Lawlink NSW: 4. Prejudice to criminal proceedings

4.1 In Discussion Paper 43 (“DP 43”), the Commission examined several aspects of the test for liability for sub judice contempt.¹ We put forward a number of proposals for legislative reform and invited submissions on these proposals. We now re-examine our proposals in light of the views expressed in consultation.

4.2 We note that this chapter focuses specifically on prejudice to criminal, as opposed to civil, proceedings. In contrast, Chapter 4 of the Discussion Paper was entitled, “Prejudicial Publications” and included a more general discussion about the test for liability for sub judice contempt. We have deferred more detailed discussion of prejudice to civil proceedings to Chapter 6 of this Report, and now look specifically at restrictions to publications relating to criminal proceedings. However, our recommendation for a reformulation of the “tendency” test, discussed in this chapter, is not limited to publications relating to criminal proceedings. In theory, it applies equally to restrict publications relating to civil proceedings, although for reasons discussed below and in Chapter 6 of this report, it will have less practical relevance to such publications.

4.3 Proposal 3 of DP 43 put forward a reformulation of the existing common law test for liability for sub judice contempt, except in so far as publications imposing pressure on parties are concerned. The proposal read as follows:

4.5 It should also be noted that a publication might constitute a contempt if it is found to have a tendency to impose improper pressure on a party to proceedings.² While this aspect of the sub judice rule generally has greater practical application in the context of publications concerning civil proceedings, it also applies to publications concerning criminal proceedings and, in the Commission’s view, should continue to do so. Proposal 3 of the Discussion Paper was not intended to affect the existing law concerning improper pressure on a party. In Chapter 6, we discuss as a separate issue the restrictions that should apply to publications that impose such pressure.

4.6 The current test for liability for sub judice contempt is generally formulated in terms of “tendency”.³ A publication must be shown to have a real and definite tendency, as a matter of practical reality, to prejudice or embarrass particular legal proceedings, in order to constitute a contempt.⁴ The prosecution bears the burden of proving the necessary tendency, beyond a reasonable doubt.

4.7 Liability for sub judice contempt depends on the potential effect of a publication on legal proceedings, rather than on proof of any actual effect the publication may have had. The court assesses the tendency of a publication to cause prejudice by examining the nature of the publication and the circumstances surrounding it, as they appeared at the time of publication. It does not have regard to later events that actually occurred following the publication. For example, in assessing whether or not a publication had a tendency to prejudice proceedings, a court would not take into account the fact that the proceedings were not in fact affected by the publication because the accused died before the trial or elected at the trial to plead guilty.⁵

4.8 The tendency test has been widely criticised for being imprecise and unclear, as well as too broad.⁶ The test is said to be imprecise in so far as “tendency” is a vague and general notion on which to base criminal liability, since it is often impossible to know whether a particular statement, if published, will be found to have a tendency to prejudice proceedings. Given that sub judice contempt is generally considered to be a criminal offence, the limits of its liability should be defined with sufficient precision to allow members of the public to know, with a satisfactory degree of certainty, the conduct that will expose them to criminal liability.

4.9 It may also be argued that the test for liability is too broad in so far as it sets too low a threshold for the prosecution to prove contempt, by requiring no more than a “tendency” to prejudice. Publications may be prohibited which have a tendency to prejudice but which do not pose any serious risk to the administration of justice. The test may then be considered to tilt the balance further than is necessary to protect the fairness of legal proceedings at the expense of protecting freedom of discussion. There is also the risk that the media will be overly cautious in publishing material because they find the tendency test unclear. This, when combined with the more fundamental objection that the tendency test sets too low a threshold, may result in an unnecessary intrusion on freedom of discussion.

4.10 Proposal 3 reformulated the test for liability in terms of a “substantial risk of prejudice”, rather than a tendency to prejudice proceedings. In fact, several Australian judges have in the past applied the “substantial risk” formulation in determining cases of sub judice contempt.⁷ Chief Justice Mason (as he then was) favoured the “substantial risk” formulation because, in his view, it balanced more appropriately the competing interests in free speech and the administration of justice. He considered the “tendency” formulation to be vague and uncertain, and perhaps to be seen to place too much weight on the protection of the administration of justice over freedom of speech.⁸ Several common law jurisdictions have adopted some form of the “substantial risk” test in favour of the tendency test.⁹ It has also been the preferred formulation of three law reform bodies who have reviewed the law of sub judice contempt, though the formulations proposed have varied in their terms from a substantial risk of prejudice, to a substantial risk of serious prejudice.¹⁰

4.11 It is open to question whether there is in fact any real difference in meaning between “substantial risk” and “tendency”.¹¹ The courts have at times used both phrases interchangeably, and Chief Justice Mason considered that there might be no substantive difference between the two.¹² The case law does not provide a particularly precise definition of “tendency” in this context. The courts have noted that the degree of likelihood required by the word is not one of probability, but rather a “real possibility” of interference.¹³ The High Court has said that the degree of possibility required must be more than a remote possibility that justice will be interfered with.¹⁴ The formulation requires that the tendency be real and definite, and must exist as a “matter of practical reality”. It is also worth noting that the “tendency” formulation makes no reference to any requirement for “serious” prejudice, but simply requires a tendency to prejudice, or embarrass, proceedings.

4.12 The word “substantial” in “substantial risk” would seem to require a higher degree of likelihood than one that is simply more than remote. The drafters of the Contempt of Court Act 1981 in the United Kingdom, in adopting the “substantial risk” test as the legislative formulation for liability for sub judice contempt, certainly appeared to intend the word “substantial” to mean something which is serious, considerable, and real, imposing a high threshold on liability in order to restrict its intrusion on freedom of expression.¹⁵ However, the English courts have since given a broader interpretation to “substantial”, stating that it does not mean “weighty”, but means “not insubstantial” or “not minimal”.¹⁶ On other occasions, it has been held that “substantial” simply excludes a risk that is only remote.¹⁷

4.13 Proposal 3 reflects the Commission’s tentative view, stated in DP 43,¹⁸ that the “substantial risk” test was preferable to the tendency test for determining liability for sub judice contempt. We considered that the “substantial risk” test was somewhat more precise, and therefore allowed the media to determine with a greater degree of certainty which publications would expose them to prosecution. We took the view that the “substantial risk” test required a high degree of risk in order to restrict the publication of material. This was desirable in order to ensure that the limits imposed on freedom of discussion were only as much as were truly necessary to protect the fairness of legal proceedings. While the degree of possibility that is required by the “tendency” test is unclear, the Commission certainly intended the “substantial risk” test to require something more than a risk that is not remote. While the English courts may have interpreted the “substantial risk” test as imposing a lower threshold for liability than was originally intended, the Commission was not convinced that this would be the experience in New South Wales. If all that is required by the tendency test is a risk that is more than remote, then Proposal 3 can be said, at least, to raise the bar for the prosecution to meet in proving liability for sub judice contempt.

4.14 It is also worth noting that Proposal 3 requires a substantial risk of prejudice, rather than a substantial risk of serious prejudice. The Commission considered both formulations, but tentatively concluded that the first was preferable to the second. We took the view that prejudice to the fairness of legal proceedings is, by its nature, serious, and in the context of a fair trial, there are not degrees of prejudice. We could not imagine a case of “trivial” prejudice, and concluded that it was superfluous to add “serious” to the proposed formulation.

4.15 Most people in consultation supported Proposal 3, although with suggestions for some amendments. Eight groups and individuals expressed clear agreement,¹⁹ while two were more hesitant in their support,²⁰ though tending to lean in favour of the proposal rather than retaining the common law. Four were opposed to the proposal, on the basis that it was preferable to retain the common law.²¹

4.16 There was some difference of opinion about whether the term “substantial risk” did in fact set a higher standard for the prosecution to meet to prove liability for contempt than the “tendency” test in the existing common law formulation. In supporting the proposal, the Australian Press Council considered that it raised the standard from that set by the existing formulation. The government lawyers who met with the Commission, in expressing their tentative support for Proposal 3, also considered that it set a higher standard, and noted that the Commission needed to make clear whether or not this was our intention. Similarly, representatives of the New South Wales Bar Association, in opposing Proposal 3, took the view that it lifted the bar for the prosecution to meet. On the other hand, some groups and individuals opposed Proposal 3 on the basis that there was no difference between the “substantial risk” and “tendency” tests, and that it was therefore preferable to retain the existing common law formulation for the sake of certainty and uniformity with other Australian states.²²

4.17 Those who supported Proposal 3 generally considered that it should be modified in some way. Three organisations took the view that the test should require a substantial risk of serious prejudice, rather than simply prejudice.²³ The Victorian Bar Council asserted that the adjective “serious” emphasises the need for considerable prejudice, without which there may in practice be more focus placed on the assessment of risk required by the phrase “substantial risk”, at the expense of proper consideration of the degree of prejudice also involved. The Director of Public Prosecutions supported Proposal 3 on the provision that, instead of a substantial risk that “the fairness of the proceedings would be prejudiced”, there be a substantial risk that the fairness of the proceedings could be prejudiced. The ABC Legal Services considered that clause (a) of the proposed formulation be changed from “members, or potential members, of a jury … could (i) encounter the publication; and (ii) recall the contents of the publication at the material time” to “members, or potential members, of a jury … would” etc. Several of the government lawyers preferred the phrase “real risk” to “substantial risk”. The Australian Press Council submitted that clause (b) of the proposed formulation should instead be made a third element of clause (a), as an additional element to be proven of the effect on jurors or potential jurors. The prosecution would then need to prove a substantial risk that members of a jury could “(i) encounter the publication; (ii) recall the contents of the publication at the material time; and (iii) be influenced by the publication in a way that seriously prejudiced the fairness of the proceedings”.²⁴ The Solicitor General considered that it might be useful to add to clause (b) of Proposal 3 words such as “notwithstanding any directions that might be given by the trial judge in relation to the publication”. However, several other government lawyers took the view that this qualification was already implicit in the test for liability.

4.19 To a large extent, Recommendation 2 reproduces the wording of Proposal 3 in DP 43. There are a couple of minor modifications that aim to make the recommendation clearer, rather than effect any substantive changes. The first of these is reference in the recommendation to the “publication of matter” and “publishing the matter”. These seek to overcome any ambiguity that may have arisen from the Commission’s use of the word “publication” in DP 43, which at times could be taken to refer to the published material the subject of the contempt charge, and at other times the act of publishing that material. We have also changed the wording in clause (a)(ii) from “the contents” to “the content”.

4.20 There are several substantive changes appearing in the recommendation, and other matters worth noting.

4.21 First, Recommendation 2 does not exclude from the scope of liability for sub judice contempt the publication of matter that may influence civil juries in defamation proceedings. This represents a reversal of policy from that underlying Proposal 3 in the Discussion Paper, and is examined in detail in Chapter 6 of this Report.

4.22 Secondly, the recommendation refers to the substantial risk that participants in the proceedings will “become aware of” the contents of published material. Proposal 3 of DP 43 referred to the risk that participants would “encounter” the material. The Commission prefers the phrase “become aware of” because it more clearly encompasses both a direct and indirect encounter with the contemptuous material. Jurors or witnesses may become aware of the material by reading, watching or listening to it themselves, or may be told about it by, for example, a friend, family member, or another juror. Our recommended test for liability aims to include both situations within its scope.

4.23 Thirdly, Recommendation 2 imposes liability for a substantial risk that participants “will” become aware of the material, “will” recall the contents of the material, and, consequently, that the fairness of the proceedings “will” be prejudiced. Use of the word “will” creates a higher threshold of liability for the prosecution to meet than was originally put forward in Proposal 3 of DP 43. We have come to conclusion that a high degree of probability is required in order to justify curtailment of freedom of discussion, and that the wording of Proposal 3 set the threshold for liability too low.

4.24 Fourthly, the Commission rejected the suggestion that the test should require a substantial risk of serious prejudice. After consideration, the Commission does not agree that the word “serious” should be added to the formulation. The threshold for liability would be too high if the prosecution were required to prove a substantial risk of serious prejudice, and it would be almost impossible to prove a substantial risk of serious prejudice beyond a reasonable doubt. We are satisfied that, in order to amount to prejudice, the potential interference with legal proceedings must be more than merely trivial interference. We note that the current “tendency” test does not refer to serious prejudice, and therefore our recommended reformulation does not lower the threshold for liability in any way.

4.25 Fifthly, we considered the suggestion of the Australian Press Council that clause (b) of the reformulation be made a third subclause of clause (a), so that the test would require a substantial risk, first, that the participants would encounter the material, secondly, would recall the content of the material, and thirdly, that the fairness of the proceedings could be prejudiced. We have not adopted this suggestion, because we do not consider that it really adds anything to the formulation, and is more a matter of form than of substance.

4.26 Lastly, it was suggested that there be a separate rule dealing with the publication of photographs of the accused. This suggestion essentially stemmed from a concern that the recommended reformulation would not be interpreted widely enough to include a substantial risk of subconscious recall. Clearly, the sub judice rule should prohibit the publication of photographs where there is a substantial risk that, for example, a witness will subconsciously recall a photograph from a newspaper or on television when identifying the accused. We are satisfied, however, that reference in the recommended reformulation to a substantial risk of recall is wide enough to cover both conscious and subconscious recall, and therefore there is no need for a separate rule dealing with the publication of photographs.

4.27 Proposal 3 in DP 43 required the prosecution to prove both that there was a substantial risk that jurors, or a witness, could encounter the publication, and that they could recall the contents of the publication at the material time. These two elements are expressed as separate requirements for the prosecution to prove in addition to proving the risk of influence. The current test for liability does not separate out these three elements. The court may give consideration to the likelihood of the publication coming to the attention of participants in the proceedings, as a factor that affects its tendency to prejudice those proceedings.²⁵ The courts appear to place greater weight on this factor, as a possible reason for finding that the requisite tendency has not been proved, in respect of publications that occur before a trial has commenced.²⁶ They seem more reluctant to attach much weight to this matter as a factor reducing the requisite tendency in respect of publications occurring after a trial has commenced.²⁷ The findings of Managing Prejudicial Publicity, a Justice Research Centre Report on juries (“JRC Report”) bears this out in so far as it suggests that recall of pre-trial publicity is piecemeal, whereas recall of in-trial publicity occurred for at least one juror in every case that was investigated.²⁸

4.28 In DP 43, the Commission considered that it might be useful to include in the basic test for liability separate requirements to consider the likelihood of a juror’s or witness’ contact with the publication in question, and the risk that they would then recall that publication. We made it clear that Proposal 3 did not require proof of actual contact and recall, but simply proof of a “substantial risk” of contact and recall. In this way, we hoped to direct the courts’ attention to giving express consideration to these factors in determining liability. While they may now be generally subsumed in the basic formulation for liability, there is the danger that the courts may blur these two notions if they are not explicitly spelled out as separate elements of liability.

4.29 Only two submissions gave express consideration to the Commission’s proposed approach in respect of the risk of contact and the risk of recall. The Law Society agreed with the Commission.²⁹ Mr Norris, of the Crown Solicitor’s Office, saw the formulation of these factors as specific considerations as unexceptionable, though doubted that they would in fact bring about any change in the way that courts deal with contempts. He made particular reference to the difficulties that might arise in trying to prove a substantial risk of contact with Internet publications.³⁰

4.30 We are satisfied that the test for liability should be reformulated to require, as separate elements, a substantial risk of encounter and of recall. We do not consider that this reformulation makes any substantive changes to the common law. It does, however, focus the court’s attention on these elements as two separate elements to assess.

4.31 In addition to reformulating the basic test for liability in terms of “substantial risk”, Proposal 3 made it clear that published material may constitute a contempt on the basis of its potential effect on a witness, as opposed to or as well as on members of a jury. This reflects the present position at common law, which assumes that witnesses may be susceptible to influence by media publicity. There is little empirical or psychological evidence to test the validity of this assumption. The courts instead appear to draw on their own experience and assumptions about human behaviour to find contempt in cases where it is feared that publicity may influence a witness either by deterring him or her from coming forward to give evidence, or by contaminating the evidence that a witness gives.³¹

4.32 In DP 43, the Commission examined the common law’s assumption about the potentially prejudicial effect of publicity on witnesses.³² We questioned whether liability for sub judice contempt should continue to be imposed on the basis of possible influence on a witness, or whether the risk to the administration of justice in such a case did not justify retaining influence on a witness as a ground for restricting the publication of information. Previous reviews of the law of sub judice contempt have considered this issue.³³ In particular, the Australian Law Reform Commission took the view that liability for contempt should not arise from the possibility of influence on a witness, with the exception that, in criminal proceedings, the law should continue to prohibit the publication of a photograph, sketch or description of the physical attributes of a person in circumstances from which it could reasonably be inferred that the person was charged with or suspected of having committed an offence.³⁴

4.33 We did refer to some general psychological studies about the malleability of memory that suggest the potential for memory to be embellished or even transformed by new information received after an event. We considered these to offer some support to restricting the publication of material that may have an effect on a witness’s recollection of events.

4.34 We noted that, while contempt can still in theory arise in this context, the common law appears to have become increasingly reluctant to restrict the publication of information on the basis that it may influence a witness in civil proceedings. The courts seem now generally to place greater faith in the honesty of witnesses and the power of cross-examination to expose prejudice and inconsistencies.³⁵

4.35 Contrary to the recommendations of the Australian Law Reform Commission, we tentatively concluded in DP 43 that there was sufficient reason for concern that a witness may be influenced by media publicity to justify retaining this ground of liability. We invited submissions on our tentative conclusion.

4.36 Only three submissions discussed the issue of possible influence on witnesses.³⁶ All three agreed that the law should continue to restrict the publication of information in this context. One submission suggested that Proposal 3 might not be appropriately formulated to deal with some forms of possible influence on witnesses. Proposal 3 requires the prosecution to prove (among other things) a substantial risk that a witness could encounter the publication in question and recall its contents at the material time. It was submitted that a publication of, for example, a photograph may contaminate a witness’ testimony whether or not the witness later has any specific recollection of the publication, and that, in fact, there may be a greater risk of prejudice to a fair trial where the witness has no separate recollection of the publication and is unconsciously influenced by it. It was submitted that there should instead be a clear, general rule prohibiting or restricting the publication of film or photographs of a person charged with an offence in any circumstances, to avoid the risk of influence, or suggestion of influence, on a witness’ testimony.³⁷

4.37 The Commission concludes that the sub judice rule should continue to prohibit the publication of material on the basis of potential influence on a witness. Recommendation 2 reflects this. Those few submissions that addressed the issue supported our conclusion, and it could be argued from the general lack of response in submissions to this issue that it is considered not to be a contentious area. As for the suggestion regarding a separate legislative provision to deal with (subconscious) recollection through photographs, we do not consider this to be necessary, for the reasons set out above.

4.38 Recommendation 2 does not include within the scope of liability for sub judice contempt the publication of material that may influence judicial officers. This reflects the Commission’s view, put forward in DP 43,³⁸ that the possibility of influence of a judicial officer should not generally form a basis for prohibiting the publication of material under the sub judice rule. The exception to this is the possibility of influence of a judicial officer following a guilty verdict or plea in a criminal case before sentence is handed down. We discuss our reasons for a general position regarding the possibility of influence of judicial officers, and our more specific views regarding influence at the sentencing stage, in Chapter 7 of this report.

4.39 Proposal 3 includes within the scope of liability for sub judice contempt publications that may influence jurors or witnesses in civil proceedings, except for jurors empanelled under s 7A of the Defamation Act 1974 (NSW). In Chapter 6, which deals generally with the application of sub judice restrictions in relation to civil proceedings, the Commission examines the submissions received on this issue.

4.40 In DP 43, the Commission proposed that legislation set out a list of the types of statements that may give rise to liability for sub judice contempt.³⁹ Proposal 4 puts forward an illustrative list of statements that, if published, may constitute a contempt. The Commission did not intend for this list to provide the sole criterion for liability. The prosecution would need to satisfy the basic test for liability in Proposal 3 in order to prove contempt. The list of statements in Proposal 4 would provide additional guidance to the media on the types of statements that are typically considered to be prejudicial. The list would be illustrative only. That is, a statement could fall outside the categories in Proposal 4 and still constitute a contempt if it met the requirements of the basic test for liability in Proposal 3.

4.41 We considered the main purpose of our proposed list to be educative, and to provide the media with a quick reference point for examples of the types of statements that they should avoid publishing. We discussed whether it would be preferable for the list to be exhaustive. That is, a statement would only constitute a contempt if it fell within one of the categories listed (and also satisfied the basic test for liability in Proposal 3). The main advantage of an exhaustive list of statements is that it provides greater certainty and clarity for the media, since they would know that a statement falling outside one of the listed categories would not constitute a contempt. We took the view that such an approach would be too inflexible, as it might be difficult to predict and enshrine in legislation every type of statement that should properly be considered to prejudice a fair trial. There may be statements that fall outside the listed categories that are later found to be prejudicial. We also considered whether it would be preferable for Proposal 4 to provide the sole basis for liability. That is, a publication would only constitute a contempt if it fell within one of the categories listed in Proposal 4, with no additional requirement to satisfy the test set out in Proposal 3. This approach has the advantage of providing the media with greater certainty about when a publication will give rise to liability. However, we rejected it on the basis that it placed an unjustifiable restriction on freedom of discussion: a statement may fall within one of the listed categories but not pose a substantial risk of prejudice to proceedings in the circumstances of a particular case. Previous reviews have differed in their conclusions about the appropriate approach to take.⁴⁰

4.42 We included in Proposal 4 those categories of statements that typically give rise to liability at common law. We have since analysed all prosecutions for sub judice contempt in Australia since 1980 to determine the number of cases, if any, which do not fall within one of the categories included in Proposal 4. A table with the result of our analysis is attached as Appendix E to this report. Almost all of the cases did fall within one of the proposed categories. It is also worth noting that some of the proposed categories, though not all of them, were also listed in the JRC Report as the types of information that jurors, judges, and counsel consider as possibly influencing juries’ perceptions of the trial.⁴¹ Other categories nominated in the JRC Report that were not included within our proposed list are: sympathetic and unsympathetic depiction of the victim, and generic publicity, that is, publications that deal with broader, general social issues that have relevance to the particular trial in question.

4.43 There was an equal division of opinion about the desirability or undesirability of including Proposal 4 in a legislative formulation of the test for liability for sub judice contempt. Four groups and individuals supported the proposal,⁴² and four opposed it.⁴³ Those who opposed it questioned the utility of including a list in legislation of categories of statements that are already well established at common law and generally known as potentially contemptuous. It was considered that a list of categories of statements might focus attention on whether or not a statement fits within one or more of the categories, rather than on determining the more fundamental question of the likelihood of prejudice to legal proceedings.

4.44 Both those who supported Proposal 4 and those who opposed it made suggestions for changes to the proposal, in the event that the Commission were to recommend its inclusion in a legislative formulation. It was submitted that the final category listed in the proposal, that is “a photograph, sketch or other likeness of the accused, or a physical description of the accused”, should be considered potentially prejudicial only in cases where the identity of the accused is or might be in issue.⁴⁴ One submission questioned the need to include “previously acquitted” in the first category of potentially prejudicial statements.⁴⁵ Another submission considered that the third and fourth categories of statements were too generally phrased, and should be redrafted in a tighter form.⁴⁶ It was argued that, in their present form, they do not give rise to serious prejudice, especially in light of the findings of the JRC Report. Similarly, another submission argued that several phrases included in the proposal were too imprecise, and needed to be clarified in order to achieve the level of certainty that was sought by including a list of statements in a legislative formulation for liability.⁴⁷ These phrases were: “or from which it could be reasonably inferred”, in the first, second and third categories,⁴⁸ “could be reasonably regarded”, in the fourth category, and “illustrative”, as a description of the list as an illustrative, as opposed to exhaustive, list. It was argued in consultation that the third and fourth categories of statements should be deleted altogether from the proposal.⁴⁹ One submission pointed out that the proposed list did not provide an exception for matters referred to in court proceedings.⁵⁰ One group argued that the list should be exhaustive,⁵¹ while another submission expressed the opposing view that it should be illustrative.⁵²

4.45 We have reached the conclusion that legislation should not prescribe categories of prejudicial material. We appreciate the educative function that such a list would serve for the media. However, on balance, we consider that such a list would be too inflexible, and would complicate the formulation for liability unnecessarily. As was made clear in consultation, the types of publicity that will typically give rise to liability for contempt are well established and generally known. It could therefore be argued that it is not really necessary to enshrine a list of this kind in legislation. As we discussed above, we consider that it would be too inflexible to prescribe an exhaustive list of categories, and to include an illustrative list may simply complicate and confuse the issue of liability more than it would clarify it.

4.46 A trial judge has a discretion to dismiss a jury in a criminal trial if he or she considers that a publication concerning the trial is so prejudicial as to make the trial unfair.⁵³ The jury should be discharged if, in all the circumstances, this is necessary in the interests of ensuring a fair trial.⁵⁴ Once a jury is discharged, the trial must stop, or be aborted, and usually a new trial, with a new jury, will be fixed to commence on some later date.

4.47 In Proposal 5 of DP 43, the Commission suggested that the fact that a trial judge has decided to dismiss, or not to dismiss, a jury in a criminal trial following the publication of material about that trial should be admissible in the contempt proceedings as relevant to the issue of liability for sub judice contempt. The proposal made it clear that, while this should be admissible evidence, it should not be determinative of the question of whether or not a contempt has occurred.

4.48 It is unclear whether, at common law, a trial judge’s decision to dismiss or not to dismiss a jury is admissible evidence in the related contempt proceedings on the question of liability for contempt, though it seems settled that it is relevant to the question of penalty once a contempt has been proven.⁵⁵ Comments in a number of Australian cases suggest that it is relevant to the question of liability, and that, at the least, it is relevant to determining that the tendency of the publication to interfere with the course of justice was not fanciful.⁵⁶ While such evidence may be relevant, the court hearing the contempt case certainly does not appear to consider itself bound by the earlier decision.⁵⁷

4.49 In one recent case in New South Wales, it was held that evidence that a trial has been aborted as a result of a publication was not relevant and was inadmissible on the issue of whether a contempt had been committed.⁵⁸ This ruling may appear to conflict with earlier cases referred to above that found such evidence to be admissible. However, one commentator has suggested that this more recent ruling should not be interpreted as a statement of a general rule, but rather a conclusion based on the particular facts of the case.⁵⁹ At the least, it may be argued that the law is uncertain on this issue, and needs to be clarified. Similarly, in other jurisdictions, there are comments in cases that suggest on the one hand that such evidence is relevant to, though not determinative of, liability,⁶⁰ and on the other hand, that there are many reasons why a publication is found to be potentially so prejudicial as to amount to a contempt while the trial to which it relates is allowed to continue.⁶¹

4.50 Clearly, the question whether to discharge a jury is different from the question whether a publication amounts to a contempt. The court must apply different tests in determining each question. The trial judge, in deciding whether to dismiss the jury, must consider the publication in light of the nature and circumstances of the trial. A publication that, on its own, may not be regarded as creating a high risk of prejudice to a trial, in combination with other factors, might tilt the balance in favour of dismissing the jury. The court hearing the contempt case will not necessarily consider those same surrounding circumstances when deciding whether the publication amounts to a contempt. In DP 43, the Commission acknowledged that different issues apply to the decision faced by the trial judge and the court hearing the contempt case. However, we took the tentative view that the court hearing the contempt case should be able to consider any finding by the trial judge of the effect of the publication on the trial as part of the actual circumstances surrounding the publication. The trial judge’s finding should be relevant, but would not of itself prove contempt; the prosecution would still need to prove, beyond a reasonable doubt, that the publication created a substantial risk of prejudice in order to succeed in the contempt proceedings.

4.51 The majority of those who expressed a view on Proposal 5 did not agree with it. Two supported it,⁶² while five opposed it.⁶³ Those who opposed it had various reasons for doing so. One submission argued that judges currently tend to dismiss juries even when there is a remote likelihood of a seriously prejudicial influence, and that consequently, evidence of the decision has little probative value and should not be admissible in any related contempt proceedings.⁶⁴ Others argued that different issues are involved in deciding whether to dismiss a jury and whether contempt has been committed, and questioned how, in that case, the trial judge’s decision could be a relevant consideration for the judge in the contempt proceedings.⁶⁵ It was pointed out that a trial judge might decide to dismiss or not dismiss a jury for a number of reasons other than the effect of the publication. The trial judge and the judge hearing the contempt case will apply different tests and different standards of proof, and only one will hear submissions from the alleged contemnor. Yet if evidence of the trial judge’s decision is admitted in the contempt proceedings, there is a risk that it will effectively determine the outcome of the contempt matter. It was also considered undesirable to allow a review in subsequent proceedings of a trial judge’s reasons for dismissing or not dismissing a jury.⁶⁶

4.53 The Commission has formulated Recommendation 3 as a further exception to be inserted in s 129 of the Evidence Act 1995 (NSW), which is a general provision excluding evidence of judicial reasons. If this recommendation is adopted in legislation, it will have implications for uniformity of evidence laws of the Commonwealth and the state. While we acknowledge this concern, it is not something that can affect our decisions regarding reform to the law of sub judice contempt in New South Wales.

4.54 Recommendation 3 also requires considerations of unfair prejudice to be weighed in the decision whether or not to admit evidence of a trial judge’s decision, according to s 135(a) of the Evidence Act 1995 (NSW). We have specifically provided that, just because it is considered inappropriate to call the trial judge to give evidence in the contempt proceedings, that cannot be used as a basis for excluding evidence of his or her decision.

4.55 There is some suggestion that, at common law, pre-existing publicity may act to lessen the tendency of a current publication to prejudice proceedings.⁶⁸ It has been held that, where material on the same topic has been published previously, through another source, the current publication may be less likely to prejudice proceedings because its prejudicial information has already been made known to the public.⁶⁹ However, it has also been held that, as a matter of public policy, pre-existing publicity can only be relied on as a factor lessening the prejudicial effect of a current publication if the previous publication did not itself constitute a contempt.⁷⁰

4.56 In DP 43, the Commission was concerned about the implications of the common law’s approach to the relevance of pre-existing publicity. As a matter of policy, we considered that a publisher should not be able to avoid liability simply because of pre-existing publicity on the same topic, and that the law should not withdraw its protection from accused persons who, for whatever reason, attract a large amount of media publicity. We also questioned the logic behind the assumption that the prejudicial effect of a publication was lessened by the existence of previous publicity. We suggested that, on the contrary, the existence of previous publicity might serve to reconfirm, if not heighten, in the public’s mind, prejudice and suppositions that would otherwise have been forgotten.⁷¹ Our views reflected the conclusions reached by the Australian Law Reform Commission on this issue.⁷²

4.57 Proposal 6 sought to clarify the law by suggesting a legislative provision to make it clear that the existence of previous publicity should not be considered as a factor lessening the risk of a current publication to cause prejudice. It is worth noting that the findings of the JRC Report support the view that pre-existing publicity may actually heighten the tendency of a current publication to cause prejudice, rather than lessen it. There was limited empirical evidence to suggest that pre-existing publicity may increase the likelihood of juror recall of a pre-trial prejudicial publication.⁷³

4.58 Five groups expressed views on Proposal 6. Three opposed it,⁷⁴ and two supported it.⁷⁵ Those who opposed it argued that it was a step backwards from the current common law position, and that it was unfair to expect a subsequent publisher to take full blame for prejudice partly arising from previous publications. One submission, which disagreed with the proposal, did concede that it might be reasonable to hold a publisher responsible if it republishes material close to the time of the commencement of the trial to which the material relates.⁷⁶ Another submission, while supporting Proposal 6, suggested that it could be worded more precisely.⁷⁷ In its current form, it was argued that the proposal did not specifically remove the assumption of the relevance of background publicity to the liability of a specific publication. It was submitted that the proposal would better achieve its aims if it included an express requirement that the risk of prejudice presented by a publication was not to be taken to be reduced solely on the ground that a previous publication, containing similar contents, had already been made, or, alternatively, that previous publicity must be disregarded in making an assessment of the risk of prejudice. It was noted, however, that this alternative formulation would foreclose a finding that the risk of prejudice was greater than it would otherwise be because of extensive pre-existing publicity.

4.60 In DP 43, the Commission raised a couple of other issues relating to the test for liability without formulating any specific proposals for reform. These were, first, the relevance of remedial measures to liability for contempt and, secondly, the admissibility and utility of expert evidence to prove tendency or substantial risk. These issues were not widely discussed in consultation. However, for the sake of completeness, it is worth referring to them again here.

4.61 In DP 43,⁷⁸ the Commission considered whether the availability of remedial measures should be relevant to the question of whether a publication has a tendency to prejudice proceedings. By “remedial measures”, we meant the measures available to minimise the possible prejudicial effect of media publicity on legal proceedings, specifically, the power of the court to order an adjournment or a change of venue, or to discharge the jury, the power of the prosecutor or defence to challenge the empanelment of jurors for cause, and directions by the trial judge to the jury to disregard publicity about the case.

4.62 The Commission discussed whether the media should be able to rely on the availability of any or all of these remedial measures to excuse them from liability for sub judice contempt, on the basis that these measures lessen the tendency, or substantial risk, of a publication to prejudice proceedings. We noted that Australian courts do not generally recognise the availability of remedial measures as relevant to the question of liability, except for possibly the effectiveness of judicial warnings in minimising the potential prejudice of media publicity.⁷⁹ The Commission took the view that the courts should not be required to give greater weight than they already do to the availability of remedial measures in determining liability for sub judice contempt. To do so would place greater reliance on remedial measures as a means of overcoming the possibility of prejudice from media publicity, which would involve greater expense to the State and the accused, and greater delays in finalising criminal trials. It may also cause inconvenience, emotional upset, and hardship to the participants in the trial.

4.63 The Commission’s view was undisputed in consultation. Only one submission expressly addressed this issue, and agreed with the Commission’s approach.⁸⁰ Consequently, we see no reason to change our position.

4.64 The Commission noted in DP 43⁸¹ that the courts now appear more inclined to admit and rely on expert evidence as relevant to the question of whether a publication has the tendency, or risk, required to prove liability for contempt.⁸² However, while such evidence may be more readily admitted, its utility in proving or disproving tendency or substantial risk may often be limited. At least in a case where expert opinion is based on a survey of mock jurors, the impossibility of replicating “real-life” trial conditions may detract from the survey results as a reliable indicator of what might happen at a trial.⁸³

4.65 The Commission saw no reason in principle why expert opinion should not be admissible on the issue of the substantial risk of a publication to cause prejudice. We considered that the provisions of the Evidence Act 1995 (NSW) relating to the admissibility of expert evidence⁸⁴ were sufficiently broad to admit expert evidence on this issue, and, consequently, there was no need to propose legislative reform.

4.66 Only one submission addressed the issue of expert evidence, and agreed with our approach.⁸⁵ We therefore do not consider it necessary to recommend any changes to the existing law regarding the admissibility of expert evidence. We do acknowledge the possibility, alluded to in the submission, that greater reliance on expert evidence in contempt cases might lead to defendants in contempt prosecutions going to great expense to call experts to attempt to define empirically the degree of risk required by a “substantial risk”.

4.67 At the beginning of this chapter,⁸⁶ we referred to another way in which liability for sub judice contempt might arise, other than from the publication of matter that creates a substantial risk of influencing jurors or witnesses. A person, or organisation, might also be convicted of sub judice contempt for publishing material that tends to impose improper pressure on a party to legal proceedings as to the conduct of those proceedings.⁸⁷ Although liability on this basis typically arises in respect of improper pressure on parties to civil proceedings, there is no reason why it should not also apply in respect of pressure on parties to criminal proceedings. A publication may tend to impose pressure on the defendant, or the Crown, to conduct their case in a particular way. Chapter 6 of this report contains the Commission’s recommendation for reform, as well as a more detailed discussion of the current law in this area. While that chapter deals specifically with publications relating to civil proceedings, the recommendation makes it clear that this restriction on publishing matter has a general application, and is not limited to the publication of material relating to parties to civil proceedings.

Table of Contents

1. See NSW Law Reform Commission, Contempt by Publication (Discussion Paper 43, 2000) (“NSWLRC DP 43”) ch 4.

4. Hinch v Attorney General (Vic) (1987) 164 CLR 15 at 34 (Wilson J), at 46 (Deane J), at 70 (Toohey J), at 88 (Gaudron J). See also Attorney General (NSW) v John Fairfax & Sons Ltd (1985) 6 NSWLR 695; Director of Public Prosecutions (Cth) v Australian Broadcasting Corporation (1987) 7 NSWLR 588; Director of Public Prosecutions (Cth) v Wran (1986) 7 NSWLR 616; Attorney General (NSW) v Dean (1990) 20 NSWLR 650; Attorney General (NSW) v Radio 2UE Sydney Pty Ltd (NSWCA, No 40236/96, 16 October 1997, unreported).

5. See Ex parte Auld; Re Consolidated Press Ltd (1936) 36 SR (NSW) 596 at 598 (Jordan CJ); Hinch v Attorney General (Vic) at 70 (Toohey J); Director of Public Prosecutions (Cth) v John Fairfax & Sons Ltd (1987) 8 NSWLR 732 at 736 (Kirby P); Attorney General (NSW) v TCN Channel Nine Pty Ltd (1990) 20 NSWLR 368 at 382; R v Glennon (1992) 173 CLR 592 at 605 (Mason CJ and Toohey J).

6. See NSWLRC DP 43 at para 4.6-4.9; S Walker, The Law of Journalism in Australia (Law Book Company, Sydney, 1989) at para 1.3.13; Australian Law Reform Commission, Contempt and the Media (Discussion Paper 26, 1986) at para 52-59; Australian Law Reform Commission, Contempt (Report 35, 1987) (“ALRC Report 35”) at para 288-295; R Pullan, “Contempt: judicial assertions but no evidence – we are ignorant about the impact of prejudicial pre trial publicity on jurors’ deliberations” (1996) 34 Law Society Journal 48 at 49; M Chesterman, “Reforming the law of contempt” (1984) 58 Law Institute Journal 380 at 381.

7. See Hinch v Attorney General (Vic) (1987) 164 CLR 15 at 27-28 (Mason CJ). See also, for example, R v Day [1985] VR 261 at 264; Attorney General (NSW) v Radio 2UE Sydney Pty Ltd (NSWCA, No 40762/91, 28 August 1992, unreported) at 3 (Priestley JA); Attorney General (NSW) v Television and Telecasters (Sydney) Pty Ltd (NSWSC, No 11752/97, Greg James J, 10 September 1998, unreported) at 13.

9. In the United Kingdom, s 2(2) of the Contempt of Court Act 1981 (UK) provides that a publication will amount to a contempt if it creates a substantial risk that the course of justice will be seriously impeded or prejudiced. In New Zealand, the courts appear to favour a test that focuses on whether the publication created a “real risk”, as a matter of practical reality, that the trial would be “likely” to be prejudiced: see Solicitor General v Radio Avon Ltd [1978] 1 NZLR 225 at 234; Solicitor General v Broadcasting Corporation of New Zealand [1987] 2 NZLR 100 at 107; Gisborne Herald Co Ltd v Solicitor General [1995] 3 NZLR 563 at 569 (Richardson J). In Canada, the courts also appear to determine liability for sub judice contempt according to whether there was a “real risk” of prejudice to the course of justice: see R v Chek TV Ltd (1987) 30 BCLR (2d) 36 at 43 (Anderson JA); Re Attorney General for Manitoba and Radio OB Ltd (1976) 31 CCC (2d) 1 at 6; R v Scozzafava [1997] OJ No 4576 (QL) (JW Quinn J); R v CHBC Television (British Columbia, Court of Appeal, No 24128, 8 February 1999, unreported) (Essen JA).

10. The Australian Law Reform Commission recommended a test in terms of whether a publication creates a substantial risk that, by virtue of the influence it might exert on the court or jury, a fair trial might be prejudiced: see ALRC Report 35 at para 295. The Phillimore Committee recommended that a publication must create a risk that the course of justice will be seriously impeded or prejudiced: see United Kingdom, Committee on Contempt of Court, Report of the Committee on Contempt of Court (Cmnd 5794, HMSO, London, 1974) at para 113. The Irish Law Reform Commission recommended that liability arise where the proceedings will be seriously impeded or prejudiced: see Ireland, Law Reform Commission, Contempt of Court (Report 47, 1994) at para 19-20. See generally NSWLRC DP 43 at para 4.22-4.25.

13. Attorney General (NSW) v John Fairfax and Sons Ltd (1985) 6 NSWLR 695 at 697-698 (Samuels JA).

14. Victoria v Australian Building Construction Employees’ and Builders Labourers’ Federation (1982) 152 CLR 25 at 56 (Gibbs CJ).

15. See United Kingdom, Parliamentary Debates (Hansard) House of Lords, 15 January 1981 at 182-184.

16. See Attorney General v News Group Newspapers Ltd [1987] 1 QB 1 at 15 (Donaldson MR); Ex Parte Telegraph Group (2001) 1 WLR 1983 (CA).

17. See Attorney General v English [1983] 1 AC 116 at 141-142 (Diplock LJ); Attorney General v Birmingham Post and Mail Ltd [1998] 4 All ER 49 at 52 (Simon Brown LJ).

19. See N Cowdery QC, Submission at 1; ABC, Submission at 1; M Sexton SC, Submission at 2; Victorian Bar Council, Submission at 2-3; Australian Press Council, Submission at 4; Law Society of NSW, Submission at para 9-11; Broadcast Media Representatives, Consultation; Print Media Representatives, Consultation 2.

20. See D Norris, Submission at para 30-36; Government Lawyers, Consultation.

21. See ACT Bar Association, Submission at 1; NSW Bar Association, Submission at para 1-6; Australian Broadcasters, Joint Submission at para 3; NSW Bar Association Representatives, Consultation.

22. NSW Bar Association, Submission at para 1-6; Australian Broadcasters, Joint Submission at para 3. While they agreed with Proposal 3, representatives of the print media also expressed some concern that there was no real difference between the “substantial risk” and “tendency” tests: Print Media Representatives, Consultation 2. David Norris also took the view that there was no difference between the two tests, but considered that it might be good to settle on a single statutory formulation for the sake of certainty, clarity, and consistency: see D Norris, Submission at para 31-32.

23. Broadcast Media Representatives, Consultation; Print Media Representatives, Consultation 1; Victorian Bar Council, Submission at para 9-10. The Australian Press Council, in suggesting a reformulation of Proposal 3, also included reference to the fairness of proceedings being “seriously prejudiced”.

26. See Attorney General (NSW) v John Fairfax & Sons Ltd (1985) 6 NSWLR 695. See also Attorney General v MGN Ltd [1997] 1 All ER 456 at 465 (Schiemann LJ).

27. In several cases, the court hearing the contempt charge has been provided with evidence that the jury did not, or was unlikely to have, come into contact with the offending publication, but has nevertheless found the publisher guilty of sub judice contempt: see Registrar, Court of Appeal v John Fairfax Group Pty Ltd (NSWCA, No 40478/92, 21 April 1993, unreported); R v David Syme & Co Ltd [1982] VR 173; R v Pearce (1992) 7 WAR 395; Attorney General (NSW) v Radio 2UE Sydney Pty Ltd (NSWCA, No 40236/96, 16 October 1997, unreported). See also M Chesterman, “Media prejudice during a criminal jury trial: stop the trial, fine the media, or why not both?” (1999) 1 University of Technology Sydney Law Review 71.

28. M Chesterman, J Chan and S Hampton, Managing Prejudicial Publicity (Justice Research Centre, Law and Justice Foundation of NSW, 2001) at para 508.

31. See NSWLRC DP 43 at para 4.36-4.39 for a discussion of the types of cases in which the courts have typically found a publication to constitute a contempt on the basis of its possible influence on a witness.

36. See NSW Bar Association, Submission at para 7-8; D Norris, Submission at para 37-38; Law Society of NSW, Submission at para 12.

40. The Australian Law Reform Commission recommended that legislation prescribe, as an exhaustive list, the categories of statements to give rise to liability for contempt: see ALRC Report 35 at para 291. The Irish Law Reform Commission recommended that legislation set out the types of statements that might give rise to liability as an illustrative, non-exhaustive list: see Ireland, Law Reform Commission, Contempt of Court (Report 47, 1994) at para 6.9.

42. See N Cowdery QC, Submission at 1; ABC, Submission at 1; Law Society of NSW, Submission at para 14; Print Media Representatives, Consultation 2.

43. D Norris, Submission at para 44-45; Australian Press Council, Submission at para 6; Australian Broadcasters, Joint Submission at para 4; Broadcast Media Representatives, Consultation.

44. ABC, Submission at 1; Australian Press Council, Submission at 4; Print Media Representatives, Consultation 1.

48. It was similarly argued in consultation that the phrase “or from which it could reasonably be inferred” should be deleted from the first and second categories: see Print Media Representatives, Consultation 1.

51. Print Media Representatives, Consultation 1. There was some support among members of this group for an illustrative list, provided that the third and fourth categories were deleted.

54. See R v George (1987) 9 NSWLR 527; R v Murdoch (1987) 37 A Crim R 118; R v Glennon (1992) 173 CLR 592.

55. See Registrar, Court of Appeal v Willesee (1985) 3 NSWLR 650; R v Sun Newspapers Pty Ltd (1992) 58 A Crim R 281; Attorney General (NSW) v Television and Telecasters (Sydney) Pty Ltd (NSWSC, No 11752/97, Greg James J, 10 September 1998, unreported); R v Day [1985] VR 261.

56. See Registrar, Court of Appeal v Willesee (1985) 3 NSWLR 650 at 663 (Hope JA); R v Day at 264 (Gobbo J). But see the consideration of R v Day by Greg James J in Attorney General (NSW) v Television and Telecasters (Sydney) Pty Ltd at 9-10.

57. In Attorney General (NSW) v John Fairfax Publications Pty Ltd [1999] NSWSC 318 (9 April 1999) (Barr J), a trial judge had refused to adjourn the commencement of a criminal trial because of a publication in a newspaper. Evidence of the trial judge’s refusal appears to have been admitted in the contempt proceedings against the newspaper, although the judge hearing the contempt charge found that the refusal to adjourn the trial was not binding on the decision in the contempt proceedings.

60. See Attorney General v Birmingham Post and Mail Ltd [1998] 4 All ER 49 at 59 (Simon Brown LJ).

61. See R v CHBC Television (British Columbia, Court of Appeal, No 24128, 8 February 1999, unreported) at 75 (Esson JA).

62. N Cowdery QC, Submission at 1; D Norris, Submission at para 53. Mr Norris’ view was perhaps not completely in line with Proposal 5. He agreed that evidence that a trial judge has discharged a jury should be admissible evidence in contempt proceedings. He did not, however, address the question of whether a decision not to discharge a jury should also be considered relevant evidence.

63. ABC, Submission at 1; Australian Press Council, Submission at para 7; Australian Broadcasters, Joint Submission at para 5; Government Lawyers, Consultation; Print Media Representatives, Consultation 1. The Law Society was ambiguous in its view on this proposal. It stated that it agreed with the Commission on this issue, but described the Commission’s position as viewing evidence of a decision to discharge a jury as relevant to the question of penalty for contempt, rather than liability. It stated that, notwithstanding a trial judge’s decision to discharge a jury, the prosecution should still need to prove that a publication created a substantial risk of prejudice in order to prove contempt, but did not express a view on whether the trial judge’s decision should be admissible evidence relevant to proving liability for contempt: see Law Society of NSW, Submission at para 15.

65. Australian Broadcasters, Joint Submission at para 5; Government Lawyers, Consultation.

69. Attorney General (NSW) v John Fairfax & Sons Ltd (1985) 6 NSWLR 695. See also Attorney General v MGN Ltd [1997] 1 All ER 456 at 463 (Schiemann LJ).

70. Director of Public Prosecutions (Cth) v Wran (1986) 7 NSWLR 616 at 628-629.

71. See Attorney General (NSW) v TCN Channel Nine Pty Ltd (1990) 20 NSWLR 368 (Hunt J). It was held in that case that, although in general the influence of a later publication would become merged in the effect of previous publications, a particular publication of information may be “so dramatic as to cause the effect of that publication to persevere whereas the effect of the previous publication would otherwise have been erased by the passage of time”: see 5 BR 10 at 20.

74. Australian Press Council, Submission at para 8; Australian Broadcasters, Joint Submission at para 6; Print Media Representatives, Consultation 1.

79. See Waterhouse v Australian Broadcasting Corporation (1986) 6 NSWLR 716 at 736 (Mahoney JA); Attorney General (NSW) v TCN Channel Nine Pty Ltd (1990) 20 NSWLR 368 at 383. See also Attorney General v News Group Newspapers Ltd [1987] 1 QB 1 at 16 (Donaldson MR); Ex parte Telegraph Plc [1993] 2 All ER 971 at 978 (Taylor CJ); Attorney General v MGN Ltd [1997] 1 All ER 456 at 461.

82. See Attorney General (NSW) v John Fairfax Publications Pty Ltd [1999] NSWSC 318.

83. Justice Barr reached the conclusion that the survey on which much of the expert opinion was based was of limited value because of the difficulties of replicating real life conditions in a mock trial. Consequently, he considered that the survey results could not form the basis for any reasonable conclusion that there was a small likelihood of the material prejudicing the administration of justice: see Attorney General (NSW) v John Fairfax Publications Pty Ltd at para 95, 102.

87. See Commercial Bank of Australia Ltd v Preston [1981] 2 NSWLR 554; Harkianakis v Skalkos (1997) 42 NSWLR 22; Attorney General v Times Newspapers Ltd [1973] 1 QB 710; Hammersley Iron Pty Ltd v Lovell (1998) 19 WAR 316.