expressly forbidden from doing so, without any apparent detriment to the trial process. In Scotland, where jury directions are quite brief, even in complex trials, very little reference is made to the evidence beyond that which is necessary to identify the issues and to lay the basis for any appropriate warning.

Greater encouragement to trial judges to reduce this aspect of the summing-up to the minimum that is necessary to focus the jury on the real issues and on the significant aspects of the prosecution and defence cases would be a helpful starting point in achieving the brevity, accuracy and comprehensibility that should be the objective of a summing-up. 51

Advantage could be taken more often of the amendment to the Crimes Act, now incorporated in the Criminal Procedure Act, 52 permitting a trial judge to exercise a discretion not to summarise the evidence if, in all of the circumstances of the trial, an opinion is reached that a summary is not necessary. 53

Lengthy summaries of the addresses are also unlikely to provide much assistance. What, it may be asked, is the reason for judges summarising the addresses of counsel which the jury have only recently heard? This carries the risk of diluting their effect or, worse, becoming an occasion for judges to venture into the arena by advancing additional arguments, 54 or being tempted to respond to the arguments or to make them more intelligible, leading, on occasions, to applications for a discharge of the jury, on the basis that the case of one party has not been fairly left.

Generally, it is the practice for judges to provide an outline of the defence case, 55 although, strictly speaking, no rule of law or practice dictates that they must do so. 56 Where the case for the prosecution is summarised, then fairness will obviously require that the case for the defence also be summarised accurately and fairly. 57 However, save in special circumstances, there appears to be little advantage in the judge providing a detailed regurgitation of the addresses.


Quite apart from the content of the directions, there are some other factors related to the way that judges sum up to juries which potentially reduce their value. They include:

  • the timing of the summing-up, which means that the jury is commonly required to sit through a morass of evidence without sufficient guidance as to what is ultimately to be of importance for their decision-making (particularly is this so where the issues have not been identified pre-trial or where the defence elects not to present any opening address); and
  • the fact that the instructions are often given orally in a somewhat routine, if not ritualistic, way that does not take account of the fact that not all people absorb oral material well, and that does not make much effort to marshal the evidence of competing cases against specific issues.

This gives rise to the question whether the jury should be given greater assistance by appropriate directions at the commencement of the trial, to ensure that they can understand the relevance of the evidence as it unfolds, and place it into context. At present, there is a real risk, unless the jurors are particularly astute, of missing what is ultimately important, and of being deprived of a proper opportunity for assessing, as the evidence unfolds, whether it supports the prosecution case or not.

This would require a degree of cooperation by counsel in making a proper disclosure of the issues at the commencement of the trial. Although some counsel maintain the right to hold back their defence, in the belief that there is merit in trial by ambush, other counsel seem to have no problem in giving a defence opening.

The adoption of a practice of this kind, in addition to the preliminary observations which are conventionally given, had the support of Lord Justice Auld in the 2001 Review of the Criminal Courts of England and Wales, in the form of his suggested case and issues summary. It was also seen as helpful in the New Zealand Jury Research Project.

Lord Phillips of Worth Matravers has noted with approval the recommendations made in this Report, which would see the jury being given, at the start of the trial, a summary of the case and of the questions which they are to decide, supported by a written aide-m'moire as to the relevant law. If these recommendations were taken up, His Lordship has observed, it should no longer be necessary for the jury to be directed at the end of the evidence on the law, or given a detailed summary of the evidence. The summing-up, it was suggested, could be reduced to a reminder of the issues and a summary of the respective cases, accompanied by a series of written questions, the answers to which would lead to a verdict. 58

A variant of this approach involves giving to the jury, before the addresses commence, a written set of instructions setting out the law and the relevant issues. Although this requires a degree of cooperation by counsel in agreeing upon the document that is provided to the jury, it would help in focusing the addresses.

A somewhat more radical approach would be to abandon the traditional general verdict and to replace it with a system calling for special verdicts. 59 It would have the benefit of protecting the jury from the law and of confining its task to that of fact-finding, as the jurors' oath or affirmation requires, by requiring the jury to bring back verdicts, in the form of answers to a series of specific questions framed in terms of the elements of the offence charged, and of the ingredients of any relevant defences.

So framed, the answers would lead to a general verdict in a way that could spare the jury from working its way through a task that currently expects its members to understand and apply the law, as explained to them, to the evidence as they recall it.

Not only would this provide a greater focus to their deliberations, but it could reduce the risk of error, since the jury would have in front of it specific questions framed with an eye to the law and to the issues in the case which call for essentially factual conclusions.

There might still be a need for some general explanation of the law, perhaps at the beginning of the trial, as well as an identification of where the onus of proof lies, although that could be picked up in the framing of the questions.

Apart from focusing the jury, a process of this kind could also assist in sentencing, for example, where an alternative verdict of manslaughter was potentially available by more than one route.

Admittedly, there would be some added complexity in those jurisdictions which permit majority verdicts, as well as a degree of intrusion into the sanctity of jury deliberations. The latter, however, seems more theoretical than real, and the former should be capable of resolution without much difficulty by advice to the jurors, at the appropriate time (which currently differs between States), of their entitlement to answer the questions on which they have failed to achieve unanimity by a majority verdict.

In addition, it would have to be accepted that such a system could impinge on the notion of 'jury equity', which recognises the entitlement of juries to bring back a verdict of not guilty in the teeth of the evidence or of the judge's directions as a safeguard against unfairly aggressive prosecutions. Whether that might be, or not be, a good outcome is open for debate, although it was condemned by Lord Justice Auld in his Review. 60


There is a case for greater assistance to be given to juries by way of a supplement to the oral summing-up, in the form of:

  • evidence folders;
  • written legal directions, which have been permissible at common law as a supplement or aide moire, but which are now given a statutory standing in New South Wales. 61 Most desirably, they should be settled in advance with counsel;
  • chronologies;
  • statements of the relevant issues;
  • flow charts or decision trees;
  • a copy of the transcript of evidence;
  • a copy of the summing-up, so far as it deals with the law.

It may be noted that Justice Bleby in R v Dunn62emphasised the essential oral nature of summing up, and of the need for caution in relation to written directions. On the other hand, the use of such material more closely approximates the way in which modern jurors are accustomed to absorbing, retaining and using information. They tend to be unfamiliar with lengthy oral presentations, being more attuned to acquiring information through the internet and PowerPoint technology.

It is not to be overlooked in this respect that several studies reveal the limited extent to which information given orally is retained in memory, and the even more limited extent to which it is retained correctly. In this respect, juries are currently expected to absorb, retain and then apply the evidence without a transcript in a way which is not expected of judges when reaching a decision.


There are some notoriously complex areas of the criminal law which would benefit from legislative clarification or resolution by appellate courts. Examples include:

  • the problematic concept of 'malice', the application of which has been described as calling for 'a meticulous analysis and fine and impractical distinctions to be made by the jury (for which task such a body is quite ill-suited)'. 63 In New South Wales, this has recently been addressed by legislation substituting the mental elements of recklessness or intention for 'malice'as the relevant element for the offences concerned; 64
  • the concept of recklessness, which now has a wider role to play in New South Wales, and also a significant role in the Commonwealth Criminal Code, has its own problems since it has been recognised that, depending on the context, it can encompass several different mental states; 65
  • the potential complications which can arise in cases involving multiple accused, where the prosecution case is pursued on the basis of the several alternatives available, including liability as an accessory, joint enterprise, extended joint enterprise, and even conspiracy or felony murder. These complications are multiplied where the possibility of a conviction on some lesser alternative offence has to be left. Although guidance has been given by the High Court as to the principles to be applied, 66 it remains an exceedingly complex matter for jurors to understand, and one in which judges are prone to error. This is the subject of a separate current reference to the NSW Law Reform Commission, 67 calling for a review of the common law of complicity, and the arguments for and against codification. For that purpose, it will be necessary to review the recent report of the Law Commission (England and Wales), 68 as well as developments in other Australian and international jurisdictions. The significance of this reference, and the possibility it opens up for the adoption of a uniform legislative approach, is made all the more important by reason of the different approaches to extended joint enterprise which apply, dependent upon whether the case is to be determined according to the common law or under the Model Criminal Code. 69 In this respect, it may also be noted that two different tests currently operate in the codified criminal jurisdictions of Australia.
  • evidence of multiple acts in sexual offence cases, some of which are charged and others of which are uncharged, and where the evidence has been left to the jury variously:

    o as tendency or coincidence evidence in uniform Evidence Act States; 70

    o as similar fact evidence in other States; 71

    o as relationship or context evidence; 72

    o as guilty passion for (or sexual interest in) the complainant; 73

    o as evidence in rebuttal of good character when that is opened by the accused; 74

    o as evidence designed to explain the reason for delay in making a complaint; and

    o even as evidence advanced by an accused to suggest that the complainant has exploited his or her past conduct in generating a new and false complaint.

Some consideration was given in Tully v The Queen75 to the basis on which this kind of evidence should be treated. However, an authoritative determination of the issues which arise in relation to its admission and to the directions which need to be given may have to await the decisions in two pending cases. 76

Equally confusing for juries are the directions which need to be given that they are to deal with multiple counts separately, and in relation to the way in which a doubt concerning a complainant's evidence for one count may be used for other counts. 77

There are many other contexts where multiple and complex directions are required which call upon a great deal of mental dexterity, for example, in relation to conspiracy, provocation, substantial impairment by abnormality of mind (diminished responsibility), self-defence, consent in sexual assault cases, 78 and the like, which can require significant mental gymnastics as a result of the several elements involved.

No less challenging are the directions required for many offences arising under the Commonwealth Criminal Code, with its somewhat complex distinction between physical elements and mental elements, its classification of the circumstances in which there is no criminal responsibility, and its sometimes cumbersome framing of offences, for example, in the area of terrorism.

Equally difficult of comprehension are the directions given in relation to circumstantial proof, particularly in relation to the distinction between indispensable intermediate facts which need to be established beyond reasonable doubt ('links in a chain') and other circumstances which do not need to be proved to such standard, being part of the mosaic ('strands in a cable') available to the jury in deciding the ultimate question of guilt. 79

Some of the problems for trial judges in giving meaningful directions in these areas could be overcome by encouraging the legislature, when framing new offences or amending existing offences, to avoid using terms which are not in everyday use, or which call for extensive supplementary explanation as to their meaning or reach. For example, the concept of 'being knowingly concerned in'certain prohibited activities, without further explanation or guidance being given in the relevant Act, has been notoriously productive of error. The use of expressions such as 'suffers', which are not in common parlance, in the context of certain conduct associated with the possession, manufacture or handling of drugs, is problematic, as can be seen in the case of R v Sheen, 80 where it was held that the jury should have been instructed that it was not only necessary for the prosecution to show that the accused knowingly allowed the third party to carry out the relevant act, but additionally that he had the right or capacity to prevent it.

Similarly, there is a case for the legislature avoiding the framing of offences or defences employing expressions of indeterminate reference such as 'reasonable cause', leaving it to the courts to provide the context. 81


The problem with directions in jury trials is not new. Sixty years ago, Judge Learned Hand observed:

It is exceedingly doubtful whether a succession of abstract propositions of law, pronounced staccato, has any effect but to give [jurors] a dazed sense of being called upon to apply some esoteric mental processes beyond the scope of their daily experience. 82

To similar effect were the observations of the English Criminal Law Revision Committee's 1972 Report on Evidence:

The present law requires Judges to direct juries to achieve certain mental feats which some Judges think impossible for any lawyers to achieve, and it is no answer to criticisms of this kind to say, as is sometimes said, that there is no difficulty in directing the jury in the way which Courts have said that they should be directed. There may be no difficulty in saying the right words: the question is what the jury makes of them and nobody can be sure of that. 83

As Lord Mackay of Clashfern has observed:

to make it obligatory on trial judges to give directions in law which are so elaborate that the ordinary member of the jury will have great difficulty in following them, and even greater difficulty in retaining them in his own memory for the purpose of application in the jury room, is no service to the cause of justice. 84

Chief Justice Gleeson has also observed:

Directions to juries are made for the purpose of enabling them to perform their constitutional function, and to return a just verdict on the issues and evidence in the trial. They are not ritualistic incantations. They are not primarily intended to preserve the reputation or dignity of the trial judge by disarming counsel in a possible future appeal. They are meant to be fair, relevant and informative. In the Australian tradition, the criminal law is meant to be administered by juries. If it is so complex that it is impossible to explain to a lay person, then there is something wrong with the law. 85

The risk is that, if the jury members do not understand what clearly is intended to be an important part of the summing-up, they will tend to disregard the whole, and instead rely on their own values, assumptions and experiences to reach a verdict, or, alternatively, respond to their individual emotional reactions to the facts of the case.

Sufficient concern has been identified, through jury studies, through the rising incidence of appeals in criminal trials and orders for retrials, and through the increasing complexity of the criminal law as the legislature strives to promote law and order while preserving fundamental civil liberties, for a new approach to be taken in relation to the summing-up. The simplification of the directions that need to be given, and the development of a structure for a summing-up that will genuinely assist a modern jury in its task would go a long way in producing a more satisfactory trial process.

Significantly, it could assist to overcome the approach that attracted the following comment of Justice McClellan in Norris v The Queen: 86

I would further note that the present case is an example of the extent to which the summing up has developed to the point of almost taking over from all other aspects of a criminal trial. The evidence in the trial took 117 pages of transcript over a period of two days with addresses. The summing up took over a day and comprised 88 pages of transcript. There was so much information given to the jury by way of warnings, directions, comments and review of evidence and arguments that it is little wonder that they had not properly grasped the concept of beyond reasonable doubt by the time they retired to consider their verdicts. With respect, there was an unnecessarily detailed review of the evidence, including cross-examination, in what was a short trial with very narrow issues in a situation where the jury had heard all the evidence and addresses uninterrupted over a period of just two days before the summing up commenced. It is a good example of both the need for a trial judge to tailor the summing up to the realities of the trial and the need for a review by the legislature of the obligations of a trial judge in summing up a sexual assault case.

The NSWLRC has a significant amount of work to undertake in responding to its current reference. Apart from the engagement of the Plain English Foundation previously mentioned, it has also commissioned a survey of jurors which is being carried out with the assistance of the Bureau of Crime Statistics and Research (BOCSAR), whose purpose is to explore the extent to which the summing-up, in a selection of trials, was regarded by the jury as helpful or, alternatively, occasioned difficulty in its comprehension. Hopefully, it might throw some more light on the factors which led to the disturbing findings of a survey conducted by BOCSAR in which jurors in 25 sexual assault trials were asked post-verdict what the verdict had been. In only six trials did all the jurors give the correct answer.

Of some concern is the fact that several studies have revealed that, although jurors report that they had no difficulty in comprehending the directions given to them, further investigation in relation to specific directions has revealed serious misunderstandings. This is not a state of affairs that can be tolerated if jury trial is to continue to be the norm.


1. Alford v Magee (1952) 85 CLR 437 at 466.

2. The problems which juries have in understanding and applying highly technical instructions were recently examined by Professors Ogloff and Rose in their article 'The Comprehension of Judicial Instructions', in K D Williams and N Brewer (ed), Psychology and Law: An Empirical Perspective (Guildford Press, 2005).

3. For example, Lord Phillips of Worth Matravers, 'Constitutional Reform: one year on', the Judicial Studies Board Annual Lecture, Inner Temple (22 March 2007); Lord Phillips, 'Trusting the Jury', the Criminal Bar Association Kalisher Lecture (23 October 2007); Justice Eames, 'Tackling the complexity of criminal trial directions: what role for appellate courts?'(2007) 29 Australian Bar Review 161; Lord Justice Auld, Review of the Criminal Courts of England and Wales, Report (2001); and see Zoneff v The Queen (2000) 200 CLR 234, [66]-[67] where Kirby J noted some of the research which indicated that jurors were having difficulties in understanding and applying instructions.

4. Justice M Kirby, 'Why has the High Court become more involved in criminal appeals?'(2002) 23 Australian Bar Review 4.

5. Weiss v The Queen (2005) 224 CLR 300.

6. Weiss, [45]-[46]; and see Darkan v The Queen (2006) 227 CLR 373.

7. The undesirability of trial judges simply reading the relevant section of an Act or of giving the jury a copy of it was noted in Safwan v The Queen (1986) 8 NSWLR 97 and, more recently, in Pengilley v The Queen [2006] NSWCCA 163, [41].

8. For a review of the extent to which the current practices may not be employing the most relevant and helpful methods of communication with jurors, see J Horan, 'Communicating With Jurors in the Twenty-first Century'(2007) 29 Australian Bar Review 75.

9. Barca v The Queen (1975) 133 CLR 82; R v Bikic [2002] NSWCCA 227.

10. A concept which Hunt AJA observed in R v El-Hassan [2007] NSWCCA 148, [33] is expressed in 'decidedly non-jury friendly language'.

11. R v Jones (2006) 161 A Crim R 511.

12. See R v Chai (2002) 76 ALJR 628; Williams v The Queen (1990) 50 A Crim R 213, 226-227; and R v Meher [2004] NSWCCA 355.

13. Domican v The Queen (1992) 173 CLR 555; R v Smith [1984] 1 NSWLR 462; and R v Riscuta [2003] NSWCCA 6.

14. Longman v The Queen (1989) 168 CLR 79; Crampton v The Queen (2000) 206 CLR 161; Doggett v The Queen (2001) 208 CLR 343; Crofts v The Queen (1996) 186 CLR 427; R v Murray (1987) 11 NSWLR 12; and Tully v The Queen (2006) 81 ALJR 391.

15. Pollitt v The Queen (1992) 174 CLR 558.

16. Davies v Director of Public Prosecutions [1954] AC 378; Jenkins v The Queen (2004) 79 ALJR 252; and R v Kanaan (2005) 64 NSWLR 527.

17. R v Stewart (2001) 52 NSWLR 301.

18. Azzopardi v R (2001) 205 CLR 50; R v Wilson (2005) 62 NSWLR 346; and Mule v R (2005) 79 ALJR 1573 for those cases where the accused has participated in an ERISP pre-trial, but has not given evidence at the trial.

19. Evidence Act 1995 s 116 and s 165; and R v Clarke (1997) 97 A Crim R 414.

20. Criminal Procedure Act 1986 (NSW) s 306X.

21. Criminal Procedure Act 1986 (NSW) s 306ZI.

22. R v Stewart (2001) 52 NSWLR 301, [139].

23. Bromley v The Queen (1986) 161 CLR 315; and R v Stewart (2001) 52 NSWLR 301, [72]-[83] and [92]-[98].

24. See Relc v The Queen (2006) 167 A Crim R 484, [80].

25. R v Stewart (2001) 52 NSWLR 301, [117]-[119].

26. For example, Evidence Act 1995 s 165B; Criminal Procedure Act 1986 (NSW) s 294 and s 294AA, although Tully v The Queen (2006) 81 ALJR 391 makes it clear that the common law will continue to require a warning where there would otherwise be a perceptible risk of a miscarriage of justice.

27. For a recent analysis of the circumstances which will require a warning in relation to the evidence in this type of case, see Wade v The Queen (2006) 164 A Crim R 583; and KJR v The Queen [2007] NSWCCA 165.

28. Edwards v The Queen (1993) 178 CLR 193; Zoneff v The Queen (2000) 200 CLR 234; and Dhanhoa v The Queen (2003) 217 CLR 1.

29. CSR Ltd v Maddalena (2006) 80 ALJR 458; and Kamm v The Queen [2007] NSWCCA 201, [50].

30. Fox v Percy (2003) 214 CLR 118, [31].

31. R v Zorad (1990) 19 NSWLR 91; and Green v The Queen (1971) 126 CLR 28, 34.

32. See B v R (1992) 175 CLR 599; R v Meher [2004] NSWCCA 355; and R v Marr (1989) 90 CrAppR 154.

33. See the observation of Handley JA in R v Heron [2000] NSWCCA 312, [79]-[80].

34. [2006] NSWCCA 119; and see also RPS v The Queen (2000) 199 CLR 620, 637 and Azzopardi v R (2001) 205 CLR 50, 70.

35. For an example where error arose in this respect, see R v Rajakaruna(No 2) (2006) 15 VR 592.

36. Whether the decision in Yager v The Queen (1977) 139 CLR 28, which accepted the regularity of a directed verdict to convict, would withstand scrutiny today is questionable: see P Gillies and A Dahdal, 'Directions to Convict'(2007) 31(5) Criminal Law Journal 295.

37. Lord Phillips, 'Trusting the Jury', 14; and 'Constitutional Reform: one year on', 13; G Taylor, 'Judicial Reflections on the Defence Case in the Summing Up'(2005) 26 Australian Bar Review 70, 83-87; and D Wolchover, 'Should Judges Sum Up on the Facts?'(1989) Criminal Law Review 781, 787-788.

38. For example, Dawson v The Queen (1961) 106 CLR 1; Green v The Queen (1971) 126 CLR 28; Van Leeuwen v The Queen (1981) 55 ALJR 726; R v Reeves (1992) 29 NSWLR 109; and R v Southammavong [2003] NSWCCA 312.

39. See Norris v R [2007] NSWCCA 235 for a recent example of a case where the jury requested clarification of the legal definition of reasonable doubt, asking whether it means 'we need to be one-hundred per cent sure either way'. In R v McNamara [1998] QCA 405, Thomas J noted at [19] that the fact that the jury has asked for an explanation does not mean that a judge must answer it by stepping beyond the accepted limits. See W v R (2006) 16 TasR 1 for a detailed analysis of several different approaches which have been taken in an attempt to provide a more meaningful explanation of what is involved in proof beyond reasonable doubt.

40. Southammavong, [23].

41. R v Wanhalla [2007] 2 NZLR 573.

42. R v Wanhalla [2007] 2 NZLR 573, [49].

43. R v Bentley [2001] 1 CrAppR 307.

44. R v Lifchus [1997] 3 SCR 320.

45. M Zander, 'The Criminal Standard of Proof 'How Sure is Sure?'(2000) 150 New Law Journal 1517; J W Montgomery, 'The Criminal Standard of Proof'(1998) 148 New Law Journal 582. But see C N Heffer, 'The Language of Conviction and the Convictions of Certainty: Is 'Sure'an Impossible Standard of Proof?'(2007) 5(1) International Commentary on Evidence (Article 5).

46. (1971) 124 CLR 107 as considered more recently in Gillard v The Queen (2003) 219 CLR 1; Gilbert v The Queen (2000) 201 CLR 414; and Fingleton v The Queen (2005) 227 CLR 166. See also the recent decision of the House of Lords in R v Coutts (2006) 1 WLR 2154; but compare the approach taken in Scotland since R v Johnston (1998) SLT 788. See also R v Saad (2005) 156 A Crim R 533.

47. In R v Kanaan (2005) 64 NSWLR 527, the NSW Court of Criminal Appeal summarised the requirements of the law in this regard. See also Fingleton v The Queen.

48. R v Brown (1987) 32 A Crim R 162, 175.

49. The relevance of which, for example, in relation to warnings concerning identification evidence and lies was noted in Dhanhoa v R (2003) 217 CLR 1.

50. R v Zorad (1990) 19 NSWLR 91, 105; and Piazza v The Queen (1997) 94 A Crim R 459.

51. See the encouragement given to trial judges to follow this approach, for example, in Alvarez-Pizalla v State of Western Australia [2007] WASCA 224; and Norris v R [2007] NSWCCA 235.

52. Criminal Procedure Act 1986 (NSW) s 161.

53. For examples of its use, see R v Williams (1999) 104 A Crim R 260; and R v Davis [1999] NSWCCA 15.

54. R v Meher [2004] NSWCCA 355.

55. R v Zorad (1990) 19 NSWLR 91, 105.

56. R v Smart [1963] NSWR 706, 713; and see Thorne v The Queen [2007] NSWCCA 10.

57. Domican v The Queen (1992) 173 CLR 555; and see R v Davis [1999] NSWCCA 15, [24].

58. Lord Phillips, 'Trusting the Jury', 15.

59. Its use was discussed by Professor E Griew in an article 'Summing Up the Law'(1989) Criminal Law Review 768 that was cited by Auld LJ in his Review of the Criminal Law.

60. See also an article by J R Spencer 'Does Our Present Criminal Appeal System Make Sense?'(2006) Criminal Law Review 677.

61. Jury Act 1977 (NSW) s 55B.

62. (2006) 94 SASR 177.

63. R v Coleman (1990) 19 NSWLR 467, 472.

64. Crimes Amendment Act 2007 (NSW) Sch 1.

65. See Pengilley v The Queen [2006] NSWCCA 163; Banditt v The Queen (2005) 224 CLR 262; R v BBD [2007] 1 Qd R 478; and an article by Adam Webster 'Recklessness: Awareness, Indifference or Belief'(2007) 31 Criminal Law Journal 272.

66. McAuliffe v The Queen (1995) 183 CLR 108; Gillard v The Queen (2003) 219 CLR 1; Clayton v R (2006) 81 ALJR 439; R v Taufahema (2007) 228 CLR 232; and see also R v Powell [1999] 1 AC 1 for an analysis of some of the problems involved.

67. Complicity in Criminal Cases, July 2007.

68. Law Commission (England and Wales), Inchoate Liability for Assisting and Encouraging Crime, Report 300 (2006).

69. Criminal Code Act 1995 (Cth).

70. Gardiner v The Queen (2006) 162 A Crim R 233.

71. Pfennig v The Queen (1995) 182 CLR 461; and Phillips v The Queen (2006) 225 CLR 303.

72. KRM v R (2001) 206 CLR 221. See the editorial by Stephen Odgers QC, 'Relationship Evidence'(2007) 31 Criminal Law Journal 269 for a discussion of this vexed topic; see also Leonard v The Queen (2006) 67 NSWLR 545; and Qualtieri v The Queen [2006] NSWCCA 95.

73. R v Ball [1911] AC 47; and R v Beserick (1993) 30 NSWLR 510.

74. BRS v The Queen (1997) 191 CLR 275.

75. (2006) 81 ALJR 391.

76. SB v The Queen [2007] HCA Trans 253; and HML v The Queen [2007] HCA Trans 253, in which special leave was given by the High Court on 12 June 2007. See also Leonard v The Queen (2007) 67 NSWLR 545; and Qualtieri v The Queen [2006] NSWCCA 95.

77. See R v Markuleski (2001) 52 NSWLR 82; and KRM v R (2001) 206 CLR 221.

78. The law in this regard will be significantly altered when the Crimes Amendment (Consent - Sexual Assault Offences) Act 2007 commences and substitutes a s 61HA for the existing consent provisions.

79. Shepherd v The Queen (1990) 170 CLR 573.

80. (2007) 170 A Crim R 533.

81. See Taikato v The Queen (1996) 186 CLR 454; and Fingleton v The Queen (2005) 227 CLR 166, [108]-[109].

82. United States v Cohen 145 F 2d 82, 93 (2 nd Cir, 1944).

83. England and Wales, Criminal Law Revision Committee, Evidence (General), Report 11, Cmnd 4991 (1972).

84. R v Adomako [1995] 1 AC 171, 189.

85. Gleeson CJ, 'The Role of a Judge in a Criminal Trial', Law Asia Conference, Hong Kong (6 June 2007).

86. [2007] NSWCCA 235, [61].